Human Sleep and Cognition, Part II, Volume 190: Clinical and Applied Research (Progress in Brain Research)

SERIES EDITORS

STEPHEN G. WAXMAN Bridget Marie Flaherty Professor of Neurology Neurobiology, and Pharmacology; Director, Center for Neuroscience & Regeneration/Neurorehabilitation Research Yale University School of Medicine New Haven, Connecticut USA

DONALD G. STEIN Asa G. Candler Professor Department of Emergency Medicine Emory University Atlanta, Georgia USA

DICK F. SWAAB Professor of Neurobiology Medical Faculty, University of Amsterdam; Leader Research team Neuropsychiatric Disorders Netherlands Institute for Neuroscience Amsterdam The Netherlands

HOWARD L. FIELDS Professor of Neurology Endowed Chair in Pharmacology of Addiction Director, Wheeler Center for the Neurobiology of Addiction University of California San Francisco, California USA

To Judith and Emmy

Elsevier Radarweg 29, PO Box 211, 1000 AE Amsterdam, The Netherlands Linacre House, Jordan Hill, Oxford OX2 8DP, UK 360 Park Avenue South, New York, NY 10010-1710 First edition 2011 Copyright Ó 2011 Elsevier B.V. All rights reserved No part of this publication may be reproduced, stored in a retrieval system or transmitted in any form or by any means electronic, mechanical, photocopying, recording or otherwise without the prior written permission of the publisher Permissions may be sought directly from Elsevier’s Science & Technology Rights Department in Oxford, UK: phone (þ44) (0) 1865 843830; fax (þ44) (0) 1865 853333; email: [email protected]. Alternatively you can submit your request online by visiting the Elsevier web site at http://elsevier.com/locate/permissions, and selecting Obtaining permission to use Elsevier material Notice No responsibility is assumed by the publisher for any injury and/or damage to persons or property as a matter of products liability, negligence or otherwise, or from any use or operation of any methods, products, instructions or ideas contained in the material herein. Because of rapid advances in the medical sciences, in particular, independent verification of diagnoses and drug dosages should be made Library of Congress Cataloging-in-Publication Data A catalog record for this book is available from the Library of Congress British Library Cataloguing in Publication Data A catalogue record for this book is available from the British Library ISBN: 978-0-444-53817-8 ISSN: 0079-6123

For information on all Elsevier publications visit our website at elsevierdirect.com

Printed and bound in Great Britain 11 12 13 14

10 9 8

7

6 5

4

3

2 1

Contributors T. Åkerstedt, Departement of Clinical Neuroscience, Karolinska Institutet and Stress Research Institute, Stockholm University, Stockholm, Sweden M.S. Aloia, Department of Medicine, National Jewish Health, Denver, CO, USA M. Barnes, Institute for Breathing and Sleep, Austin Health, Heidelberg, Melbourne, VIC, Australia G. Belenky, Sleep and Performance Research Center, Washington State University, Spokane, WA, USA C. Cajochen, Centre for Chronobiology, Psychiatric Hospital of the University of Basel, Basel, Switzerland J.A. Caldwell, Jr., Fatigue Science, Honolulu, HI, USA J.L. Caldwell, Air Force Research Laboratory, Wright-Patterson Air Force Base, Dayton, OH, USA A. Capelli, Univ. de Bordeaux, Sommeil, Attention et Neuropsychiatrie, USR 3413, Bordeaux, France M.A. Carskadon, Department of Psychiatry and Human Behavior, Warren Alpert Medical School of Brown University, Providence, RI, USA S.L. Chellappa, Centre for Chronobiology, Psychiatric Hospital of the University of Basel, Basel, Switzerland and CAPES Foundation/Ministry of Education of Brazil, Brasilia-DF, Brazil A.M.L. Coenen, Donders Centre for Cognition, Radboud University Nijmegen, Nijmegen, The Netherlands M.C.M. Gordijn, Department of Chronobiology, University of Groningen, Groningen, The Netherlands M.E. Howard, Institute for Breathing and Sleep, Austin Health, Heidelberg, Melbourne, VIC, Australia S.R. Hursh, Institutes for Behavior Resources, Inc., Baltimore, MD, USA M.L. Jackson, Sleep and Performance Research Center, Washington State University, Spokane, WA, USA G. Kecklund, Departement of Clinical Neuroscience, Karolinska Institutet and Stress Research Institute, Stockholm University, Stockholm, Sweden S.J.G. Lewis, Parkinson’s Disease Research Clinic, Ageing Brain Centre, Brain & Mind Research Institute, The University of Sydney, Sydney, NSW, Australia M.M. Lorist, Department of Experimental Psychology, University of Groningen and BCN-NeuroImaging Center, University Medical Center Groningen, Groningen, The Netherlands E.E. Matthews, College of Nursing, University of Colorado, Aurora, CO, USA L.L. McGee-Koch, Circadian Rhythms & Sleep Research Laboratory, Northwestern University, Chicago, IL, USA S.L. Naismith, Healthy Brain Ageing Clinic, Ageing Brain Centre, Brain & Mind Research Institute, The University of Sydney, Sydney, NSW, Australia P. Philip, Univ. de Bordeaux, Sommeil, Attention et Neuropsychiatrie, USR 3413, Bordeaux, France T.G. Raslear, Office of Research and Development, Federal Railroad Administration, Washington, DC, USA K.J. Reid, Department of Neurology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA v

vi

N.L. Rogers, Chronobiology and Sleep Group, Brain & Mind Research Institute, The University of Sydney, Sydney, NSW, Australia J. Snel, Department of Psychonomics, University of Amsterdam, Amsterdam, The Netherlands H.P.A. Van Dongen, Sleep and Performance Research Center, Washington State University, Spokane, WA, USA A. Vermeeren, Department of Neuropsychology and Psychopharmacology, Maastricht University, Maastricht, The Netherlands L.J. Wu, Sleep and Performance Research Center, Washington State University, Spokane, WA, USA P.C. Zee, Department of Neurology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA

Preface It is increasingly recognized that there is a critical, bidirectional relationship between sleep and cognition. The literature in this area, encompassing rich and multidimensional research foci, is dotted with important discoveries—but surprisingly scattered. For example, dozens of recent papers in about as many different scientific journals are hotly debating what happens in the human brain during sleep and during sleep deprivation and how this affects cognitive performance. For psychologists, sleep researchers, neuroscientists, clinicians, and others working in a spectrum of disciplines where sleep and cognition are highly relevant, it is difficult to get a good overview of the basic principles, latest discoveries, and outstanding challenges. Perhaps because of the interdisciplinary nature of the science of sleep and cognition, it is not a clearly defined subdiscipline in any established academic field, and as yet not well integrated. Human Sleep and Cognition (Parts 1 and 2) aims to bring together and make accessible cutting-edge research on the subject in the basic, clinical, and applied sciences, to review current knowledge and understanding, provide a starting point for researchers and practitioners entering the field, and build a platform for further research and discovery. Part 1: Basic Research reviewed the basic aspects, primarily based on human data but also including some relevant animal research. Part 2: Clinical and Applied Research builds on the foundation provided by Part 1 and adds chapters with direct relevance and application in the real world. Part 2 is organized into the following three sections: “Sleep disorders and cognitive functioning,” “Cognitive effects of clinical interventions in sleep and wakefulness,” and “Sleep and cognition in real-world settings.” In Part 2: Clinical and Applied Research, Reid, McGee-Koch, and Zee (2011) review sleep disorders caused by the endogenous biological clock, and what is known about their effects on cognition. Neurodegenerative disease tends to affect sleep, cognition, and their interaction all at the same time, yielding a complex clinical picture comprehensively described by Naismith, Lewis, and Rogers (2011). Jackson, Howard, and Barnes (2011) discuss the effects on cognitive functioning of sleep-related breathing disorders such as sleep apnea, due to the abnormal respiration and indirectly resulting from the disruptive influence thereof on sleep. Moving on to treatment of sleep disorders, Matthews and Aloia (2011) discuss the effects of positive airway pressure, a choice therapy for sleep apnea, on cognition—which can be beneficial but varies in effectiveness. Vermeeren and Coenen (2011) review the side effects on cognition of different types of hypnotics used to improve sleep. Snel and Lorist (2011), on the other hand, discuss the impact on cognition of caffeine used to promote wakefulness and alertness. Another way to promote alertness is bright-light exposure, which is addressed in the context of cognition by Chellappa, Gordijn, and Cajochen (2011). The importance of good sleep for optimal cognition is clearly seen in adolescents, whose early school times interfere with getting enough sleep and result in sleepiness, impaired performance, and learning difficulties, as pointed out by Carskadon (2011). Sleep loss is common throughout society these days, but Van Dongen, Caldwell, and Caldwell (2011) demonstrate that not everyone is equally vulnerable to the adverse consequences this has on cognitive performance. Impending cognitive impairment from loss of sleep and/or temporal misalignment relative to the endogenous biological clock can be predicted ix

x

mathematically, and recent advances in this area are presented by Raslear, Hursh, and Van Dongen (2011). Such progress notwithstanding, sleep loss remains a key contributor to accidents in the workplace and on the road, as reviewed by Åkerstedt, Philip, Capelli, and Kecklund (2011). How to best manage sleep and fatigue to reduce errors and accidents and to improve cognitive performance, safety, productivity, and well-being is a focus of current investigation and debate. Belenky, Wu, and Jackson (2011) close Part 2: Clinical and Applied Research with an introduction into a new interdisciplinary field concerned with this issue: occupational sleep medicine. A wide variety of clinical disorders and operational conditions is associated with insomnia, that is, inability to obtain sufficient sleep because of difficulty initiating or maintaining sleep. As discussed throughout the pages of Human Sleep and Cognition (Parts 1 and 2), insomnia of definable origin is typically associated with cognitive impairment. However, in the case of psychophysiological insomnia (also called primary insomnia), the underlying cause is unknown and findings on concomitant cognitive impairment are mixed (Shekleton et al., 2010). Chronic psychophysiological insomniacs report excessive fatigue, negative mood states, and cognitive dysfunction (Sateia et al., 2000), but the cognitive dysfunction has been hard to substantiate with objective observations (Orff et al., 2007; Riedel and Lichstein, 2000). It has been suggested that cognitive deficits in chronic insomniacs may be masked by compensatory effort and/or stimulation from the environment (Kloss, 2003; Varkevisser et al., 2007). A recent laboratory study found that when the cognitive performance of a group of chronic psychophysiological insomniacs and a group of healthy controls was measured repeatedly during 24 h of continuous wakefulness under strictly controlled, nonstimulating conditions, the insomniacs performed worse than the controls across the hours of the day (Varkevisser and Kerkhof, 2005). There was no evidence of heightened physiological arousal in the group of insomniacs (Varkevisser et al., 2005), which is notable because hyperarousal has been theorized to underlie the inability to obtain sufficient sleep (Bonnet and Arand, 1997). It is possible that in this population, hyperarousal is primarily an issue in the context of preparing for bedtime (Harvey, 2000; Vgontzas et al., 2001). Alternatively, sleep-related attentional bias rather than hyperarousal may be keeping patients awake (MacMahon et al., 2006; Marchetti et al., 2006). This issue continues to be a theme in research on insomnia; see Fulda and Schulz (2010) in Part 1: Basic Research for a discussion of relevant methodological considerations. Much more can be said about insomnia, its consequences for cognition, its impact on functioning in the workplace, and the various treatment options in clinical practice. Reviews can be found in the published literature (e.g., Doghramji, 2010; Riemann, 2010; Vernon et al., 2010). Focusing on a range of other topics, the chapters in Human Sleep and Cognition, Part 2: Clinical and Applied Research, describe cutting-edge developments in clinical and operational research and their applications at the intersection of sleep and cognition. Hans P.A. Van Dongen Gerard A. Kerkhof References Åkerstedt, T., Philip, P., Capelli, A., & Kecklund, G. (2011). Sleep loss and accidents—Work hours, life style and sleep pathology. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 169–188). Amsterdam: Elsevier. Belenky, G., Wu, L. J., & Jackson, M. L. (2011). Occupational sleep medicine: Practice and promise. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 189–203). Amsterdam: Elsevier.

xi Bonnet, M. H., & Arand, D. L. (1997). Hyperarousal and insomnia. Sleep Medicine Reviews, 1, 97–108. Carskadon, M. A. (2011). Sleep's effects on cognition and learning in adolescence. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 137–143). Amsterdam: Elsevier. Chellappa, S. L., Gordijn, M. C. M., & Cajochen, C. (2011). Bright light effects on sleep and cognition. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 119–133). Amsterdam: Elsevier. Doghramji, K. (2010). The evaluation and management of insomnia. Clinics in Chest Medicine, 31, 327–339. Fulda, S., & Schulz, H. (2010). How treatment affects cognitive deficits in patients with sleep disorders: Methodological issues and results. In G. A. Kerkhof & H. P. A. Van Dongen (Eds.), Human sleep and cognition. Part 1: Basic research. Progress in Brain Research, Vol. 185 (pp. 69–90). Amsterdam: Elsevier. Harvey, A. G. (2000). Pre-sleep cognitive activity in insomnia: A comparison of sleep-onset insomniacs and good sleepers. British Journal of Clinical Psychology, 39, 275–286. Jackson, M. L., Howard, M. E., & Barnes, M. (2011). Cognition and daytime functioning in sleep-related breathing disorders. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 53–68). Amsterdam: Elsevier. Kloss, J. D. (2003). Daytime sequelae of insomnia. In M. P. Szuba, J. D. Kloss & D. F. Dinges (Eds.), Insomnia: Principles and management (pp. 23–42). Cambridge: Cambridge University Press. MacMahon, K. M., Broomfield, N. M., & Espie, C. A. (2006). Attention bias for sleep-related stimuli in primary insomnia and delayed sleep phase syndrome using the dot-probe task. Sleep, 29, 1420–1427. Marchetti, L. M., Biello, S. M., Broomfield, N. M., MacMahon, K. M., & Espie, C. A. (2006). Who is pre-occupied with sleep? A comparison of attention bias in people with psychophysiological insomnia, delayed sleep phase syndrome and good sleepers using the induced change blindness paradigm. Journal of Sleep Research, 15, 212–221. Matthews, E. E., & Aloia, M. S. (2011). Cognitive recovery following positive airway pressure (PAP) in sleep apnea. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 71–88). Amsterdam: Elsevier. Naismith, S. L., Lewis, S. J. G., & Rogers, N. L. (2011). Sleep-wake changes and cognition in neurodegenerative disease. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 21–52). Amsterdam: Elsevier. Orff, H. J., Drummond, S. P. A., Nowakowski, S., & Perlis, M. L. (2007). Discrepancy between subjective symptomatology and objective neuropsychological performance in insomnia. Sleep, 30, 1205–1211. Raslear, T. G., Hursh, S. R., & Van Dongen, H. P. A. (2011). Predicting cognitive impairment and accident risk. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 155–167). Amsterdam: Elsevier. Reid, K. J., McGee-Koch, L. L., & Zee, P. C. (2011). Cognition in circadian rhythm sleep disorders. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 3–20). Amsterdam: Elsevier. Riedel, B. W., & Lichstein, K. L. (2000). Insomnia and daytime functioning. Sleep Medicine Reviews, 4, 277–298. Riemann, D. (2010). Hyperarousal and insomnia: State of the science. Sleep Medicine Reviews, 14, 9–15. Sateia, M. J., Doghramji, K., Hauri, P. J., & Morin, C. M. (2000). Evaluation of chronic insomnia. Sleep, 23, 1–21. Shekleton, J. A., Rogers, N. L., & Rajaratnam, S. M. (2010). Searching for the daytime impairments of primary insomnia. Sleep Medicine Reviews, 14, 47–60. Snel, J., & Lorist, M. M. (2011). Effects of caffeine on sleep and cognition. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 105–117). Amsterdam: Elsevier. Van Dongen, H. P. A., Caldwell, J. A., & Caldwell, J. L. (2011). Individual differences in cognitive vulnerability to fatigue in the laboratory and in the workplace. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 145–153). Amsterdam: Elsevier. Varkevisser, M., & Kerkhof, G. A. (2005). Chronic insomnia and performance in a 24-hour constant routine study. Journal of Sleep Research, 14, 49–59. Varkevisser, M., Van Dongen, H. P. A., & Kerkhof, G. A. (2005). Physiologic indexes in chronic insomnia during a constant routine: Evidence for general hyperarousal? Sleep, 28, 1588–1596.

xii Varkevisser, M., Van Dongen, H. P. A., Van Amsterdam, J. G. C., & Kerkhof, G. A. (2007). Chronic insomnia and daytime functioning: An ambulatory assessment. Behavioral Sleep Medicine, 5, 279–296. Vermeeren, A., & Coenen, A. M. L. (2011). Effects of the use of hypnotics on cognition. In H. P. A. Van Dongen & G. A. Kerkhof (Eds.), Human sleep and cognition. Part 2: Clinical and applied research. Progress in Brain Research, Vol. 190 (pp. 89–103). Amsterdam: Elsevier. Vernon, M. K., Dugar, A., Revicki, D., Treglia, M., & Buysse, D. (2010). Measurement of non-restorative sleep in insomnia: A review of the literature. Sleep Medicine Reviews, 14, 205–212. Vgontzas, A. N., Bixler, E. O., Lin, H. M., Prolo, P., Mastorakos, G., Vela-Bueno, A., et al. (2001). Chronic insomnia is associated with nyctohemeral activation of the hypothalamic-pituitary-adrenal axis: Clinical implications. Journal of Clinical Endocrinology & Metabolism, 86, 3787–3794.

H. P. A. Van Dongen and G. A. Kerkhof (Eds.) Progress in Brain Research, Vol. 190 ISSN: 0079-6123 Copyright Ó 2011 Elsevier B.V. All rights reserved.

CHAPTER 1

Cognition in circadian rhythm sleep disorders Kathryn J. Reid{,*, Lori L. McGee-Koch{ and Phyllis C. Zee{ {

Department of Neurology, Feinberg School of Medicine, Northwestern University, Chicago, IL, USA { Circadian Rhythms & Sleep Research Laboratory, Northwestern University, Chicago, IL, USA

Abstract: Circadian rhythms in physiology and behavior exist in all living organisms, from cells to humans. The most evident rhythms are the recurrent cycles of sleep and wake as well as changes in alertness and cognitive performance across the 24 h. Clearly, sleep pressure can exert a strong influence on cognitive performance, but the influence of circadian modulation of alertness and cognitive function is evident even when the pressure for sleep is high. Circadian rhythms also influence more complex cognitive tasks, such as selective attention and executive function, which are important for work performance and safety. The circadian timekeeping system also ensures that circadian rhythms are appropriately synchronized to the external physical environment and work and social schedules. Circadian misalignment is the basis for all circadian rhythm sleep disorders. These disorders are often associated with impairments of cognitive performance that can have adverse effects on school and work performance, overall quality of life, and safety. Keywords: circadian; sleep disorders; sleep; cognitive performance.

Introduction

function has adverse consequences for sleep, health, and cognitive performance. The recurring daily cycle of sleep and wakefulness is the most apparent rhythm that is regulated by the circadian timekeeping system. Therefore, disruption of circadian timing is often accompanied by alterations in sleep and daytime alertness. The International Classification of Sleep Disorders, 2nd edition (ICSD-2) describes nine circadian rhythm sleep disorders (CRSDs) defined by a persistent or recurrent pattern of sleep disturbance resulting from either alterations of the circadian timekeeping system or misalignment between the

The circadian system regulates or modulates the temporal organization of most physiological and behavioral processes across the 24-h day, including sleep, wakefulness, mood, and cognitive performance. The timing and strength of circadian rhythms are influenced by both endogenous and environmental factors. Disruption of circadian *Corresponding author. Tel.: þ1 312 503 1528; Fax: þ1 312 503 5679 E-mail: [email protected] DOI: 10.1016/B978-0-444-53817-8.00001-3

3

4

endogenous circadian rhythm and exogenous factors that affect the timing and duration of sleep (ICSD-2, 2005). This chapter provides a review of (1) basic principles of circadian biology and its modulating role on sleep, alertness, and cognitive function; (2) the impact of circadian rhythm disturbance in relation to the six most common CRSDs on cognition, psychomotor performance, and alertness. Circadian biology Understanding of the relationship between cognitive performance and circadian rhythm disturbances requires knowledge of the fundamentals of circadian biology. Endogenous circadian (near 24 h) rhythms are generated by a master clock, located in the suprachiasmatic nuclei (SCN) of the hypothalamus, which modulates the activity of other brain regions and peripheral clocks located throughout the body (Dibner et al., 2010). In the absence of external cues, the endogenous free-running period of the human circadian clock is slightly longer than 24 h (Czeisler et al., 1999). Therefore, in order to maintain a 24-h daily cycle in physiology and behavior, circadian rhythms need to be synchronized or entrained to the 24-h light and dark cycle. The timing of the circadian clock is adjusted by internal cues (melatonin) and external cues (light, social and physical activity, sound; Baehr et al., 2003; Barger et al., 2004; Goel, 2005). The primary synchronizing agents in humans are light and melatonin. The ability of these agents to alter the timing of circadian rhythms (phase resetting) is highly dependent upon the biological timing of their presentation (Khalsa et al., 2003; Kripke et al., 2007; Lewy et al., 1998).

Role of light and melatonin in circadian timing In humans, light is the strongest synchronizing agent for the circadian clock. Photic information from the eye is primarily sent from intrinsically

photosensitive retinal ganglion cells that contain the blue light-sensitive photopigment melanopsin (Berson et al., 2002; Freedman et al., 1999; Gooley et al., 2001; Hattar et al., 2002; Ruby et al., 2002) to the SCN via a direct pathway, the retinohypothalamic tract (RHT; Gooley et al., 2001), and an indirect pathway via the intergeniculate leaflet (IGL; Harrington, 1997). The rods and cones via their synaptic connections with retinal ganglion cells can also provide photic input to the SCN (Berson et al., 2010; Wong et al., 2007). The timing, intensity, duration, and wavelength of light determine the direction and magnitude of light-induced phase shifts of the circadian clock (Duffy and Wright, 2005). For example, bright light exposure in the evening results in phase delays (shifts the biological clock later), whereas light exposure in the early morning results in phase advances (shifts the biological clock earlier). Light-induced phase shifts of circadian rhythms are most sensitive to blue light, in the range of 446–477 nm (Lockley et al., 2003; Warman et al., 2003). However, it remains unclear whether blueenriched bright light is more effective than broad spectrum white light at shifting the circadian clock in humans or in patients with CRSDs (Smith and Eastman, 2009; Smith et al., 2009). Melatonin is another important synchronizing agent for circadian rhythms (Lewy et al., 1992). Melatonin is produced in the pineal gland, and its secretion during the night by the pineal gland is regulated by the SCN via the sympathetic superior cervical ganglion (Moore, 1996). Melatonin secretion is inhibited directly by the presence of bright light (Brainard et al., 1988; Lockley et al., 2003). Like light-induced circadian phase shifts, light-induced melatonin suppression is most sensitive to blue light, in the range of 446–477 nm (Lockley et al., 2003; Warman et al., 2003). Exogenous melatonin and melatonin agonists via their action on melatonin receptors in the SCN have been shown to induce phase shifts of the circadian clock (Dubocovich, 2007). Similar to light (but reversed), the biological timing and dose of melatonin or melatonin-receptor agonist administered

5

determines the direction and the magnitude of phase shift (Burgess et al., 2008, 2010; Lewy et al., 1998; Richardson et al., 2008). When taken in the late afternoon or early evening, melatonin advances circadian timing, whereas when administered in the morning, it will delay circadian timing (Burgess et al., 2008; Lewy et al., 1998). Circadian regulation of sleep and wakefulness In addition to its prominent role in temporal organization, the circadian system plays an important role in the regulation of sleep and wake propensity, as well as fluctuations in alertness and performance across the day. The timing and propensity for sleep and wakefulness is regulated by a complex interaction of endogenous circadian (C) and sleep homeostatic (S) processes, as well as the influence of external behavioral and environmental factors (Edgar et al., 1993). As a function of time awake, the homeostatic pressure for sleep increases during the day. In opposition to the buildup of process S, the circadian alerting signal peaks in the early evening to maintain wakefulness until bedtime (Fig. 1). Even when the pressure to sleep is high (following prolonged wakefulness), the propensity to sleep and cognitive performance is attenuated in the early evening (Dijk and Czeisler, 1994; Lavie, 1986, 1997). After reaching its high point in the early evening, circadian alertness decreases, corresponding to the nocturnal secretion of pineal melatonin and the decline in core body temperature (Cajochen et al., 1999). Under normal conditions, the low circadian signal for alertness in the evening and early morning hours and high melatonin levels help to facilitate sleep (Dijk and Czeisler, 1994). Indeed, exogenous melatonin and melatonin-receptor agonist drugs have been used to help promote sleep (Markwald et al., 2010; Rajaratnam et al., 2009; Wyatt et al., 2006). Proper functioning of the circadian timekeeping system and proper alignment between circadian timing and sleep and wake behaviors are essential for performance. Understanding the

Two-process model of sleep regulation

Process C

Process S

Fig. 1. The two-process model of sleep regulation. The top panel is a schematic of the circadian alerting signal or Process C. The bottom panel is a schematic of the homeostatic component to sleep–wake regulation or Process S. Process S builds up with wakefulness and is dissipated during sleep (black bars). Redrawn and modified from Beersma and Gordijn, 2007 with permission.

interplay between the circadian and homeostatic control of sleep–wake has important implications for understanding the mechanisms that underlie the cognitive impairments associated with CRSDs, as well as their treatments. Effect of light and melatonin on cognitive function In addition to its ability to reset the timing of circadian rhythms, bright light has direct effects on neurobehavioral function both during the day (Phipps-Nelson et al., 2003) and at night (Cajochen et al., 2005; Lockley et al., 2008). These effects are modulated by the nonvisual effects of light. Findings from neuroimaging studies suggest that light can alter the activity of cortical and subcortical structures (hypothalamus, brainstem, thalamus) that regulate alertness (Vandewalle et al., 2006). The alerting effect of light is dependent on the wavelength and duration of the light exposure (Cajochen, 2007). Cajochen showed that exposure

6

to 2 h of monochromatic light at 460 nm in the late evening resulted in significant increases in subjective alertness, core temperature, and reductions in melatonin compared to light exposure at 550 nm (Cajochen et al., 2005). A longer pulse (6.5 h) of 460 nm light at night resulted in significant improvements in reaction time on an auditory vigilance task compared to 555 nm light (Lockley et al., 2006). During the day (12–5 p.m.), light has been shown to improve alertness and psychomotor vigilance performance (Phipps-Nelson et al., 2003). The alerting effect of bright light likely plays a role in the symptoms experienced by those with CRSDs, since they are typically exposed to bright light at inappropriate circadian times. Since the regulation of melatonin is controlled via the SCN and melatonin is high during the biological night, performance and alertness are usually impaired when endogenous melatonin levels are high compared to when melatonin is low (Cajochen et al., 1999). This relationship between melatonin and alertness also holds true for exogenous administration of melatonin given when melatonin levels would normally be low (Reid et al., 1996; Rogers et al., 1998, 2003). Reid et al. (1996) showed that daytime administration of melatonin at 2 p.m. resulted in a significant reduction in sleep-onset latency (SOL) on the multiple sleep latency test compared to placebo (Reid et al., 1996). This increase in sleepiness is also associated with performance impairments. Rogers et al. (1998) showed that performance on a tracking task and on response and reaction time scores for visual choice and extended two-choice visual tasks were impaired following the administration of 5 mg of melatonin at 12:30 p.m. (Rogers et al., 1998). For these reasons, caution should be used if using exogenous melatonin in the early morning in the treatment of advanced sleep-phase disorder (ASPD; Zee, 2008). Cognition and circadian rhythms The influence of circadian timing on cognitive performance was described by Kleitman in the early

1930s (Kleitman, 1933). Kleitman showed a relationship between the diurnal fluctuations in the speed and accuracy of psychomotor performance, including complex cognitive domains, with the rhythm in core body temperature. He noted that the best performance occurred in the afternoon, when temperature was high, whereas, lower performance in the early morning and late evening were seen at the temperature minimum (Kleitman, 1963; Kleitman and Jackson, 1950). Kleitman's hypothesis that alertness and performance is modulated by a daily rhythm that is independent of sleep has been confirmed by studies using time isolation and forced desynchrony protocols. Aschoff and Wever (1976) demonstrated that diurnal performance rhythms are modulated by the endogenous circadian system in humans. This was the first evidence that, when there is adequate sleep, both the circadian system and the imposed sleep–wake cycle (i.e., duration of prior wakefulness in this case) contribute approximately equally to the variation in cognitive performance. Constant routine (CR) protocols have also been used to unmask endogenous circadian rhythms (Mills et al., 1978). As the name suggests, CR protocols control environmental conditions (lighting, temperature, feeding) and remove the influence of sleep. Specifically these conditions include constant levels of ambient light and temperature, constant body posture position (semirecumbent posture in bed), constant food intake (hourly isocaloric snacks), and constant wakefulness. This technique has been used in many studies to determine that there is a parallel between variations in performance, core body temperature, and melatonin level (Cajochen et al., 1999; Dijk et al., 1992; Johnson et al., 1992). It should be noted that this paradigm allows a buildup of the homeostatic sleep drive by inducing wakefulness for extended periods ( 40 h), thereby resulting in a linear decrease in alertness and performance. However, the underlying circadian rhythms are still observed. Figure 2 is an example of data from a CR protocol, although performance declines during the

7 (a)

(b) Relative clock time (h) 16 8 16 24 8

8

Relative clock time (h) 16 24 8 16

Core body temperature

37.0

More tired

⬚C

37.5

36.5

7

Karolinska sleepiness scale

5

3

200

Psychomotor vigilance performance

Plasma melatonin

100

ms

pmol/L

6000

10% slowest reaction time Mean reaction time Median reaction time 10% fastest reaction time

1000 500

15 200 10

Eye blink rate

5

60 # of attempts

# per 30-s epoch

0

16

0 3 2 1

Stage 1 sleep

0 0

8

16

24

32

Elapsed time awake (h)

Cognitive throughput 30

Slow eye movements

4 # of correct pairs

% of 5-min epochs

8

45

2

Probed recall memory test 0 0

8

16

24

32

Elapsed time awake (h)

Fig. 2. The relationship between time of day, time since wake, and measures of circadian rhythms, cognitive performance, and alertness during 32 h of continuous wakefulness. Reductions in nearly all the performance and alertness measures occur at around 16 h of wakefulness and coincide with the increase in endogenous melatonin and reduction in core body temperature rhythm. Panel (a) shows core body temperature, endogenous plasma melatonin, mean eye blink rate, and incidences of slow eye movements (SEMs) and of stage 1 sleep. Panel (b) shows subjective sleepiness assessed by Karolinska sleepiness scale (KSS; highest possible score ¼ 9, lowest possible score ¼ 1), psychomotor vigilance performance, cognitive throughput, and memory performance. Figure taken from Cajochen et al. (1999) with permission.

8

biological night after about 24 h of wakefulness performance begins to improve again because of the increase in the circadian alerting signal. It should be noted that if wakefulness were to continue, performance would not return to the same level as the previous day due to the buildup of the homeostatic sleep drive. Circadian preference and cognitive performance Extreme morning (lark) or evening (owl) type preference does not indicate a CRSD, although those with delayed sleep-phase disorder (DSPD) or ASPD tend to be extreme evening (Chang et al., 2009) or morning types, respectively (Reid et al., 2001). There are several physiological and neurobehavioral differences between morning and evening types that impact performance. Circadian chronotype is typically defined by a person's preferred time to conduct activities, and/or the timing of sleep, assessed by questionnaires (Horne and Östberg, 1976) or Munich chronotype (Zavada et al., 2005) questionnaires). One of the most prominent features of the difference in morning/evening types is that they exhibit different timing to patterns of sleep–wake. Several studies involving students have reported that morning types have earlier bed times and arising times than evening types (Foret et al., 1982; Horne and Östberg, 1976; Kerkhof, 1985; Webb and Bonnet, 1978). Further, Breithaupt et al. (1978) report that morning types have less flexibility, particularly in their rising times, than evening types. Circadian preference can be modulated by age, such that, older people tend to be more morning type and younger adults tend to be more evening type (Roenneberg et al., 2007). Much like those with DSPD tend to be younger than those with ASPD. Morning- and evening-type individuals differ in the endogenous circadian phase of their biological clock, for example, the maximum and nadir in core temperature occurs earlier in morning types than in evening types (Andrade et al., 1992;

Baehr et al., 2000; Breithaupt et al., 1981; Folkard and Monk, 1979; Waterhouse et al., 2001). In addition, Hildebrandt and Stratmann (1979) reported that there was a greater circadian amplitude of the body temperature rhythm for evening types than for morning types (Hildebrandt and Stratmann, 1979). These differences in circadian phase parallel the diurnal course of their neurobehavioral function (Kerkhof and Van Dongen, 1996; Van Dongen et al., 1998). Some people are consistently at their best in the morning, whereas others are more alert and perform better in the evening. In addition to the modulation of circadian phase on neurobehavioral function, there is evidence that there are alterations in the dissipation and buildup of homeostatic drive in those with extreme circadian preference (Mongrain et al., 2006a,b). Studies have shown that evening types with an intermediate circadian phase have lower levels of slow-wave activity and slower homeostatic decay of sleep pressure than morning types with an intermediate circadian phase. Interestingly, this relationship is not evident in morning and evening types with extreme circadian phases (Mongrain et al., 2006a). This alteration in homeostatic drive could contribute to the differences observed in neurobehavioral function. Circadian rhythm sleep disorders CRSDs are characterized by a persistent or recurrent pattern of sleep disturbance resulting in alterations to the timing and duration of sleep, due to the misalignment between endogenous circadian timing and the 24-h physical and social environment. This circadian-related sleep disruption leads to complaints of insomnia or excessive daytime sleepiness with impairment in important areas of functioning and quality of life (ICSD-2, 2005). Based on the underlying pathophysiology, there are two major types of CRSDs, those that are a result of alterations in the external environment relative to the internal circadian timing

9

system (e.g., jet lag and shift-work disorder (SWD)); and those resulting from an intrinsic alteration of the endogenous circadian clock or its input and output pathways (e.g., DSPD, ASPD, irregular sleep–wake rhythm, free-running type). However, the type and severity of symptoms of most CRSDs are influenced by a combination of physiologic, behavioral, and environmental factors. Data on the direct impact of CRSDs on cognitive performance are limited, particularly for those disorders due to alterations in the endogenous circadian system. However, CRSDs result not only in circadian misalignment but also, in many cases, sleep loss. Both of these states are associated with performance impairment. It is often difficult to separate the contribution of the circadian misalignment from the sleep loss associated with the misalignment in these disorders. For example, when an individual with DSPD has to get up early in the morning to go to work/school, sleep will be truncated, leading to chronic sleep loss, which in turn compounds the already impaired performance due to circadian misalignment. The following sections describe and summarize what little is known about the cognitive impairment associated with each of the CRSDs.

Delayed sleep-phase disorder DSPD is characterized by a delay in the sleep–wake cycle of 3–6 h later than desired (Fig. 3), resulting in insomnia, excessive sleepiness, and functional impairments, particularly during the morning hours (ICSD-2, 2005). In addition to sleep, other markers of the endogenous circadian clock such as melatonin (Chang et al., 2009; Shibui et al., 1999) and temperature (Ozaki et al., 1996; Uchiyama et al., 2000a) are also delayed. However, there are conflicting reports on whether there is misalignment between sleep and other circadian phase markers (melatonin onset and core body temperature) in patients with DSPD when compared to controls (Chang

et al., 2009). In a study of a clinical population with DSPD, Chang et al. (2009) reported that there was no circadian misalignment between sleep and either melatonin or core body temperature in those with DSPD when compared to controls. This is most likely due to those with DSPD sleeping at a delayed phase. In contrast, several studies do report a greater interval between core body temperature nadir and sleep offset and between melatonin and sleep propensity rhythms in those with DSPD compared to controls (Uchiyama et al., 2000a,b). There is also evidence that there may be alterations to the input to the circadian clock in DSPD, as they have a hypersensitive suppression of melatonin in response to bright light exposure during the night (Aoki et al., 2001). In addition to alterations in the circadian timing system there is evidence that at least in some individuals with DSPD, there is also an alteration in the homeostatic control of sleep. For example, those with DSPD are unable to compensate as well as controls following sleep loss. Those with DSPD were unable to sleep well until the circadian night following 24 h of sleep deprivation on an ultra-short sleep–wake cycle protocol, where as the controls were able to sleep at most circadian times (Uchiyama et al., 2000b). This inability to recover from sleep loss until the circadian night is likely to have a significant impact on cognitive performance, especially since attempts to maintain a conventional sleep–wake cycle often results in sleep deprivation for those with DSPD. There is currently no direct evidence that DSPD in itself results in cognitive performance impairment compared to controls. However, treatment studies of DSPD have reported improvement in levels of objective sleepiness in the morning (Rosenthal et al., 1990) and behavior problems (Szeinberg et al., 2006). Bright light (2500 lux) administration in the morning for 2 h between 6 and 9 a.m. for 2 weeks increased SOL in the first two multiple sleep latency tests (a test used to determine the ability to fall asleep) administered after wake. Treatment of adolescents with DSPD

10

16:00

20:00

24:00

Clock time 04:00 08:00

12:00

16:00

Advanced sleep-phase disorder Delayed sleep-phase disorder Irregular sleep–wake type

Free-running type

{

Shift work disorder Jet lag

Fig. 3. Schematic representation of circadian rhythm sleep disorders. Black bars indicate sleep periods, white indicates wake, the gray bar indicates typical/conventional sleep time, the blue triangle indicates melatonin onset for the conventional sleep phase, and the yellow triangle indicates core body temperature minimum for the conventional sleep phase. If circadian rhythms and the sleep–wake cycle are in phase (gray bar), the melatonin onset (upward facing triangle) would occur about 2 h prior to sleep onset, and the core body temperature minimum (downward facing triangle) would occur about 2 h prior to wake. Advanced sleep-phase disorder is characterized by a sleep period that is several hours earlier than conventional sleep time and is associated with early evening sleepiness and difficult sleeping to desired wake time. Delayed sleep-phase disorder is characterized by a sleep period that is several hours later than conventional sleep time and is associated with difficulty falling asleep and morning sleepiness. Irregular sleep–wake type is characterized by the absence of a single consolidated sleep period and three of more sleep periods per 24-h period and is associated with sleepiness. Free-running type is characterized by a gradual delay of the sleep–wake cycle by about an hour a day and difficulty falling asleep and sleepiness depending on where sleep is initiated in regard with sleep–wake. The example for shift-work disorder is for a typical sleep period following the night shift and would be associated with difficulty maintaining sleep and sleepiness during the night. The example for jet lag is for a sleep period when traveling in an eastward direction where the sleep period is advanced, and would be associated with difficulty falling asleep and sleepiness upon awakening.

using oral melatonin administration of 3–5 mg has been reported to result in less behavioral and social problems (Szeinberg et al., 2006).

Advanced sleep-phase disorder ASPD is characterized by an advance in the sleep–wake cycle (Fig. 3), excessive sleepiness in the early evening and an inability to sleep until

the desired time in the morning (ICSD-2, 2005). As a result of early morning awakening and staying up until a socially desirable bedtime, those with ASPD often experience sleep loss. The advanced sleep phase is believed to be the result of an advance of the circadian timing system, which is supported by several studies that indicate an advance in other circadian rhythms such as melatonin (Reid et al., 2001) and core body temperature (Lack et al., 2005).

11

There is currently no direct evidence that ASPD in itself results in cognitive performance impairment compared to controls. However, given the advanced phase of the endogenous circadian clock and the reduction of the circadian alerting signal in the evening, those with ASPD are most likely to experience sleepiness and performance decrements in the evening, especially if they attempt to remain awake until desired or conventional times.

Free-running type Free-running type is a CRSD characterized by a delay in the sleep–wake cycle of about an hour a day (Fig. 3). Those with free-running type report complaints of insomnia and/or excessive sleepiness depending on the phase of the endogenous circadian clock in relation to (attempted) sleep (ICSD-2, 2005). Functional impairments are mostly reported when waking coincides with a time when the circadian alerting signal is low. It is thought that those with free-running type are unable to entrain to the 24-h light–dark cycle, due to either damage of the input pathways of light or the ability to process other circadian time cues. Free-running type is most often seen in blind people (Sack et al., 1992), but has also been reported in sighted individuals (Boivin et al., 2003; Hayakawa et al., 2005). In a study of 52 blind individuals with and without circadian disruption, results indicated that performance was worse when participants were awake during the biological night (Lockley et al., 2008). There were no differences between the four groups (normally entrained, delayed, advanced, and free-running) on overall performance on an auditory four-choice serial reaction time task. There was also no difference between the normally entrained and free-running group, although there were significant differences between the advanced and delayed groups in performance relative to hours since waking. In a another study detailing a series of sighted individuals with free-running type, there was no

direct evidence of cognitive impairment, but several patients reported an inability to maintain schooling which is likely due to the unpredictable sleep–wake schedule (Hayakawa et al., 2005). Irregular sleep–wake type Irregular sleep–wake type is characterized by the lack of a clearly discernable sleep–wake rhythm. There are typically at least three short sleep episodes, ranging from 1 to 4 h, throughout the 24-h day (Fig. 3). Complaints are often a combination of sleep-onset insomnia, poor sleep maintenance at night, or excessive daytime sleepiness with frequent napping (ICSD-2, 2005). Irregular sleep–wake disorder is most common in institutionalized elderly and in those with cognitive impairment (Hoogendijk et al., 1996; Palm et al., 1991; Witting et al., 1990). There is some evidence that in demented elderly there are neuroanatomical changes in the SCN that could explain the circadian disruption (Swaab et al., 1985). In a group of home-dwelling older adults, Oosterman et al. (2009) reported that fragmentation of the rest–activity cycle (Fig. 4) is associated with cognitive performance impairment. Specifically, those with high intradaily variability in rest–activity had worse cognitive function than those with low intradaily variability in rest–activity (Fig. 4; Oosterman et al., 2009). Intervention studies using a combination of melatonin and bright light administration have shown that improvements in the sleep–wake cycle are associated with cognitive function improvement in institutionalized elderly with dementia. The mechanism of the improvements in cognitive function is unclear and may be the result of both improved sleep and/or circadian alignment (Riemersma-van der Lek et al., 2008). Shift-work disorder Traditional work schedules are typically 40 h per week performed during the daylight hours from between 8 and 9 a.m. until 4–5 p.m. However, many

09:00

15:00

21:00

03:00

09:00

15:00

21:00

03:00

09:00

15:00

21:00

03:00

09:00

09:00

15:00

21:00

03:00

09:00

15:00

21:00

03:00

09:00

15:00

21:00

03:00

09:00

03:00

21:00

15:00

09:00

03:00

21:00

15:00

09:00

03:00

21:00

15:00

09:00

03:00

21:00

15:00

60 50 40 30 20 10 0 09:00

Minutes with activity

12

60 50 40 30

03:00

21:00

15:00

09:00

03:00

21:00

15:00

09:00

03:00

21:00

15:00

09:00

03:00

21:00

15:00

20 10 0 09:00

Minutes with activity

Time of day

Time of day Fig. 4. Representative circadian rest–activity rhythm in two older adults with low (upper panel) and high (lower panel) intradaily variability (IV) in activity. The upper panel displays the rest–activity rhythm of a participant with low IV who showed above average cognitive test performance. The lower panel displays the rest–activity rhythm of a participant with high IV who performed below average on the cognitive tests. Figure taken from Oosterman et al. (2009) with permission.

industries operate outside these times. As a result of economic and customer demand, a range of different work schedules have evolved. Shift work is a term that applies to a wide variety of nonstandard work schedules. Sleep problems and sleepiness are most commonly seen in relation to night shifts (Åkerstedt, 1984) and early morning shifts (Kecklund et al., 1997; Waterhouse et al., 1993) and usually persist for the duration of the work period. But sleep problems can persist on days off when the worker attempts to return to a conventional sleep–wake schedule. SWD is characterized by complaints of insomnia, excessive sleepiness, and impaired performance in relation to the work schedule (ICSD-2, 2005). SWD occurs in vulnerable individuals because the endogenous circadian clock is out of synchrony with the sleep–wake cycle (Fig. 3).

The consequences of the fatigue (sleepiness) associated with shift work not only affect the individual but also society as a whole. Many serious transportation-related accidents have been attributed to fatigue, including airline pilots failing to properly monitor flight instruments, railroad workers falling asleep during their shift, truck drivers falling asleep at the wheel, and ships running aground due to poor vigilance. These accidents have resulted in serious injury or even death and can have significant personal and economic costs (Lauber and Kayten, 1988; Leger, 1994; Mitler et al., 1988). In most cases, these accidents have several common factors including extended hours of work, short sleep prior to the accident, and they occurred in the early morning or mid-afternoon hours (coinciding with the circadian low points for alertness/performance).

13

While there are limited data on the direct impact of SWD on cognitive function (Czeisler et al., 2005; Gumenyuk et al., 2010), there are more reports on the cognitive impairment associated with shift work per se (Åkerstedt, 2007). The impairment reported in shift work is believed to be due to two major factors; sleep loss and being awake during the circadian low point for alertness. The sleep of shift workers is often 1–4 h shorter than that for day workers (Åkerstedt, 2003; Drake et al., 2004), due to sleeping when the circadian alerting signal is high. This sleep disruption can result in the buildup of a substantial sleep debt by the end of the shift rotation. Studies clearly show that this type of repetitive sleep loss results in significant cognitive impairment (Belenky et al., 2003; Van Dongen et al., 2003). In combination, the sleep debt and circadian disruption mean that shift workers are at particular risk for performance deficits that may increase risk for serious accidents or injury depending on the profession. For this reason, many at-risk industries regulate the hours of service for shift workers, for example, the airline industry, trucking, rail, shipping, and more recently the medical profession. An area that is not regulated and is often overlooked is the commute home for workers following the end of a work period. There is evidence to suggest that the incidence of traffic accidents following extended work hours and night shifts is quite high (Barger et al., 2005). Driving while fatigued not only puts the workers themselves at risk but also endangers the lives of others on the road. There are several studies that indicate the performance impairment following 24 h of sustained wakefulness is comparable to that seen at a blood alcohol concentration of 0.08% (above or at the legal limit for most countries; Dawson and Reid, 1997; Lamond and Dawson, 1999). This duration of wakefulness is not uncommon in shift workers on their way home after the first night shift (Knauth and Rutenfranz, 1980). The data for actual job performance in real shift workers are limited due to the difficulty in

obtaining reliable data. The bulk of the evidence for performance impairments due to shift work is from either simulated or laboratory-based performance tasks in the field. There are a few classic studies that report performance on tasks during the work period for night workers: for example, errors on meter reading for gas company workers peak at night (Bjerner et al., 1955), and the speed at which telephone operators connected calls was reduced at night (Brown, 1949). The bulk of the data is from simulated shiftwork studies or from studies that use laboratorybased performance tasks in the field to determine impairment levels. The problem with both these types of studies is that results vary depending on the task, much as impairment at work depends on the task and the profession. The types of task used also vary from those that test simple reaction time (Ferguson et al., 2010), executive function (Wojtczak-Jaroszowa et al., 1978), and simulation of work tasks (Gillberg et al., 1996). While slowed reaction time is a cause for concern for a truck driver, it may not be as much of a safety issue for someone working in an office. There is a need for more data on whether there are significant impairments to actual work performance, in particular, outcome measures that occur before an accident or catastrophic error. In addition, due to the huge variability in the timing (day, night, evening, morning, rotating) and hours of work (8, 12, 24, etc.), it is difficult to generalize the findings from one study to another work schedule. Given this in general, the principles of sleep and circadian biology and the response to sleep loss and disruption can be applied to estimate impairment in shift work. Several models exist that attempt to do this and are described in the Proceedings of the Fatigue and performance modeling workshop (2004). Individual factors have been identified that are associated with reduced tolerance, poor sleep, and performance impairment in shift workers including age, circadian preference, and psychosocial factors (Chung et al., 2009; Folkard et al., 1978; Härmä et al., 1994; Reid and Dawson, 2001).

14

Since there are limited data on the demographics of SWD, it is unclear whether these factors play a role in the presentation of this disorder (Czeisler et al., 2009; Drake et al., 2004). Age is one of the risk factors for greater performance impairment in shift workers, although there may be some degree of self-selection, as a worker with SWD may not continue to do shift work. Or, it could be that as shift workers age, they are more likely to develop SWD. Exposure to shift work may also impact cognitive performance (Rouch et al., 2005). A study by Rouch and colleagues suggests that men with 10–20 years of employment as a shift worker have worse memory performance, even after adjusting for sleep duration, compared to those with 1–4 years of shift work. They also suggest that this impairment may be reversed 4 or more years after the cessation of the shift-work schedules (Rouch et al., 2005). The risks associated with shift work can be managed in several ways, including speeding the adaption of the circadian clock to the shift schedule using administration of bright light or exogenous melatonin, reducing the homeostatic drive for sleep using naps, or improving alertness with bright light and pharmacology (i.e., caffeine, modafinil, and armodafinil). While the chronic use of caffeine is considered a socially acceptable way of improving alertness, the use of prescription medications regulated by the US Federal Drug Administration (FDA) such as modafinil or armodafinil on a chronic basis is somewhat more controversial in the shift-work arena. Studies examining the cognitive impairment specifically in those with SWD are limited, the only study that directly compares those with SWD to shift workers without SWD and day workers is by Gumenyuk et al. (2010). This study used auditory event-related brain potential (ERP) measures (mismatch negativity and P3a) to evaluate the impact of SWD on the neurophysiology of memory and attention (Gumenyuk et al., 2010). They found that there is significant impairment in those with SWD compared to shift

workers without SWD and day-working controls, and that this impairment is similar to that seen with sleep deprivation. Another interesting finding was that even when given the same sleep opportunity (8 h in bed during the day), polysomnographically recorded sleep duration was significantly shorter in those with SWD compared to shift workers without SWD (5.96 vs. 7.0 h). This reduction in sleep duration was the result of a less consolidated sleep period (more wake after sleep onset). Data on the cognitive function of those with SWD are also available from treatment studies. For example, the treatment of SWD with alerting medications like modafinil (Czeisler et al., 2005) and armodafinil (Czeisler et al., 2009) has resulted in significant improvements in cognitive performance (including simple reaction time, psychomotor vigilance test, attention, delayed word recall) and sleepiness (multiple sleep latency test; Czeisler et al., 2005, 2009). However, while the increases in SOL were significantly different than placebo, the resulting SOL was still in the pathological range (mean of < 6 min). Since these studies only examine those with SWD, it remains unclear whether the cognitive impairment in this group of participants is any worse than shift workers without SWD or day workers.

Jet lag Jet lag is the result of the external environment being temporarily altered in relation to the internal clock by traveling across time zones (Fig. 3). As the endogenous circadian timing system adapts slowly to new time cues, the phase relationship between biological rhythms and external time cues are out of synchronization for a period of time. This disturbance of circadian rhythms has been shown to impair physical and psychological health (Boulos et al., 1995; Leger et al., 1993). For the first few days, people who travel across at least two time zones complain of a variety of subjective symptoms including insomnia, excessive

15

daytime sleepiness, fatigue, headache, performance impairments, and gastrointestinal upset (Boulos et al., 1995; ICSD-2, 2005). The severity of symptoms often depends on the number of time zones traveled and the direction of travel, with eastward travel (requiring a phase advance of the circadian clock) taking the longest to adapt to. Travel in the eastward direction typically leads to difficulty initiating sleep, while westward travel is more often associated with difficulty staying asleep (Boulos et al., 1995). There is substantial data of the impact of jet lag on physical/sports performance (Reilly et al., 2005) and less so on cognitive performance (Beh and McLaughlin, 1997; Klein et al., 1970, 1980). After a 6-h time zone change, impairments in reaction time, letter cancellation, and digit summation tasks were reported. There is a single study that even suggests only eastward flight caused impairments in cognitive performance (Hauty, 1967). A study of flight attendants who had repeated exposure to jet lag indicates that there is an impact on cognitive function (Cho, 2001; Cho et al., 2000). A comparison of flight attendants with and without jet lag exposure indicated that there were lower cognitive performance and higher cortisol and brain atrophy for those flight attendants with 4 or more years of exposure. Concluding remarks Disruption of the balance between the circadian and homeostatic control of sleep–wake and other behaviors by CRSDs can have a significant impact on cognitive function on a chronic basis. These impairments in cognition not only negatively impact quality of life and daily functioning of those with these disorders, but they also pose a potential risk to the public. Given this, it is surprising that there are little data available on the specifics of how these disorders influence cognitive performance. Future studies comparing various aspects (memory, attention, executive function) of performance in those with these

conditions to normal controls will be useful in determining not only what the impairment level is but whether individuals with these conditions have some inherent difference in basic cognitive functioning. An extension of these studies would be to develop treatment strategies for the disorders that target specific areas of functioning. Abbreviations ASPD CRSD DSPD ICSD-2 SCN SWD

advanced sleep-phase disorder circadian rhythm sleep disorder delayed sleep-phase disorder International Classification of Sleep Disorders, 2nd edition suprachiasmatic nucleus shift-work disorder

References Åkerstedt, T. (1984). Work schedules and sleep. Experientia, 40, 417–422. Åkerstedt, T. (2003). Shift work and disturbed sleep/wakefulness. Occupational Medicine (London), 53, 89–94. Åkerstedt, T. (2007). Altered sleep/wake patterns and mental performance. Physiology and Behavior, 90, 209–218. Andrade, M. M., Benedito-Silva, A. A., & Menna-Barreto, L. (1992). Correlations between morningness-eveningness character, sleep habits and temperature rhythm in adolescents. Brazilian Journal of Medical and Biological Research, 25, 835–839. Aoki, H., Ozeki, Y., & Yamada, N. (2001). Hypersensitivity of melatonin suppression in response to light in patients with delayed sleep phase syndrome. Chronobiology International, 18, 263–271. Aschoff, J., & Wever, R. (1976). Human circadian rhythms: A multioscillatory system. Federation Proceedings, 35, 232–236. Baehr, E. K., Eastman, C. I., Revelle, W., Olson, S. H., Wolfe, L. F., & Zee, P. C. (2003). Circadian phase-shifting effects of nocturnal exercise in older compared with young adults. American Journal of Physiology: Regulatory, Integrative and Comparative Physiology, 284, R1542–R1550. Baehr, E. K., Revelle, W., & Eastman, C. I. (2000). Individual differences in the phase and amplitude of the human circadian temperature rhythm: With an emphasis on morningness-eveningness. Journal of Sleep Research, 9, 117–127.

16 Barger, L. K., Cade, B. E., Ayas, N. T., Cronin, J. W., Rosner, B., Speizer, F. E., et al. (2005). Extended work shifts and the risk of motor vehicle crashes among interns. The New England Journal of Medicine, 352, 125–134. Barger, L. K., Wright, K. P., Jr., Hughes, R. J., & Czeisler, C. A. (2004). Daily exercise facilitates phase delays of circadian melatonin rhythm in very dim light. American Journal of Physiology: Regulatory, Integrative and Comparative Physiology, 286, R1077–R1084. Beersma, D., & Gordijn, M. (2007). Circadian control of the sleep-wake cycle. Physiology and Behavior, 90, 190–195. Beh, H. C., & McLaughlin, P. (1997). Effect of long flights on the cognitive performance of air crew. Perceptual and Motor Skills, 84, 319–322. Belenky, G., Wesensten, N. J., Thorne, D. R., Thomas, M. L., Sing, H. C., Redmond, D. P., et al. (2003). Patterns of performance degradation and restoration during sleep restriction and subsequent recovery: A sleep dose-response study. Journal of Sleep Research, 12, 1–12. Berson, D. M., Castrucci, A. M., & Provencio, I. (2010). Morphology and mosaics of melanopsin-expressing retinal ganglion cell types in mice. The Journal of Comparative Neurology, 518, 2405–2422. Berson, D. M., Dunn, F. A., & Takao, M. (2002). Phototransduction by retinal ganglion cells that set the circadian clock. Science, 295, 1070–1073. Bjerner, B., Holm, Å., & Swensson, Å. (1955). Diurnal variation of mental performance. A study of three-shift workers. British Journal of Industrial Medicine, 12, 103–110. Boivin, D. B., James, F. O., Santo, J. B., Caliyurt, O., & Chalk, C. (2003). Non-24-hour sleep-wake syndrome following a car accident. Neurology, 60, 1841–1843. Boulos, Z., Campbell, S. S., Lewy, A. J., Terman, M., Dijk, D. J., & Eastman, C. I. (1995). Light treatment for sleep disorders: Consensus report. VII. Jet lag. Journal of Biological Rhythms, 10, 167–176. Brainard, G. C., Lewy, A. J., Menaker, M., Fredrickson, R. H., Miller, L. S., Weleber, R. G., et al. (1988). Dose-response relationship between light irradiance and the suppression of plasma melatonin in human volunteers. Brain Research, 454, 212–218. Breithaupt, H., Hildebrandt, G., Dohre, D., Josch, R., Sieber, U., & Werner, M. (1978). Tolerance to shift of sleep, as related to the individual's circadian phase position. Ergonomics, 21, 767–774. Breithaupt, H., Hildebrandt, G., & Werner, A. (1981). Circadian type questionnaire and objective circadian characteristics. In A. Reinberg, N. Vieux & P. Andlauer (Eds.), Night and shift work: Biological and social aspects (pp. 435–440). Oxford: Pergamon Press. Brown, R. C. (1949). The day and night performance of teleprinter switchboard operators. Journal of Occupational Psychology, 23, 121–126.

Burgess, H. J., Revell, V. L., & Eastman, C. I. (2008). A three pulse phase response curve to three milligrams of melatonin in humans. Journal of Physiology, 586, 639–647. Burgess, H. J., Revell, V. L., Molina, T. A., & Eastman, C. I. (2010). Human phase response curves to three days of daily melatonin: 0.5 mg versus 3.0 mg. The Journal of Clinical Endocrinology and Metabolism, 95, 3325–3331. Cajochen, C. (2007). Alerting effects of light. Sleep Medicine Reviews, 11, 453–464. Cajochen, C., Khalsa, S. B., Wyatt, J. K., Czeisler, C. A., & Dijk, D. J. (1999). EEG and ocular correlates of circadian melatonin phase and human performance decrements during sleep loss. The American Journal of Physiology, 277, R640–R649. Cajochen, C., Munch, M., Kobialka, S., Krauchi, K., Steiner, R., Oelhafen, P., et al. (2005). High sensitivity of human melatonin, alertness, thermoregulation, and heart rate to short wavelength light. The Journal of Clinical Endocrinology and Metabolism, 90, 1311–1316. Chang, A. M., Reid, K. J., Gourineni, R., & Zee, P. C. (2009). Sleep timing and circadian phase in delayed sleep phase syndrome. Journal of Biological Rhythms, 24, 313–321. Cho, K. (2001). Chronic 'jet lag' produces temporal lobe atrophy and spatial cognitive deficits. Nature Neuroscience, 4, 567–568. Cho, K., Ennaceur, A., Cole, J. C., & Suh, C. K. (2000). Chronic jet lag produces cognitive deficits. The Journal of Neuroscience, 20, RC66. Chung, M. H., Chang, F. M., Yang, C. C., Kuo, T. B., & Hsu, N. (2009). Sleep quality and morningness-eveningness of shift nurses. Journal of Clinical Nursing, 18, 279–284. Czeisler, C. A., Duffy, J. F., Shanahan, T. L., Brown, E. N., Mitchell, J. F., Rimmer, D. W., et al. (1999). Stability, precision, and near-24-hour period of the human circadian pacemaker. Science, 284, 2177–2181. Czeisler, C. A., Walsh, J. K., Roth, T., Hughes, R. J., Wright, K. P., Kingsbury, L., et al. (2005). Modafinil for excessive sleepiness associated with shift-work sleep disorder. The New England Journal of Medicine, 353, 476–486. Czeisler, C. A., Walsh, J. K., Wesnes, K. A., Arora, S., & Roth, T. (2009). Armodafinil for treatment of excessive sleepiness associated with shift work disorder: A randomized controlled study. Mayo Clinic Proceedings, 84, 958–972. Dawson, D., & Reid, K. (1997). Fatigue, alcohol and performance impairment. Nature, 388, 235. Dibner, C., Schibler, U., & Albrecht, U. (2010). The mammalian circadian timing system: Organization and coordination of central and peripheral clocks. Annual Review of Physiology, 72, 517–549. Dijk, D. J., & Czeisler, C. A. (1994). Paradoxical timing of the circadian rhythm of sleep propensity serves to consolidate sleep and wakefulness in humans. Neuroscience Letters, 166, 63–68.

17 Dijk, D. J., Duffy, J. F., & Czeisler, C. A. (1992). Circadian and sleep/wake dependent aspects of subjective alertness and cognitive performance. Journal of Sleep Research, 1, 112–117. Drake, C. L., Roehrs, T., Richardson, G., Walsh, J. K., & Roth, T. (2004). Shift work sleep disorder: Prevalence and consequences beyond that of symptomatic day workers. Sleep, 27, 1453–1462. Dubocovich, M. L. (2007). Melatonin receptors: Role on sleep and circadian rhythm regulation. Sleep Medicine, 8(Suppl. 3), 34–42. Duffy, J. F., & Wright, K. P.Jr., (2005). Entrainment of the human circadian system by light. Journal of Biological Rhythms, 20, 326–338. Edgar, D. M., Dement, W. C., & Fuller, C. A. (1993). Effect of SCN lesions on sleep in squirrel monkeys: Evidence for opponent processes in sleep-wake regulation. The Journal of Neuroscience, 13, 1065–1079. Ferguson, S. A., Paech, G. M., Dorrian, J., Roach, G. D., & Jay, S. M. (2010). Performance on a simple response time task: Is sleep or work more important for miners? Applied Ergonomics, 42, 210–213. Folkard, S., & Monk, T. (1979). Shiftwork and performance. Human Factors, 21, 483–492. Folkard, S., Monk, T. H., & Lobban, M. C. (1978). Short and long-term adjustment of circadian rhythms in 'permanent' night nurses. Ergonomics, 21, 785–799. Foret, J., Benoit, O., & Royant-Parola, S. (1982). Sleep schedules and peak times of oral temperature and alertness in morning and evening 'types'. Ergonomics, 25, 821–827. Freedman, M. S., Lucas, R. J., Soni, B., von Schantz, M., Munoz, M., David-Gray, Z., et al. (1999). Regulation of mammalian circadian behavior by non-rod, non-cone, ocular photoreceptors. Science, 284, 502–504. Gillberg, M., Kecklund, G., & Åkerstedt, T. (1996). Sleepiness and performance of professional drivers in a truck simulator—Comparisons between day and night driving. Journal of Sleep Research, 5, 12–15. Goel, N. (2005). Late-night presentation of an auditory stimulus phase delays human circadian rhythms. American Journal of Physiology: Regulatory, Integrative and Comparative Physiology, 289, R209–R216. Gooley, J. J., Lu, J., Chou, T. C., Scammell, T. E., & Saper, C. B. (2001). Melanopsin in cells of origin of the retinohypothalamic tract. Nature Neuroscience, 4, 1165. Gumenyuk, V., Roth, T., Korzyukov, O., Jefferson, C., Kick, A., Spear, L., et al. (2010). Shift work sleep disorder is associated with an attenuated brain response of sensory memory and an increased brain response to novelty: An ERP study. Sleep, 33, 703–713. Härmä, M. I., Hakola, T., Åkerstedt, T., & Laitinen, J. T. (1994). Age and adjustment to night work. Occupational and Environmental Medicine, 51, 568–573.

Harrington, M. E. (1997). The ventral lateral geniculate nucleus and the intergeniculate leaflet: Interrelated structures in the visual and circadian systems. Neuroscience and Biobehavioral Reviews, 21, 705–727. Hattar, S., Liao, H. W., Takao, M., Berson, D. M., & Yau, K. W. (2002). Melanopsin-containing retinal ganglion cells: Architecture, projections, and intrinsic photosensitivity. Science, 295, 1065–1070. Hauty, G. T. (1967). Individual differences in phase shifts of the human circadian system and performance deficit. Life Sciences and Space Research, 5, 135–147. Hayakawa, T., Uchiyama, M., Kamei, Y., Shibui, K., Tagaya, H., Asada, T., et al. (2005). Clinical analyses of sighted patients with non-24-hour sleep-wake syndrome: A study of 57 consecutively diagnosed cases. Sleep, 28, 945–952. Hildebrandt, G., & Stratmann, I. (1979). Circadian system response to night work in relation to the individual circadian phase position. International Archives of Occupational and Environmental Health, 43, 73–83. Hoogendijk, W. J., van Someren, E. J., Mirmiran, M., Hofman, M. A., Lucassen, P. J., Zhou, J. N., & Swaab, D. F. (1996). Circadian rhythm-related behavioral disturbances and structural hypothalamic changes in Alzheimer's disease. International Psychogeriatrics, 8(Suppl. 3), 245–252 discussion 269–272. Horne, J. A., & Östberg, O. (1976). A self-assessment questionnaire to determine morningness-eveningness in human circadian rhythms. International Journal of Chronobiology, 4, 97–110. ICSD-2. (2005). The International Classification of Sleep Disorders: Diagnostic and Coding Manual. Westchester, IL: American Academy of Sleep Medicine. Johnson, M. P., Duffy, J. F., Dijk, D. J., Ronda, J. M., Dyal, C. M., & Czeisler, C. A. (1992). Short-term memory, alertness and performance: A reappraisal of their relationship to body temperature. Journal of Sleep Research, 1, 24–29. Kecklund, G., Åkerstedt, T., & Lowden, A. (1997). Morning work: Effects of early rising on sleep and alertness. Sleep, 20, 215–223. Kerkhof, G. A. (1985). Inter-individual differences in the human circadian system: A review. Biological Psychology, 20, 83–112. Kerkhof, G. A., & Van Dongen, H. P. (1996). Morning-type and evening-type individuals differ in the phase position of their endogenous circadian oscillator. Neuroscience Letters, 218, 153–156. Khalsa, S. B., Jewett, M. E., Cajochen, C., & Czeisler, C. A. (2003). A phase response curve to single bright light pulses in human subjects. Journal of Physiology, 549, 945–952. Klein, K. E., Bruner, H., Holtmann, H., Rehme, H., Stolze, J., Steinhoff, W. D., et al. (1970). Circadian rhythm of pilots'

18 efficiency and effects of multiple time zone travel. Aerospace Medicine, 41, 125–132. Klein, K. E., Wegmann, H. M., Scheving, L., & Halberg, F. (1980). The effect of transmeridian and transequatorial air travel on psychological well-being and performance. In L. S. F. Halberg (Ed.), Chronobiology: Principles and applications to shifts in schedules (pp. 339–352). Alphen aan den Rijn, Netherlands: Sijthoff and Noordhoff. Kleitman, N. (1933). Diurnal variation in performance. The American Journal of Physiology, 104, 449–456. Kleitman, N. (1963). Sleep and wakefulness. Chicago: The University of Chicago Press. Kleitman, N., & Jackson, D. P. (1950). Body temperature and performance under different routines. Journal of Applied Physiology, 3, 309–328. Knauth, P., & Rutenfranz, J. (1980). Night and shiftwork. Biological and social aspects. In A. Reinberg, N. Vieux & P. Andlauer (Eds.), Advances in the biosciences (pp. 161–168). Oxford: Pergamon Press. Kripke, D. F., Elliott, J. A., Youngstedt, S. D., & Rex, K. M. (2007). Circadian phase response curves to light in older and young women and men. Journal of Circadian Rhythms, 5, 4. Lack, L., Wright, H., Kemp, K., & Gibbon, S. (2005). The treatment of early-morning awakening insomnia with 2 evenings of bright light. Sleep, 28, 616–623. Lamond, N., & Dawson, D. (1999). Quantifying the performance impairment associated with fatigue. Journal of Sleep Research, 8, 255–262. Lauber, J. K., & Kayten, P. J. (1988). Sleepiness, circadian dysrhythmia, and fatigue in transportation system accidents. Sleep, 11, 503–512. Lavie, P. (1986). Ultrashort sleep-waking schedule. III. 'Gates' and 'forbidden zones' for sleep. Electroencephalography and Clinical Neurophysiology, 63, 414–425. Lavie, P. (1997). Melatonin: Role in gating nocturnal rise in sleep propensity. Journal of Biological Rhythms, 12, 657–665. Leger, D. (1994). The cost of sleep-related accidents: A report for the National Commission on Sleep Disorders Research. Sleep, 17, 84–93. Leger, D., Badet, D., & de La Giclais, B. (1993). The prevalence of jet-lag among 507 traveling businessmen. Sleep Research, 22, 409. Lewy, A. L., Ahmed, S., Jackson, J. M. L., & Sack, R. L. (1992). Melatonin shifts human circadian rhythms according to a phase-response curve. Chronobiology International, 9, 380–392. Lewy, A. J., Bauer, V. K., Ahmed, S., Thomas, K. H., Cutler, N. L., Singer, C. M., et al. (1998). The human phase response curve (PRC) to melatonin is about 12 hours out of phase with the PRC to light. Chronobiology International, 15, 71–83. Lockley, S. W., Brainard, G. C., & Czeisler, C. A. (2003). High sensitivity of the human circadian melatonin rhythm to

resetting by short wavelength light. The Journal of Clinical Endocrinology and Metabolism, 88, 4502–4505. Lockley, S. W., Dijk, D. J., Kosti, O., Skene, D. J., & Arendt, J. (2008). Alertness, mood and performance rhythm disturbances associated with circadian sleep disorders in the blind. Journal of Sleep Research, 17, 207–216. Lockley, S. W., Evans, E. E., Scheer, F. A., Brainard, G. C., Czeisler, C. A., & Aeschbach, D. (2006). Short-wavelength sensitivity for the direct effects of light on alertness, vigilance, and the waking electroencephalogram in humans. Sleep, 29, 161–168. Markwald, R. R., Lee-Chiong, T. L., Burke, T. M., Snider, J. A., & Wright, K. P.Jr., (2010). Effects of the melatonin MT-1/MT-2 agonist ramelteon on daytime body temperature and sleep. Sleep, 33, 825–831. Mills, J. N., Minors, D. S., & Waterhouse, J. M. (1978). Adaptation to abrupt time shifts of the oscillator(s) controlling human circadian rhythms. Journal of Physiology, 285, 455–470. Mitler, M. M., Carskadon, M. A., Czeisler, C. A., Dement, W. C., Dinges, D. F., & Graeber, R. C. (1988). Catastrophes, sleep, and public policy: Consensus report. Sleep, 11, 100–109. Mongrain, V., Carrier, J., & Dumont, M. (2006a). Circadian and homeostatic sleep regulation in morningnesseveningness. Journal of Sleep Research, 15, 162–166. Mongrain, V., Carrier, J., & Dumont, M. (2006b). Difference in sleep regulation between morning and evening circadian types as indexed by antero-posterior analyses of the sleep EEG. The European Journal of Neuroscience, 23, 497–504. Moore, R. Y. (1996). Neural control of the pineal gland. Behavioural Brain Research, 73, 125–130. Oosterman, J. M., van Someren, E. J., Vogels, R. L., Van Harten, B., & Scherder, E. J. (2009). Fragmentation of the rest-activity rhythm correlates with age-related cognitive deficits. Journal of Sleep Research, 18, 129–135. Ozaki, S., Uchiyama, M., Shirakawa, S., & Okawa, M. (1996). Prolonged interval from body temperature nadir to sleep offset in patients with delayed sleep phase syndrome. Sleep, 19, 36–40. Palm, L., Blennow, G., & Wetterberg, L. (1991). Correction of non-24-hour sleep/wake cycle by melatonin in a blind retarded boy. Annals of Neurology, 29, 336–339. Phipps-Nelson, J., Redman, J. R., Dijk, D. J., & Rajaratnam, S. M. (2003). Daytime exposure to bright light, as compared to dim light, decreases sleepiness and improves psychomotor vigilance performance. Sleep, 26, 695–700. Proceedings of the Fatigue and Performance Modeling Workshop, June 13–14, 2002, Seattle, Washington, USA. Aviation, Space and Environmental Medicine, 75, A1–A199. Rajaratnam, S. M., Polymeropoulos, M. H., Fisher, D. M., Roth, T., Scott, C., Birznieks, G., et al. (2009). Melatonin agonist tasimelteon (VEC-162) for transient insomnia after

19 sleep-time shift: Two randomised controlled multicentre trials. Lancet, 373, 482–491. Reid, K. J., Chang, A. M., Dubocovich, M. L., Turek, F. W., Takahashi, J. S., & Zee, P. C. (2001). Familial advanced sleep phase syndrome. Archives of Neurology, 58, 1089–1094. Reid, K., & Dawson, D. (2001). Comparing performance on a simulated 12 hour shift rotation in young and older subjects. Occupational and Environmental Medicine, 58, 58–62. Reid, K., Van den Heuvel, C., & Dawson, D. (1996). Day-time melatonin administration: Effects on core temperature and sleep onset latency. Journal of Sleep Research, 5, 150–154. Reilly, T., Waterhouse, J., & Edwards, B. (2005). Jet lag and air travel: Implications for performance. Clinical Journal of Sports Medicine, 24, 367–380 xii. Richardson, G. S., Zee, P. C., Wang-Weigand, S., Rodriguez, L., & Peng, X. (2008). Circadian phase-shifting effects of repeated ramelteon administration in healthy adults. Journal of Clinical Sleep Medicine, 4, 456–461. Riemersma-van der Lek, R. F., Swaab, D. F., Twisk, J., Hol, E. M., Hoogendijk, W. J., & Van Someren, E. J. (2008). Effect of bright light and melatonin on cognitive and noncognitive function in elderly residents of group care facilities: A randomized controlled trial. The Journal of the American Medical Association, 299, 2642–2655. Roenneberg, T., Kuehnle, T., Juda, M., Kantermann, T., Allebrandt, K., Gordijn, M., et al. (2007). Epidemiology of the human circadian clock. Sleep Medicine Reviews, 11, 429–438. Rogers, N. L., Kennaway, D. J., & Dawson, D. (2003). Neurobehavioural performance effects of daytime melatonin and temazepam administration. Journal of Sleep Research, 12, 207–212. Rogers, N. L., Phan, O., Kennaway, D. J., & Dawson, D. (1998). Effect of daytime oral melatonin administration on neurobehavioral performance in humans. Journal of Pineal Research, 25, 47–53. Rosenthal, N. E., Joseph-Vanderpool, J. R., Levendosky, A. A., Johnston, S. H., Allen, R., Kelly, K. A., et al. (1990). Phaseshifting effects of bright morning light as treatment for delayed sleep phase syndrome. Sleep, 13, 354–361. Rouch, I., Wild, P., Ansiau, D., & Marquie, J. C. (2005). Shiftwork experience, age and cognitive performance. Ergonomics, 48, 1282–1293. Ruby, N. F., Brennan, T. J., Xie, X., Cao, V., Franken, P., Heller, H. C., et al. (2002). Role of melanopsin in circadian responses to light. Science, 298, 2211–2213. Sack, R. L., Lewy, A. J., Blood, M. L., Keith, L. D., & Nakagawa, H. (1992). Circadian rhythm abnormalities in totally blind people: Incidence and clinical significance. The Journal of Clinical Endocrinology and Metabolism, 75, 127–134. Shibui, K., Uchiyama, M., & Okawa, M. (1999). Melatonin rhythms in delayed sleep phase syndrome. Journal of Biological Rhythms, 14, 72–76.

Smith, M. R., & Eastman, C. I. (2009). Phase delaying the human circadian clock with blue-enriched polychromatic light. Chronobiology International, 26, 709–725. Smith, M. R., Revell, V. L., & Eastman, C. I. (2009). Phase advancing the human circadian clock with blue-enriched polychromatic light. Sleep Medicine, 10, 287–294. Swaab, D. F., Fliers, E., & Partiman, T. S. (1985). The suprachiasmatic nucleus of the human brain in relation to sex, age and senile dementia. Brain Research, 342, 37–44. Szeinberg, A., Borodkin, K., & Dagan, Y. (2006). Melatonin treatment in adolescents with delayed sleep phase syndrome. Clinical Pediatrics (Philadelphia), 45, 809–818. Uchiyama, M., Okawa, M., Shibui, K., Kim, K., Tagaya, H., Kudo, Y., et al. (2000). Altered phase relation between sleep timing and core body temperature rhythm in delayed sleep phase syndrome and non-24-hour sleep-wake syndrome in humans. Neuroscience Letters, 294, 101–104. Uchiyama, M., Okawa, M., Shibui, K., Liu, X., Hayakawa, T., Kamei, K., et al. (2000). Poor compensatory function for sleep loss as a pathogenic factor in patients with delayed sleep phase syndrome. Sleep, 23, 553–558. Van Dongen, H. P., Kerkhof, G. A., & Souverijn, J. H. (1998). Absence of seasonal variation in the phase of the endogenous circadian rhythm in humans. Chronobiology International, 15, 623–632. Van Dongen, H. P., 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, 117–126. Vandewalle, G., Balteau, E., Phillips, C., Degueldre, C., Moreau, V., Sterpenich, V., et al. (2006). Daytime light exposure dynamically enhances brain responses. Current Biology, 16, 1616–1621. Warman, V. L., Dijk, D. J., Warman, G. R., Arendt, J., & Skene, D. J. (2003). Phase advancing human circadian rhythms with short wavelength light. Neuroscience Letters, 342, 37–40. Waterhouse, J. M., Folkard, S., & Minors, D. S. (1993). Effects of a change in shift-work on health. Occup Med (Lond)., 43, 167–168. Waterhouse, J., Folkard, S., Van Dongen, H., Minors, D., Owens, D., Kerkhof, G., et al. (2001). Temperature profiles, and the effect of sleep on them, in relation to morningnesseveningness in healthy female subjects. Chronobiology International, 18, 227–247. Webb, W. B., & Bonnet, M. H. (1978). The sleep of 'morning' and 'evening' types. Biological Psychology, 7, 29–35. Witting, W., Kwa, I. H., Eikelenboom, P., Mirmiran, M., & Swaab, D. F. (1990). Alterations in the circadian rest-activity rhythm in aging and Alzheimer's disease. Biological Psychiatry, 27, 563–572. Wojtczak-Jaroszowa, J., Makowska, Z., Rzepecki, H., Banaszkiewicz, A., & Romejko, A. (1978). Changes in

20 psychomotor and mental task performance following physical work in standard conditions, and in a shift-working situation. Ergonomics, 21, 801–809. Wong, K. Y., Dunn, F. A., Graham, D. M., & Berson, D. M. (2007). Synaptic influences on rat ganglion-cell photoreceptors. Journal of Physiology, 582, 279–296. Wyatt, J. K., Dijk, D. J., Ritz-de Cecco, A., Ronda, J. M., & Czeisler, C. A. (2006). Sleep-facilitating effect of exogenous

melatonin in healthy young men and women is circadianphase dependent. Sleep, 29, 609–618. Zavada, A., Gordijn, M. C., Beersma, D. G., Daan, S., & Roenneberg, T. (2005). Comparison of the Munich chronotype questionnaire with the Horne-Östberg's morningnesseveningness score. Chronobiology International, 22, 267–278. Zee, P. C. (2008). Melantonin for the treatment of advanced sleep phase disorder. Sleep, 31, 923.

H. P. A. Van Dongen and G. A. Kerkhof (Eds.) Progress in Brain Research, Vol. 190 ISSN: 0079-6123 Copyright Ó 2011 Elsevier B.V. All rights reserved.

CHAPTER 2

Sleep–wake changes and cognition in neurodegenerative disease Sharon L. Naismith{,*, Simon J. G. Lewis{ and Naomi L. Rogers} { {

Healthy Brain Ageing Clinic, Ageing Brain Centre, Brain & Mind Research Institute, The University of Sydney, Sydney, NSW, Australia Parkinson’s Disease Research Clinic, Ageing Brain Centre, Brain & Mind Research Institute, The University of Sydney, Sydney, NSW, Australia } Chronobiology and Sleep Group, Brain & Mind Research Institute, The University of Sydney, Sydney, NSW, Australia

Abstract: With the increasing aging population, neurodegenerative disorders will become more common in clinical practice. These disorders involve multiple pathophysiological mechanisms that differentially affect cognition, mood, and physical functions. Possibly due to the involvement of common underlying neurobiological circuits, sleep and/or circadian (sleep–wake) changes are also common in this disease group. Of significance, sleep–wake changes are often a prodromal feature and are predictive of cognitive decline, psychiatric symptoms, quality of life, need for institutional care, and caregiver burden. Unfortunately, in neurodegenerative disease, few studies have included detailed polysomnography or neuropsychological assessments although some data indicate that sleep and neurocognitive features are related. Further studies are also required to address the effects of pharmacological and nonpharmacological treatments on cognitive functioning. Such research will hopefully lead to targeted early intervention approaches for cognitive decline in older people. Keywords: sleep; cognition; neurodegenerative disease; dementia; aging; depression.

Introduction With global increases in life expectancy and population size, neurodegenerative disorders affecting older adults will soon become a global health problem and will be more prominent in clinical practice.

*Corresponding author. Tel.: þ61 2 9351 0781; Fax: þ61 2 9351 0855 E-mail: [email protected] DOI: 10.1016/B978-0-444-53817-8.00002-5

21

22

Depending on the nature and distribution of pathology, these disorders present with a variety of distinct features including cognitive impairment (e.g., memory loss), psychiatric features (e.g., depression, psychosis), and physical signs (e.g., tremor). Patients with neurodegenerative conditions are also commonly affected by sleep–wake (i.e., sleep and/or circadian) disturbances, with estimates being as high as 70% for early stage dementia (Rongve et al., 2010). Of significance, they are recognized to be an indicator of poor prognosis and are associated with a higher incidence of cognitive and neuropsychiatric problems (Gagnon et al., 2008; Marion et al., 2008; Naismith et al., 2009b, 2010a; Rongve et al., 2010; Vendette et al., 2007). Sleep–wake disturbances are also linked to poor quality of life (Naismith et al., 2010c) as well as caregiver burden, health, and well-being (Gallagher-Thompson et al., 1992; Happe and Berger, 2002; Mills et al., 2009). Although sleep–wake disturbances in neurodegenerative disorders can relate to a range of pathophysiological mechanisms, dysfunction of neural circuitry is likely to play a key role (Gagnon et al., 2008). This is not surprising, as the principal regulators of arousal, circadian rhythms, and sleep reside within the central nervous system. While the hypothalamus and brainstem are critical regions, the sleep–wake system has numerous cortical projections and utilizes many neurotransmitters and modulatory hormones that overlap with those affected by neurodegenerative and neuropsychiatric diseases (Fuller et al., 2006; Saper et al., 2005; Wulff et al., 2010). Therefore, it is likely that specific sleep–wake, cognitive, and behavioral features of these diseases have shared neurobiological underpinnings. It is worth noting that holistic consideration of sleep perturbations in neurodegenerative diseases must also incorporate the various external influences on sleep, which contribute to large interand intraindividual variability. Thus, they may reflect not only changes to the neural circuitry but also other confounds such as pain, anxiety, medications, sedentary lifestyle, medical comorbidities, and changes to mental state. Conversely,

disruptions to the sleep–wake system have widespread effects on cognition, mood, metabolism, immune functioning, and medical morbidities. Thus, there are likely complex multifactorial relationships influencing the clinical manifestation of sleep–wake and cognitive disturbance (Gagnon et al., 2008; Wulff et al., 2010). In this chapter, we will firstly overview the sleep–wake and circadian changes that occur with normal aging, followed by an overview of sleep–wake changes in Alzheimer’s disease (AD), Parkinson’s disease (PD), Lewy body dementia (LBD), and frontotemporal dementia (FTD). We will then address both mild cognitive impairment (MCI) and late-life depression; clinical syndromes that are considered to be “at-risk” states for dementia. While the focus in this chapter will be on cognition, there is a general dearth of detailed studies linking sleep–wake disturbances with detailed neuropsychological measures. We will then highlight some of the pertinent literature linking sleep–wake and cognition, particularly memory, as these data provide important insights into neural mechanisms of memory consolidation during sleep as well as insights into potential intervention targets. Finally, we will provide a summary of some of the treatments for sleep–wake disorders in older people, with an emphasis on those that may have relevance for neuropsychiatric and neurodegenerative diseases affecting older adults. Sleep–wake changes in normal aging With increasing age, alterations in the circadian and sleep homeostatic systems occur, and are associated with circadian phase advances, poor sleep quality, increased sleep disturbance, and insomnia. Sleep architecture changes include increased wake after sleep onset (WASO), increased time spent in stage 1 and stage 2 sleep, and reduced durations of slowwave sleep (SWS) and rapid eye movement (REM) sleep (Floyd et al., 2000). There can also be changes to nonrapid eye movement (NREM) sleep. Phase advances in the timing of circadian rhythms

23

occur, which in turn affect many behaviors coordinated by the circadian system including sleep–wake timing, body temperature, and mood. Changes in the amplitude of a number of circadian rhythms are also common with increasing age. It has been hypothesized that the reduction in the amplitude of body temperature commonly reported in older adults may be causally related to the increased sleep disturbance (particularly, in the second half of the sleep period) that is commonly reported in this age group (Floyd et al., 2000). A primary marker of the phase of the circadian system is the pineal hormone melatonin (Wright and Rogers, 2008). Secretion of this indolamine is directly controlled by the circadian system, and is not directly influenced by sleep–wake state. High levels of melatonin are typically secreted at night, with its secretion suppressed by light exposure. Nocturnal onset of melatonin secretion is temporally related to a decrease in core body temperature, reduction in alertness and cognitive function, and increased sleep propensity (Wright and Rogers, 2008). Circulating melatonin levels are typically highest during childhood and adolescence, with gradual reductions occurring at the end of pubertal development and during the early twenties (Kennaway et al., 1999; Rogers et al., 2003a). Melatonin levels are believed to be significantly lower in older adults (Van Coevorden et al., 1991) with a number of cross-sectional studies documenting an age-related decline in melatonin production (Kennaway et al., 1999; Luboshitzky et al., 2001; Touitou, 2001). However, age-related decline has not been universally reported (Kawinska et al., 2005; Zeitzer et al., 1999). In addition to light exposure, a number of other factors including menopause, medical conditions, some medications (e.g., b-blockers), exercise, and diet (e.g., foods containing high levels of tryptophan or melatonin) are known to influence melatonin production and thus may potentially contribute to sleep–wake disturbance in older adults. Currently, there is a dearth of data that has investigated the dynamic interplay between sleep–wake changes that occur with aging and

age-associated changes in cognitive functioning. Using actigraphy in a sample with age-associated cognitive decline, one study did report a relationship between rest–activity fragmentation and processing speed, memory, and executive functioning (Oosterman et al., 2009). Another study, by Nebes et al. (2009), found that when a sample of healthy older adults was classified as either good or poor sleepers, good sleepers performed better on tests of working memory, attentional set-shifting, and abstract problem-solving as compared to the poor sleepers. Importantly, this relationship was not explained by confound factors such as cerebrovascular disease, depression, or medication usage, suggesting that sleep problems may contribute to cognitive performance variability in older adults. Only a few studies of memory consolidation have been conducted in older adults; these are discussed with respect to the mechanisms of memory consolidation (see section Sleep and Cognition). Sleep–wake changes in neurodegenerative disease In addition to the alterations associated with normal aging, sleep–wake disturbances are pronounced in subsets of individuals with neurodegenerative conditions. While these have been most extensively investigated in AD, a burgeoning array of recent literature has described sleep–wake disturbances in PD and LBD and a smaller number of studies have begun to characterize changes in FTD. While not specifically addressed here, some investigators have also begun to characterize sleep–wake changes in other neurodegenerative diseases including progressive supranuclear palsy, multiple system atrophy, Huntington’s disease, corticobasal degeneration, and Creutzfeldt–Jakob disease (see review by Gagnon et al., 2008). Despite its high prevalence, no known studies have characterized sleep–wake disturbances in vascular dementia. A summary of findings of the sleep–wake changes in the major neurodegenerative diseases, and their relationship with clinical and cognitive features is presented in Tables 1 and 2.

Table 1. Summary of neurobiological and sleep–wake changes in the major neurodegenerative diseases

Disease

Neuropathology

Cognitive profile

Common sleep–wake and circadian changes

Alzheimer’s disease (AD)

b-Amyloid neuritic plaques and hyperphosporylated tau neurofibrillary tangles. Early changes seen in hippocampal and entorhinal cortex; progressing to involve limbic area and neocortex





 

Early changes in episodic memory reflecting predominant medial temporal lobe and hippocampal pathology Later changes in language and visuospatial functions Decline in executive functions, behaviour, praxis, and remote memory occur later

  



Parkinson’s disease (PD)

Severe loss of dopaminergic cells within the nigrostriatal tract. Serotonergic, cholinergic, and noradrenergic depletion occurs in early disease and diffuse spread of synuclein pathology with disease progression



 

 

Lewy body dementia (LBD)

Synuclein containing Lewy bodies in the brainstem and cortex



 

Early decline in processing speed and executive functioning including working memory, problem-solving, and set-shifting Later decline in visuospatial functions and memory Neuropsychological changes evident in around 50% of patients Dementia prevalence is around 15–20% Up to 80% may progress to dementia in a 20-year period



Early changes in attention, executive function and visuospatial skills Fluctuating cognition Accompanied by visual hallucinations, parkinsonian features and falls



   

Relationship between sleep–wake changes and cognitive and clinical features

Sundowning and alterations in circadian phase and/or amplitude Sleep-disordered breathing Hypersomnia Amplification of age-related changes in sleep architecture Occur in up to 40% of patients in early stages



Insomnia Sleep-disordered breathing REM sleep behaviour disorder Hypersomnia Occur in up to 75% of patients









  

  

REM sleep behavior disorder Insomnia Hypersomnia Occur in up to 40% of patients



 

Predictive of ongoing cognitive decline Limited evidence shows relationships with general cognitive function, verbal fluency, working memory, memory, and executive functioning May be underpinned by cholinergic changes in the basal forebrain, alterations in melatonin amplitude and timing due to disruption of the circadian system Both REM sleep behavior disorder and hypersomnia are prodromal features of PD and risk factors for dementia Few detailed studies to date linking sleep–wake disturbance with neuropsychological functions May be exacerbated by dopaminergic medications May relate to depressive symptoms May worsen with disease severity or duration No detailed studies to date linking sleep–wake disturbance with neuropsychological functions Sleep–wake disturbance more pronounced than in AD REM sleep behaviour disorder is a prodromal feature and is predictive of cognitive decline

Frontotemporal dementia (FTD)

Deposition of transactivating responsive sequence DNA-binding protein (TDP43) and Tau proteinopathy





Mild cognitive impairment (MCI)

Amnestic subtypes may have early medial temporal lobe changes and AD pathology. Nonamnestic subtypes have more diffuse etiological underpinnings including cerebrovascular disease, psychiatric disorders, medical conditions or other causes of dementia

  



  

Late-life depression (LLD)

May include a combination of cerebrovascular and neurodegenerative disease, and illness specific factors affecting inflammatory, immunological, and hypothalamic-adrenal axis functioning

 



 

Early changes in personality, social cognition, and executive functioning in behavioral variant Early changes in language in the language variants (Progressive Non-Fluent Aphasia and Semantic Dementia) primary progressive aphasia and semantic dementia



General preservation of functioning Nondemented Amnestic subtypes have early memory deficits detectable in neuropsychological tests Nonamnestic subtypes have deficits in other domains of neuropsychological functions Can have single or multiple domains affected Often accompanied by subjective memory impairment High risk of conversion to dementia



Frontosubcortical profile Changes in processing speed and executive functioning are most pronounced Mild change in memory, particularly new learning and retrieval of information Deficits often persist, despite depressive symptom resolution Neuropsychological deficits predictive of progression to dementia



 

 

 

Disturbed sleep–wake timing Occur in up to 30% of patients Limited data available



Clinician or carer rated data suggest sleep–wake disturbance in 14–59% of patients Hypersomnia Limited data available



Alterations in sleep architecture and sleep quality Disturbed sleep–wake timing Alterations in circadian phase and amplitude



 

 



 



No studies to date linking sleep–wake disturbance with neuropsychological functions Scarcity of PSG data No known differences between variants of FTD

Nocturnal arousals linked to nonverbal memory deficits, attention, and executive functions Sleep propensity linked to neuropsychological performance Neuropathological data suggests reduced cerebrospinal and pineal melatonin

Sleep-offset insomnia may relate to memory and language changes Nocturnal arousals linked to attention, memory, language, and executive function Sleep–wake changes may relate to later ages of illness onset May be a marker of treatment responsiveness, illness onset and recurrence Consider REM-suppressing effects of some antidepressants

26 Table 2. Summary of pharmacological and nonpharmacological treatments for sleep wake disturbance in neurodegenerative disease Pharmacological treatments Sedative hypnotics

  

Benzodiazepines

   

Psychotropics

 

Melatonin

    

Melatonin analogs

    

Acetylcholinesterase inhibitors

  

Less side effects than benzodiazepines May not be suitable for long-term use Little or no data on cognitive effects Not recommended for older people Not recommended for long-term use Cognitive side effects Clonazepam may be suitable for REM Sleep Behavior Disorder Antidepressants may suppress REM Antipsychotics may exacerbate sleep–wake disturbance in Alzheimer’s disease, increase cerebrovascular risk, and are associated with more severe cognitive decline and increased mortality Studied most extensively in Alzheimer’s disease with inconclusive findings Some data suggesting improved subjective sleepiness in Parkinson’s disease Some data suggesting improved sleep, cognition and mood in mild cognitive impairment Further controlled trials incorporating cognition are required May have neuroprotective effects Those also targeting serotonin may be beneficial due to effects on mood Minimal side effects Data in younger samples suggests beneficial effects on sleep continuity and slow-wave sleep May have neuroprotective and neuroplasticity promoting effects Further data required in older people and specifically for those with neurodegenerative disease Benefits to sleep architecture possibly via remediation of cholinergic systems critical to REM sleep May have beneficial effects on memory in healthy older people and Alzheimer’s disease Further data regarding effects on cognition in other neurodegenerative diseases is required

Nonpharmacological treatments     

Treatment options include bright-light therapy, education aimed at increasing physical activity, more structured bedtimes and wake times, minimizing noise, decreasing awakenings and use of regulated and controlled napping Can be used in conjunction with pharmacological treatments May be more suitable for long-term sustainability Multimodal approaches may be optimal Little or no data on cognitive effects

Alzheimer’s disease AD is the world’s leading cause of dementia and is characterized by a progressive decline in memory, language, visuospatial functioning, and executive functioning. The cognitive decline typically follows the progression of the disease from the hippocampal and entorhinal cortex to the limbic

area and neocortex. Changes in memory are often prominent early, have a temporal gradient (changes predominantly affect the acquisition of new or recent memories), and remote memory is eventually affected (Dorrego et al., 1999). Given the wide range of neural circuitry and neurotransmitter systems infiltrated by the pathological changes of AD, it is not surprising that

27

sleep–wake disturbance is a prominent clinical feature of the disorder. Indeed, clinical and community prevalence studies suggest that around 40% of patients with AD have sleep disturbance. Behaviorally, this is characterized by frequent daytime napping, increased nighttime wakefulness, and “sundowning,” periods of agitation and other disruptive behaviors that begin in the evening or at night (Bliwise et al., 1995). In AD, sleep disruption has a significant effect on daily functioning, and even in mild to moderate AD this feature has been associated with aggression including agitation, verbal outbursts, and physical threats (Moran et al., 2005). Not surprisingly, the effects of sleep disruption on daytime functioning are associated with broader psychosocial functioning, including disability and decreased quality of life. In turn, this contributes significantly to caregiver burden and often increases the need for placement in aged care facilities (Van Someren et al., 2000).

Circadian disturbance Changes in the circadian system are most evident from the phenomenon of sundowning, a feature that affects over half of AD patients living at home (Gallagher-Thompson et al., 1992) and represents the leading cause of nursing home placement (Pollak and Perlick, 1991). While further prospective studies are required in order to ascertain the prognostic significance of circadian disturbance in early AD, data from moderate to severe AD samples suggest that this feature is predictive of more pronounced cognitive decline (see review by Wu and Swaab, 2007). Even early in the disease, researchers have linked sleep–wake disturbances with cognitive dysfunction including general cognition, working memory, verbal fluency, memory, and executive functioning (Bonanni et al., 2005). Studies using actigraphy in home-dwelling patients have shown that the deterioration of rest–activity cycles occurs early but most notably with observable progression of the disease (Hatfield et al., 2004). In addition to the behavioral manifestation

of sundowning, there are other biomarkers suggestive of circadian rhythm dysfunction, including changes in the amplitude and/or timing of melatonin and cortisol secretion, and body temperature, which, in turn, have been linked to cognitive impairment (Giubilei et al., 2001; Harper et al., 2005; Hatfield et al., 2004, and see review by Wu and Swaab, 2007). Sleep architecture Changes in sleep architecture in AD tend to show a similar profile to those seen in normal aging, but are more pronounced. That is, duration of SWS tends to be reduced, and there is increased WASO and time spent in stage 1 sleep, resulting in less restorative sleep episodes. A decreased percentage and duration of REM sleep episodes has also been described (Montplaisir et al., 1995), a finding which is not surprising given that REM sleep is in itself largely determined by cholinergic neurons (discussed below) with an onset that is driven by the circadian system (PaceSchott and Hobson, 2002). Other reported changes in sleep architecture in patients with AD relative to controls include decreased electroencephalogram (EEG) density during REM sleep periods as well as decreased density of K-complexes and sleep spindles during stage 2 sleep, which may impact on waking cognitive function and memory. The presence and relevance of reported changes in REM sleep have been inconsistent and questionable (Bliwise et al., 1989; Gagnon et al., 2008). Indeed, some data does not support differences in the number of REM episodes across the night in patients with AD relative to healthy controls (Gagnon et al., 2008). Sleep propensity is often increased during the daytime in AD, not only in advanced AD but also in milder cases (Bonanni et al., 2005). Hypersomnia A number of studies have highlighted the occurrence of hypersomnia in AD (Bonanni et al.,

28

2005; Lee et al., 2007). Significantly, this feature has a negative impact on daily function, and appears to be independent of cognitive function, suggesting that there may be differential neurobiological processes impacting on sleepiness and cognition (Lee et al., 2007). While the cause of hypersomnia in AD is unknown, it may be secondarily related to sleep fragmentation and poor sleep quality. In addition, a range of other factors such as sleep-disordered breathing (evident in around 35–63% of AD patients; Cooke et al., 2006), dysregulation of circadian timing (with increased sleep propensity no longer confined to nocturnal periods) or reductions in the amplitude of the circadian drive for wakefulness may all contribute to hypersomnia.

Neuropathology Pathologically, AD is characterized by the deposition in the brain of b-amyloid neuritic plaques and hyperphosphorylated tau neurofibrillary tangles. Early changes commonly target the temporal mesocortex, from where it spreads throughout other cortical regions in a fan-like manner, predominantly affecting central cholinergic pathways (Braak et al., 2006a). In regard to overlap with the sleep–wake systems, AD pathology affects the nucleus basalis of Meynert (a major cholinergic region), as well as the anterior (suprachiasmatic nucleus, SCN and ventrolateral preoptic nucleus, VLPO) and posterior hypothalamus, the intralaminar nucleus of the thalamus, the locus coeruleus, the raphe nuclei, and the central autonomic regulators (Stopa et al., 1999; Swaab et al., 1992) even in early stages (Simic et al., 2009). Neuropathological studies have found decreased melatonin in the pineal gland and suggest that prominent pineal changes likely underpin circadian change (see review by Wu and Swaab, 2007). The role of circadian “clock” genes for circadian disturbance in AD requires further examination but preliminary data suggest that there is a disrupted circadian profile of pineal

clock-gene expression during AD progression (Wu et al., 2006a). It has been postulated that melatonin may not only be associated with AD but may also play a key role in the pathophysiology. This view is largely supported by animal studies demonstrating that melatonin has potent antioxidant and neuroprotective properties and may inhibit amyloid and oxidative pathology. Animal studies have also shown that melatonin supplements may improve memory retention deficits (Zhu et al., 2004). Not surprisingly, multiple components of the sleep–wake system are affected in AD. There are likely complex interactions arising from the relationships between the SCN (the central circadian pacemaker), REM sleep (the timing of which is driven by the circadian system), and the cortical innervations of the cholinergic basal forebrain (required during REM sleep). Additionally, NREM requires disinhibition of VLPO neurons by adenosinergic inhibition of GABA-containing basal forebrain neurons and the subsequent inhibition of diencephalic and ascending brainstem arousal systems. Thus, any AD pathology affecting the VLPO and basal forebrain will have broader effects on NREM sleep, and the diffuse pathology is likely to also effect the homeostatic drivers for sleep (see review by Wulff et al., 2010).

Parkinson’s disease PD is a common neurodegenerative condition affecting at least 1% of those over the age of 65 years (de Rijk et al., 1995). Although characterized by its cardinal motor signs of bradykinesia, cogwheel rigidity, resting tremor, and postural instability, much of the burden of disease is the result of a range of nonmotor features occurring across cognitive, psychiatric, behavioral, and autonomic domains. Cognitive deficits in PD early in the disease course often reflect frontostriatal dysfunction, with difficulties in processing speed, concentration, and executive dysfunction. However, as the disease progresses,

29

visuospatial and memory impairments additionally emerge. This decline appears to reflect the dissemination of disease through temporal and parietal domains in keeping with the known increase in limbic and neocortical Lewy body pathology (Braak et al., 2006c; Halliday, 2008). Dementia is prevalent in 15–20% of PD patients and has been reported to affect 80% of those with a 20-year disease duration (Hely et al., 2008). Sleep–wake disturbance is an extremely common feature of PD, affecting around two-thirds of patients (Garcia-Borreguero et al., 2003). Disturbances of sleep in PD are associated with cognitive decline, psychiatric symptoms, caregiver burden, and reduced quality of life. These disturbances also represent a significant risk factor for institutional care (Garcia-Borreguero et al., 2003; Marion et al., 2008; Verbaan et al., 2008). There is considerable diversity in the pattern of sleep–wake disturbances seen in PD with manifestations including increased prevalence of sleep-disordered breathing, insomnia, hypersomnia, and REM sleep behavior disorder (RSBD; Gagnon et al., 2008; Garcia-Borreguero et al., 2003; Rogers and Kloss, 2005). Not surprisingly, the etiology underpinning sleep–wake disturbance in PD is likely to be multifactorial. Other factors including periodic limb movements, medication effects (e.g., dopamine agonists, the “wearing off” phenomenon), pain, discomfort, and nocturia may all be compounding the effects of neurodegenerative changes on the sleep–wake brain systems. Despite the fact that there have been an abundance of studies investigating the nature of sleep–wake disturbance in PD, surprisingly, few studies have examined in detail the relationship of these features with performance on neuropsychological test batteries. Further, most studies have utilized questionnaire measures to derive a measure of sleep disturbance, and do not generally include the “gold standard” polysomnography (PSG) assessments, nor even measures such as actigraphy or detailed circadian assessments.

Hypersomnia Hypersomnia is reported in over 50% of patients with PD (Rye et al., 2000) and is associated with age, disease severity, and dopamine agonist medication (Verbaan et al., 2008). Significantly, hypersomnia may actually predate the motor symptoms of the disease by several years (Abbott et al., 2005) and is a known risk factor for the development of dementia (Ondo et al., 2001). In PD, hypersomnia can also manifest as an acute sleep attack, which is of concern if the patient is performing cognitive tasks, operating machinery, or driving (Hobson et al., 2002).

Insomnia Up to 60% of PD patients report insomnia, which is often characterized by increased sleep latency, early awakening, and sleep fragmentation (nocturnal intermittent wakefulness). The causes of insomnia in PD are varied and may relate to physical discomfort, nocturia, or restless leg movements and tend to be related to depressive symptoms and disease duration (see reviews by Gagnon et al., 2008; Gunn et al., 2010).

REM sleep behavior disorder In PD, RSBD may be evident in up to 60% of patients and is commonly manifested by dream enactment behaviors such as talking, shouting, swearing, punching, kicking, crying, and laughing (Iranzo et al., 2009). During RSBD, there is a loss of muscle atonia that usually accompanies REM sleep. Brainstem regions responsible for the control of REM-associated muscle atonia are particularly targeted in neurological disorders with synuclein pathology, such as PD, multiple system atrophy and LBD. Of significance, RSBD may occur as a prodromal feature in PD predating motor symptoms by several years, making it a potential future

30

diagnostic biomarker. The occurrence of RSBD also appears to have particularly strong associations with cognitive and psychiatric functioning and has been shown to be predictive of dementia in longitudinal studies (Marion et al., 2008; Vendette et al., 2007). Those with RSBD may be more likely to have the akinetic-rigid or nontremor subtype of PD and, in nondemented samples, these patients may have poorer memory and poorer visuospatial and executive functioning on neuropsychological testing (Kumru et al., 2007; Vendette et al., 2007). While RSBD is diagnosed using PSG, recent studies have shown that actigraphy may also be a viable tool for recognizing this symptom early (Naismith et al., 2010b). The finding that only one-third of patients have awareness of RSBD symptoms suggests that not only are screening tools required, but there is a need to ask bed partners and/or caregivers about this particularly disruptive symptom (Iranzo et al., 2009). Clinical history and bed partner observations in those with PD or known neurodegenerative conditions may yield reasonable sensitivity and specificity (see Iranzo et al., 2009 for discussion). There is, however, evidence that the prevalence of PSGconfirmed RSBD may actually be higher than those reported on clinical history alone with rates ranging from 46% to 58% in comparison to 15% to 46% when based on clinical history alone (see Iranzo et al., 2009 for review).

from that related to motor symptomatology. The same study also reported that the occurrence of RSBD was not related to disease duration, which is in keeping with the finding that this phenomenon may predate the motor symptoms of disease by several years (Postuma et al., 2009).

Circadian disturbance To date, there is not strong evidence for circadian disruption in PD. Previous studies conducted in small groups of drug naïve patients have demonstrated a circadian profile similar to that of agematched controls (Bordet et al., 2003; Fertl et al., 1991). Additionally, although initial sleep-onset insomnia is common in PD, it does not increase significantly with disease progression (Alves et al., 2005; Maetzler et al., 2009), suggesting a lesser role for circadian disruption than that seen in AD. In interpreting the findings investigating the circadian system, it is important to note that some studies have demonstrated significant circadian changes characterized by a phase advance of the nocturnal melatonin peak (Bordet et al., 2003; Fertl et al., 1993). However, these studies often include cases with advanced PD, who are receiving chronic high-dose L-dopa treatment. Thus, circadian disturbance in PD may well be related to dopaminergic medications and their indirect modulation of the circadian system, rather than to the underlying neurodegenerative process per se.

Sleep architecture Neuropathology Data exploring sleep architecture in PD are limited and inconsistent. Only one known prospective study has been conducted in PD patients, which showed that with increasing disease duration there is a progressive reduction in sleep efficiency, total sleep time, SWS, and REM sleep (Diederich et al., 2005). These changes were not correlated with motor symptoms or medications, suggesting that these features of sleep may represent a distinct underlying pathology that differs

The major pathological finding in PD relates to a severe loss of dopaminergic cell loss within the nigrostriatal tract. In addition, there is serotonergic (dorsal raphe), cholinergic (nucleus dorsalis), and noradrenergic (locus coeruleus) depletion occurring in early disease and diffuse spread of Lewy body pathology throughout the regions of the cortex in the more advanced stages (Braak et al., 2006b). This disseminated and progressive

31

pattern of disease evolution from regions within the medulla (dorsal motor nucleus of the vagus nerve) throughout the brainstem and cortex is likely responsible for a host of the nonmotor features observed in PD, and it is not surprising that there are disturbances to widespread components of the sleep–wake system. Although the SCN is relatively devoid of pathology in PD (Langston and Forno, 1978), a number of hypothalamic regions responsible for modulating sleep–wake patterns do exhibit disturbances. Recent neuroimaging techniques have identified dopaminergic dysfunction in the hypothalamus (Politis et al., 2008), and clinicopathological studies have shown a dramatic loss of hypocretin (orexin) cells within the hypothalamus (Fronczek et al., 2007; Thannickal et al., 2007). Investigators have previously highlighted the similarities in orexinergic cell loss in narcolepsy, a condition characterized by abrupt transitions between wake and sleep, and the hypersomnia that occurs in PD (Fronczek et al., 2007; Thannickal et al., 2007), raising the possibility of a therapeutic target for this symptom based on progress in narcolepsy symptom treatment. However, cerebrospinal fluid measurements conducted in advanced PD patients with hypersomnia have not consistently shown reduced levels of the neuropeptide orexin A (Overeem et al., 2002) and have also failed to show correlation with disturbances in sleep architecture (Compta et al., 2009).

Lewy body dementia LBD is one of the most common forms of dementia after AD. A review of prevalence studies in LBD studies noted rates of up to 5% of the general population (Zaccai et al., 2005). It is characterized by visual hallucinations, Parkinsonism, and fluctuations in cognition and alertness. While it shares common pathology (synucleinopathy) and clinical features with PD, dementia is a very early feature in LBD occurring within 12 months of symptom onset. Cognitive impairment in LBD also has a

different profile to PD and is characterized by early and prominent visuospatial, attentional, and executive dysfunction (see review by Troster, 2008). The occurrence of synuclein containing Lewy bodies in the brainstem and cortex is a necessary pathological feature of LBD (see Iranzo et al., 2009 for a review). In many cases, patients with LBD have the associated presence of cortical b-amyloid changes, although even when this AD associated pathology is prominent, memory performance is superior to those with AD alone, whereas visuospatial impairment is more pronounced (see review by Troster, 2008). This has been purported to reflect a lower burden of neurofibrillary tangles and less cholinergic compromise in the medial temporal regions in those with LBD (see review by Troster, 2008). As with PD, patients with LBD also experience prominent sleep disturbance. Due to the underlying pathology being synucleinopathy, it is not surprising that RSBD is also a prodromal feature of LBD. There are few data available on prevalence, but estimates suggest that around 40% of LBD patients have this feature, which may be more common in men (Rongve et al., 2010). The presence of RSBD may also be an important feature assisting with the differential diagnosis of LBD and AD. A recent questionnaire-based study of sleep disturbances in early dementia concluded that sleep disturbances were more commonly found in LBD, in comparison to AD (Rongve et al., 2010). While there is a paucity of polysomnographic data available in LBD, disturbances of REM sleep have been postulated to be due to cholinergic neuron loss (Grace et al., 2000).

Frontotemporal dementia FTD is thought to be the second most common form of dementia under the age of 65 years (Ratnavalli et al., 2002). This disease includes a spectrum of dementias characterized by focal atrophy of frontal and anterior temporal regions with pathological changes that are distinct from AD and PD. The pathological changes observed

32

in FTD are varied including the deposition of transactivating responsive sequence DNA-binding protein (TDP43) and tau proteinopathy and may be related to the presence of specific genetic defects. The distribution and nature of this pathology is believed to contribute to the range of clinical features observed. Patients with FTD may experience problems with social behavior, including disinhibited behavior (behavioral variant) or language impairment (language variants; progressive nonfluent aphasia and semantic dementia). Sleep disturbances in FTD have been reported in 30% of patients (Fernandez Martinez et al., 2008) and have demonstrated the existence of specific circadian rhythm disturbances that could be differentiated from healthy controls and AD patients (Anderson et al., 2008). A prior study conducted in 30 men with AD and FTD (Harper et al., 2001) found differential circadian changes in both groups, with AD patients showing a phase delay in activity and body temperature rhythms, and FTD patients showing a fragmented activity rhythm and phase advance in activity relative to core temperature—suggestive of internal desynchrony. However, both of these studies were restricted to prolonged actigraphy recordings and sleep diaries. Neither included detailed clinical nor neuropsychological assessments and neither included assessment of melatonin nor objective assessment of sleep (i.e., PSG). In one preliminary study that included PSG recordings of five FTD patients, data were reported to qualitatively suggest reduced SWS as well as decreased REM duration and total sleep time, anomalies that tended to increase with disease duration (data cited in Autret et al., 2001).

Vascular dementia While there are no known studies examining the relationship between sleep–wake disturbance and cognitive impairment due to cerebrovascular disease, there are a number of stroke studies that highlight the role of key brain regions for sleep.

Dysfunction in the ascending arousal system can be incurred by hemorrhagic or ischemic lesions at the level of the reticular formation and brainstem generally, which has widespread noradrenergic, cholinergic, and histaminergic cortical connections (with projections originating from the locus coeruleus, reticular substance, and posterior hypothalamus, respectively). Additionally, thalamic lesions have been associated with insomnia and hypersomnia while pontine lesions have been associated with changes to both REM and NREM sleep. No known studies have attempted to characterize the location and burden severity of white matter lesions in relation to sleep–wake disturbance (see Autret et al., 2001). One recent study by Berlow et al. (2009), however, showed that in AD, white matter brain lesions were associated with greater symptoms of nighttime disturbance. This suggests that disturbance of brain circuitry due to cerebrovascular disease contributes to sleep–wake disturbance. Further studies in this area would not only help to delineate the relationships between sleep and cognition in association with cerebrovascular disease, but would also assist with the development of neurobiological models of sleep–wake disruption.

Mild cognitive impairment MCI is a clinical syndrome whereby individuals have cognitive decline, beyond what would be expected for their age, yet do not meet criteria for dementia (Petersen and Morris, 2005). On formal neuropsychological tests, commonly used criteria require subjective cognitive complaints and a deficit of at least 1.5 standard deviations in at least one domain of functioning. Those with prominent memory impairments (amnestic subtype) are the best characterized and this group has an especially high risk of progression to AD, with studies showing that over 50% progress to dementia within 5 years. Hence, they could be seen to represent a prodromal stage of the disorder. A nonamnestic subtype of MCI has also been

33

described although the pathophysiology underlying this subtype is likely to be more diffuse and there is less data available on longitudinal progression (Gauthier et al., 2006). While there are some known clinical markers of deterioration in MCI, such as subjective memory complaints and depression (Gauthier et al., 2006), the nature and predictive capacity of sleep and/ or circadian changes as “biomarkers” for conversion to dementia have yet to be established. The lack of studies characterizing the relationship between cognitive and sleep–wake changes in this at-risk group is indeed surprising given the prominent sleep and circadian changes observed in AD. Using caregiver or clinician reports alone, a recent review (Beaulieu-Bonneau and Hudon, 2009) noted that over 15 studies have examined sleep disturbances in MCI. Overall, between 14% and 59% of patients with MCI were noted to have sleep disturbances. In nonamnestic MCI, actigraphy data suggest that the number of nocturnal “arousals” (wake bouts) is associated with poorer learning and problem-solving, while a greater duration of WASO is associated with reduced attention and executive functions. These associations appear to be independent of potential confounds such as age and depressive symptoms (Naismith et al., 2010a). In general, there is a dearth of data in MCI that explores the associations between early cognitive and sleep changes, although some data suggest that sleep–wake changes are evident early in the disease. For instance, in one study examining circadian changes in both MCI and AD, Bonanni et al. (2005) reported that people with early AD had hypersomnia and there was a negative correlation between sleep propensity and neuropsychological test scores, implying that hypersomnia in the early stages of the disease might exacerbate cognitive impairment in AD patients. Neuropathological studies have demonstrated reduced CSF and pineal levels of melatonin (Wu et al., 2003) in those that were deemed to demonstrate early AD pathology (Braak stages I–II; Braak et al.,

2006a). Overall, this preliminary body of work seems to suggest that sleep–wake disturbance in MCI may relate to early cognitive change. However, research is required to further characterize the associations with other clinical, medical, neuroimaging (e.g., white matter change or medial temporal lobe atrophy), and psychiatric features. It will also be important to longitudinally track cohorts to determine the predictive capacity of sleep–wake change as a biomarker for conversion to dementia and disease progression. Importantly, further emphasis on delineating these relationships in prodromal AD states may assist with targeted sleep–wake interventions, thereby potentially delaying cognitive decline and improving psychosocial functioning (Naismith et al., 2009a).

Late-life depression The prevalence of depressive disorders affecting older adults (late-life depression) ranges from 9% to 26% in the elderly population (Copeland et al., 2004). This disorder is commonly associated with high rates of medical morbidity, mortality, cognitive impairment, disability, high healthcare costs, and poor prognosis (Alexopoulos et al., 1993; Hickie et al., 1995; Murphy et al., 1998). Depression is an independent risk factor for dementia, and those with depression and cognitive impairment are twice as likely to develop AD. The past two decades have witnessed an abundance of research that has sought to understand the neurobiology of late-life depression with data indicating that disruptions to frontosubcortical circuitry most likely underpin symptomatology. Additionally, it appears that depression emerging later in life (i.e., onset after the age of 50; late-onset depression) may represent a subtype of depressive disorder that is etiologically distinct. This subtype has particularly pronounced cognitive impairment, neurobiological changes, vascular risk factors, white matter changes on neuroimaging, and increased rates of progression to dementia (Hickie et al., 1997).

34

The neuropsychological profile of late-life depression resembles that of frontosubcortical dysfunction, with prominent changes in executive dysfunction, processing speed, and to a lesser extent learning and memory (Naismith et al., 2003). Changes in cognition also appear to be linked to underlying volumetric changes in regions such as the hippocampus and caudate nucleus (Hickie et al., 2005; Naismith et al., 2002). Despite the detailed neurobiological characterization of this group, research in this area has generally neglected to consider the interrelationships of clinical, neuroimaging, and neuropsychological features with sleep–wake disturbance. This is surprising, as both insomnia and hypersomnia are part of the DSM-IV diagnostic criteria for major depression. Sleep–wake disturbance may also be a prodromal feature of depressive symptom onset that may persist despite symptom remission and may perpetuate the illness (Pigeon et al., 2008). In late-life depression, insomnia is the most commonly reported sleep disturbance; this includes complaints of difficulty in falling asleep, frequent nocturnal awakenings, and early morning wakefulness (Benca and Peterson, 2008; Buysse, 2004). Sleep disturbances in depression are indicative of a more severe illness and tend to be under-treated (Sunderajan et al., 2010). Epidemiological studies suggest that insomnia is a significant risk factor for depression onset and recurrence in younger and older samples and can even be observed over follow-up periods of up to 12 years (Cho et al., 2008; Dombrovski et al., 2008; Mallon et al., 2000; Perlis et al., 2006). In late-life depression, sleep disturbance appears to be most prominent at the onset of the episode (Dew et al., 1996) and has been shown to be a prognostic marker. For instance, Dew et al. (1996) found that EEG sleep abnormalities predicted a slower response to antidepressant treatment. While few detailed studies have attempted to characterize sleep–wake disturbance according to age of onset, one polysomnographic study failed to show differences between age subgroups (Buysse et al., 1988).

Although only a small number of studies have concurrently considered the interrelationships between depression, sleep–wake disturbance, and cognition, it is possible that such features reflect disruption to common neurobiological circuits. This hypothesis is supported by data showing that depression and anxiety are the most common psychiatric features to be associated with sleep disturbances in early dementia (Rongve et al., 2010). More recently, researchers have begun to investigate the relationship between sleep and cognitive dysfunction in late-life depression. In one study that examined clinician-rated sleep changes in older patients with moderate to severe major depression, late insomnia was associated with more pronounced cognitive impairment on tests of memory and verbal fluency as well as later ages of depression onset (Naismith et al., 2009b). Studies utilizing actigraphy as a more objective measure of sleep–wake behavior have further suggested that the extent of nocturnal awakenings is associated with the severity of cognitive impairment across a number of neuropsychological measures of memory and executive functioning (Naismith et al., 2011). The relationships between sleep–wake disturbance and neuropsychological functioning remained significant after controlling for depression severity, suggesting that such changes are likely to reflect common underlying neurobiological changes. One recent study conducted by Dresler et al. (2010) showed that overnight consolidation of procedural memories was impaired in patients with depression of all ages, and was particularly poor for older patients. The finding that those in remission performed at a level comparable to control subjects, however, suggests that the “state” effects of depression might be a significant additional factor influencing overnight memory consolidation. Further research delineating the links between early cognitive impairment, clinical features (e.g., treatment resistance, cerebrovascular disease), and sleep disturbance is now needed for this group, which is at-risk of developing

35

dementia. This is warranted not only to elucidate whether sleep and/or circadian changes are biomarkers for cognitive decline but also to help inform early intervention approaches (Naismith et al., 2009a).

Sleep architecture When examining depression across age groups, PSG studies have demonstrated a number of alterations in sleep architecture. These are characterized by reduced REM latency, prolongation of the first REM period, an increased proportion of REM sleep, increased REM density, and a reduced time spent in SWS (see Benca and Peterson, 2008). There is a dearth of data specifically addressing sleep disturbance through objective methods in late-life depression. One study evaluated the PSG data of 41 participants with major depressive disorder before they were admitted for electroconvulsive therapy (Grunhaus et al., 1996). The authors showed that older patients (older than 65 years) with late-onset depression (mean age of depression onset of 54.3 years) had reduced total sleep time, impaired sleep maintenance, and lower REM sleep pressure and density compared with patients younger than 65 years. As noted above, there do not, however, appear to be differences between early- and late-onset depression subtypes with regard to sleep architecture (Buysse et al., 1988).

Circadian disturbance Data suggest that there are significant alterations in the amplitude and rhythm of melatonin in those with depressive disorders (Srinivasan et al., 2006). However, the literature is inconsistent in terms of the direction of the alterations (Lewy, 2002). Several studies have reported reduced 24-h secretion of melatonin, with irregular daytime peaks and a reduced, phase-advanced nighttime peak in individuals with major

depression (Branchey et al., 1982; Claustrat, 1984; Parry and Newton, 2001; Wehr et al., 1985). In contrast, a number of studies have shown delays in circadian phase with an increased peak in nocturnal melatonin concentration, while further studies show no alterations at all (Carvalho et al., 2006; Rubin, 1992; Sekula et al., 1997). The differences in these findings may be attributed to changes in depressive symptomatology, and/or disparities in antidepressant medication use (Pacchierotti, 2001), and/or failure to consider the heterogeneity of depressive disorders, particularly, with regard to depressive subtype and other etiological factors (Naismith et al., 2003). The reported changes in rhythm and amplitude of melatonin, however, are consistent with alterations in the timing of the sleep–wake cycle and sleep physiology changes reported earlier (Buysse, 2004; Lauer et al., 1991). For example, early morning wakefulness (Buysse, 2004; Nelson et al., 2005) and shortened REM latency (Lauer et al., 1991) reported in older adults with major depression are consistent with the observations of circadian phase advance.

Medication effects Some antidepressant medications including amitriptyline, citalopram, duloxetine, and venlafaxine suppress REM sleep, while others such as buproprion, mirtazapine, or trimipramine do not appear to suppress or enhance REM sleep (see overview by Dresler et al., 2010). Studies to date have not shown broader effects of antidepressants on declarative memory (see reviews by Smith, 2001; Walker et al., 2005). However, recent data have shown that overnight consolidation of procedural memories is not hampered by REMsuppressing medications in patients with major depression, a finding that is contrary to the purported role of REM sleep in procedural memory (Rasch et al., 2009). While no sleep EEG data are available to confirm this finding, it tentatively suggests that NREM sleep may be more pertinent

36

to this cognitive function in major depression (see below for the role of REM and NREM sleep in cognition). Additional studies incorporating PSG are needed to further explore the potential benefits or limitations of antidepressant medications for cognition in depression. Indeed, such studies should be conducted in light of comprehensive pathophysiological models of depression including the role of hypothalamic–pituitary–adrenal axis functioning, hippocampal volume reduction, and depressive subtype (Dresler et al., 2010; Hickie et al., 2005; Naismith et al., 2003). Sleep and cognition Recent years have witnessed a resurgence in the animal and human literature on the critical role of sleep for neuropsychological functions. While numerous studies have documented the correlates of various aspects of sleep disturbance including circadian disruption (Rogers and Dinges, 2008), sleep deprivation (Dinges, 2005; Morris et al., 1960; Rogers et al., 2003b), and sleep-disordered breathing (e.g., due to both hypoxemia and sleep fragmentation; Naismith et al., 2004; Wong et al., 2008), there has also been significant interest in the role of overnight sleep processes for critical brain functions such as memory consolidation and neurogenesis (see reviews by Brankack et al., 2009; Walker, 2010). These data offer unique insights into brain functioning and offer promise for future therapies that may target sleep in order to facilitate neuroplasticity and improved neuropsychological functions.

The neural basis of learning and memory The neural networks underpinning declarative learning and memory primarily involve frontotemporal circuits for encoding and the hippocampus appears to be essential for memory storage. In AD, pathology primarily begins with in the hippocampus. Thus, memory storage

deficits are a prominent early sign, and indeed are evident in amnestic forms of MCI. Other dementias such as LBD and FTD tend to include more prominent cortical involvement and the learning and memory difficulties of PD are likely to involve frontosubcortical systems. Thus, learning and memory deficits in the various neurodegenerative diseases often exhibit distinct profiles depending on the distribution of pathology. Neuropsychologically, this can be conceptualized as having difficulties with encoding (i.e., frontotemporal or frontosubcortical circuitry), storage (i.e., hippocampal), and/or retrieval (i.e., frontotemporal or frontosubcortical circuitry). In addition, other aspects of nondeclarative memory, such as procedural memory and implicit memory, are likely to include a diffuse frontosubcortical network and in some cases even recruit the cerebellum (Lewis et al., 2003; Lezak, 1995; Naismith et al., 2010d).

The role of sleep in learning and memory It is likely that there are both direct and indirect effects of sleep that facilitate various aspects of learning and memory. First, it appears that different sleep stages directly and uniquely contribute to overnight memory consolidation. Although studies have primarily been conducted on younger healthy subjects, there are some inconsistencies (Rauchs et al., 2004). Empirical studies largely demonstrate that episodic declarative memory (the laying down of new semantic facts) is favored during SWS (Ferrara et al., 2008; Peigneux et al., 2004), whereas procedural learning (the reinforcement of acquired motor skills) occurs during REM sleep (Marshall and Born, 2007; Stickgold, 2005). However, such clear distinctions have not proved to be absolute and it has been suggested that other factors are relevant. For instance, the sequential stages of sleep may be fundamental to effective memory consolidation. As such, while SWS might consolidate the cortical integration of new memories by

37

reactivation of the neural networks involved in their initial encoding (Ji and Wilson, 2007; Wilson and McNaughton, 1994), REM sleep might be responsible for synaptic consolidation of these new memories (Giuditta et al., 1995). In addition, there appear to be “critical windows,” during which sleep periods must occur for optimal memory consolidation (for review, see Born et al., 2006). Of interest for elucidating the impact of sleep on memory consolidation in older people, there appear to be analogies with both aging and sleep deprivation, whereby with either phenomenon imaging studies have shown that the prefrontal cortex becomes increasingly involved in learning and memory, and there is less recruitment of posterior aspects of the hippocampus (Gutchess et al., 2005; Yoo et al., 2007). These data further suggest that with sleep deprivation, normal engagement of the medial temporal regions cannot occur, and the prefrontal cortex may exert a compensatory effect, with increased recruitment of neurons to maintain cognitive performance levels. Clearly, further studies are required in order to determine whether the neural network changes in memory systems observed with aging are associated with sleep changes, or whether the concomitant effects of both sleep disruption and aging have particularly deleterious effects. Second, sleep appears to be important for hippocampal neurogenesis, that is, new neuronal growth in the hippocampus. In this regard, adequate sleep is important for the overall integrity of neuronal structures and this process suggests one possible mechanism whereby sleep disruption may impede learning, memory, and neuroplasticity. Supporting data for this etiological mechanism largely stems from animal studies that highlight the important role of REM sleep for hippocampal neurogenesis. This observation is supported by studies showing that REM sleep loss is associated with a reduction in hippocampal proliferation, survival, and neurogenesis, whereas both NREM and REM sleep appear to modulate the number of cells that develop into mature

neurons. Thus it is possible that sleep is pertinent to memory not just for overnight consolidation but also for ensuring the integrity and optimal functioning of brain structures that support memory (Meerlo et al., 2009). Behavioral experiments that deprive participants of sleep prior to encoding tasks also lend support to this data by showing reduced ability to encode new information, which in turn is associated with reduced hippocampal activation posteriorly (Yoo et al., 2007). These data provide a further mechanism by which sleep disturbance can mediate or modulate cognitive and neuropsychiatric functioning in neurodegenerative diseases, and also provides a possible theoretical framework for the prognostic association between sleep and cognitive decline. While at a cellular level, the precise mechanism by which sleep disruption impedes neurogenesis and proliferation is unknown, glucocorticoids and other stress hormones may play some role (Meerlo et al., 2009). Overall, these findings suggest that early detection of sleep–wake disturbance in older people may offer opportunities for both primary and secondary prevention of cognitive decline (Naismith et al., 2009a). In terms of mechanisms, evidence from electrophysiological studies shows that during sleep, there are synchronized oscillatory patterns of neuronal activity that likely contribute to memory consolidation. These have been studied most extensively in NREM sleep where predominantly animal studies have shown that cortical sleep spindles are associated with excitatory activity initiated in the hippocampus. Thus, it is postulated that the co-occurrence of hippocampal sharp waves and cortical spindles plays a role in the integration of hippocampal and cortical information that is processed during sleep. This overnight neural function may thereby reallocate recently encoded memories from the hippocampus to the cortex and data from human studies support this observation (see review by Walker, 2010). These findings further posit that impairments in NREM sleep and spindle activity will compromise homeostatic functions that may,

38

in turn, reduce the capacity for memory encoding and processing the next day. That is, if there is insufficient opportunity to depotentiate saturated memory networks during NREM sleep, prefrontal and hippocampal encoding capacity will be suboptimal (see Walker, 2010). This hypothesis is consistent with data showing that sleeping prior to learning benefits memory encoding. Specifically, even mild disruptions to SWS (producing “light sleep”) can impede subsequent encodingrelated hippocampal activation, as evidenced by functional magnetic resonance imaging and memory performance (Van der Werf et al., 2009). Importantly, these data suggest that promotion of NREM sleep offers a target for interventions that may result in more effective encoding and consolidation of memories. Such interventions may be pharmacological or nonpharmacological (e.g., napping, exercise). Unfortunately, to date, there have only been a few studies conducted in older adults with neurodegenerative disease that have examined in detail the role of sleep stages for optimal memory encoding or consolidation. However, there is evidence that stage 2 sleep spindle density following procedural learning in older adults is not as pronounced as in younger samples (Peters et al., 2008). Data also suggest that there is a reduction in prefrontal sleep spindles with age (De Gennaro and Ferrara, 2003). Further research in this area may offer valuable opportunities to promote neuroplasticity in the aging brain, improve memory, and other aspects of cognition as well as addressing other associated features such as mood disturbance, quality of life, and caregiver burden. Treatments for sleep–wake disturbance in neurodegenerative diseases While a range of pharmacological and nonpharmacological options are available to address sleep–wake disturbance in older adults, data regarding effects on cognition are generally lacking. Further, there is a dearth of controlled

studies in those with neurodegenerative diseases, precluding conclusions regarding the efficacy of such interventions for cognition. To date, the most common treatment of sleep–wake disturbance in older people has been pharmacological, although this does not address the underlying mechanisms perpetuating the sleep–wake problems, and in this regard, emerging evidence is suggesting that nonpharmacological options are preferred, safer, and potentially more sustainable (Gehrman and Gooneratne, 2010).

Chronobiotic compounds The administration of melatonin (dose ranges 3–10 mg), a chronobiotic compound, has been most extensively investigated in AD, due to the prominent circadian changes observed in this group. These studies have produced some promising short-term (Cohen-Mansfield et al., 2000; Mishima et al., 2000) and long-term (Brusco et al., 2000) results for sundowning and irregular sleep cycles. Beneficial effects have also been found for reducing daytime somnolence, increasing daytime activity, and improving the ratio between daytime and nighttime sleep (Dowling et al., 2008). However, universal improvements have not been reported for the use of melatonin in AD (Singer et al., 2003) and recent metaanalytic reviews and large-scale, multisite randomized controlled trials have reported negative or inconclusive results generally (Brzezinski et al., 2005; Gehrman et al., 2009). Such negative findings may well be explained by the need to consider whether individual patients actually have any melatonin deficiency (Pandi-Perumal et al., 2005) or whether there is a significant reduction in melatonin (MT1) receptors in the SCN, thereby preventing effective melatonin action (Wu et al., 2006b). The use of melatonin in PD patients has been associated with subjective improvements in sleep but little change in objective measures such as actigraphy or PSG (Dowling et al., 2005; Medeiros et al., 2007). No

39

known studies have evaluated whether melatonin has beneficial effects on sleep in LBD, FTD, or vascular dementia. The use of melatonin early in neurodegenerative diseases may have benefits, with studies suggesting potential neuroprotective properties. There is increasing evidence regarding the role of melatonin in protection from oxidative damage (PandiPerumal et al., 2006). This protective role likely lies in its nonreceptor-mediated interactions as a potent free radical scavenger. Melatonin stimulates the expression and activity of antioxidative enzymes including glutathione peroxidase, glutathione reductase, and glucose 6-phosphate dehydrogenase (Pandi-Perumal et al., 2006). Melatonin is also capable of protecting against reductions in membrane fluidity, likely related to its free radical scavenging activity in that it inhibits the peroxidation of lipids (lipid peroxidation leads to increases in membrane rigidity; see Reiter, 1998). Of specific relevance for AD, melatonin appears to be protective against the toxicity of amyloid-b peptide (e.g., Bozner et al., 1997). For melatonin to be most efficacious as a neuroprotective agent, it may be best implemented at secondary levels of prevention, in those deemed to be atrisk of dementia, or specifically, in those with MCI. Further human studies are required in this area; however, in one small clinical sample of 10 elderly patients with MCI, melatonin was found to improve sleep, memory, and mood (JeanLouis et al., 1998). In a retrospective analysis of 50 MCI patients, only half of whom had received melatonin, those receiving 3–9 mg daily over periods of 9–18 months demonstrated superior cognitive performance on gross measures of cognition as well as improved mood and wakefulness and improved sleep quality (Furio et al., 2007). In late AD, there are preliminary data from case reports and small clinical studies suggesting that melatonin may improve cognition and memory, although further controlled studies are still required in this area (reviewed in Wu and Swaab, 2007).

Newer pharmacological agents selectively targeting melatonin receptors (Arendt and Rajaratnam, 2008) or melatonin and serotonin receptors have yet to be evaluated in neurodegenerative diseases although they may prove useful for concurrently targeting circadian and neuropsychiatric features in AD. One agent that may hold such promise is agomelatine, a melatonin receptor agonist with 5-hydroxytryptamine2C (serotonin) receptor antagonist properties. This novel antidepressant offers a unique pharmacological approach with minimal troubling side effects. Data suggest that it has beneficial effects on sleep with improvements in sleep continuity, an increase in SWS (without affecting REM sleep), and a normalization of the distribution of SWS throughout the night (Quera Salva et al., 2007). In comparison to sertraline, agomelatine demonstrates superior effects in terms of the relative amplitude of the circadian rest–activity cycle (as determined by actigraphy), as well as improved sleep latency and sleep efficiency (Kasper et al., 2010). No known studies have been published on the effects of agomelatine in older people with depression or neurodegenerative disease, and none have examined broader effects on cognition. This is unfortunate since animal studies suggest that this agent has neuroprotective and neuroplasticity promoting effects, including increases in cell proliferation and neurogenesis in the ventral dentate gyrus in the adult rat (Banasr et al., 2006) and protection from the excitotoxic effects of a glutamatergic analog (Gressens et al., 2008). Overall, however, further research utilizing melatonin and melatonin agonists is required in order to determine their efficacy, safety, and suitability for older at-risk patients as well as those with established neurodegenerative diseases. Insomnia treatments A range of pharmacological agents are used to treat insomnia in older people including sedative

40

hypnotics (“z-drugs”; e.g., zolpidem, eszopliclone, zaleplon), benzodiazepines (e.g., diazepam), antidepressants (e.g., amitriptyline, mirtazapine), antipsychotics (e.g., quetiapine, risperidone), antihistamines, and melatonin receptor agonists (e.g., ramelteon). However, despite their frequent use, few data are available regarding their longterm safety and efficacy (see review by Krystal, 2010), nor are detailed data available regarding effects on cognition. A relatively small number of studies have used PSG to examine effects on sleep architecture and there is a dearth of studies that included other outcomes, particularly mood, cognition, and functional status (see Krystal, 2010). Hypnotic medication is commonly prescribed for circadian disturbance in older adults as well as in AD, as the circadian disturbance is commonly misdiagnosed as insomnia. However, such medication usually loses its efficacy after  1 week of continuous use, and is thus unsuitable as a long-term insomnia treatment (Wu and Swaab, 2007). Further, hypnotic medication may increase the risk of falls and cognitive impairment and can impede daytime functioning in older people (Glass et al., 2005). Similarly, benzodiazepines are not advised for long-term use due to reduced efficacy with time and dependency. Administration of benzodiazepine compounds is associated with significant deficits in cognitive function immediately following administration (Rogers et al., 2003c) and with retrograde amnesia and hangover effects several hours after administration, and following waking (Uzun et al., 2010). In addition, several benzodiazepines have been reported to suppress REM sleep (Landolt and Gillin, 2000) and often lead to oversedation, thereby potentially further contributing to impaired daytime cognition. The abrupt withdrawal of benzodiazepines also needs to be carefully managed given its association with rebound insomnia. Benzodiazepines have also been associated with falls and fractures, confusional states, and nocturnal wandering. Antipsychotic medication is commonly used for those with more

severe cognitive decline but it has been noted to exacerbate sleep–wake disturbances in AD. Further, there are data to suggest the use of antipsychotic medications in this setting is associated with an increased cerebrovascular risk, more severe cognitive decline, and increased mortality (Wu and Swaab, 2007). As detailed below, combining short-term hypnotic use with nonpharmacological approaches may be optimal for longer-term management. Melatonin receptor agonists with affinity for the MT1 and MT2 receptors may also be used for insomnia in older people, and recently ramelteon has been approved in a number of countries for use in sleep-onset insomnia. However, further data on short-term and long-term efficacy are required in older patients generally, as well as in those with neurodegenerative disease.

Hypersomnia treatments Many of the nonpharmacological treatments described below have been shown to have beneficial effects on daytime sleepiness, although they were not specifically designed to address this aspect of functioning. Few pharmacological options specifically target hypersomnia, but stimulant medications may play some role. In a sample of patients with major depression who were only partially responsive to selective serotonin reuptake inhibitors, augmentation with the wake-promoting agent, modafinil, was shown to improve excessive sleepiness, fatigue, and depressive symptoms, and was well tolerated (Fava et al., 2007). However, all patients in this study were under the age of 66 years. Although its mechanism is unknown, modafinil is considered to be pharmacologically different to other stimulants, with effects on mood and cognition purported to be similar to those observed with caffeine. In support of this cognitive effect, a recent study conducted in young healthy subjects showed that modafinil was associated with reduced prefrontal brain activation when

41

completing working memory and attention tasks, as measured by functional magnetic resonance imaging (Rasetti et al., 2010). However, even in those disorders for which modafinil has approval from the U.S. Food and Drug Administration (FDA)—that is, narcolepsy, obstructive sleep apnea with residual excessive sleepiness, and shift-work disorder—effects on cognition remain poorly understood (see review by Kumar, 2008). Further, since modafinil inhibits several cytochrome P450 isoenzymes and has a number of drug interactions, reduced doses have been recommended for the elderly (Kumar, 2008). Some small trials of modafinil for hypersomnia in PD have been conducted but have produced inconsistent results (Kumar, 2008). A recent uncontrolled trial in 10 institutionalized patients with PD suggested positive results, but did not include cognition or broader measures of functioning (Lokk, 2010), while another small study showed positive effects on patient-rated sleepiness but not fatigue (Tyne et al., 2010). Clearly, further controlled studies incorporating broader outcome measures are required in this area since hypersomnia contributes to quality of life in PD (Naismith et al., 2010c).

Acetylcholinesterase inhibitors There may be some benefits of cholinesterase inhibitors for both sleep and cognition. Pharmacological manipulations with the commonly used cholinesterase inhibitors for memory loss in AD may alter sleep architecture, presumably by remediating cholinergic systems critical to REM sleep. While the mechanisms require further exploration, data suggest that M1 and M2 muscarinic receptors are involved in REM sleep regulation, with M1 playing a role in REM onset and M2 mediating REM maintenance and density (Nissen et al., 2006). In healthy older people (Schredl et al., 2001), as well as in those with AD, these medications have demonstrated improved sleep efficiency, shorter sleep latency,

and an increase in the amount of REM sleep correlating with cognitive improvements (Mizuno et al., 2004; Moraes Wdos et al., 2006). In one study, whereas REM deprivation did not appear to significantly influence performance on a declarative memory (paired associate learning) or procedural memory (mirror tracing) task, pharmacological augmentation of REM sleep by administering an acetylcholinesterase inhibitor was associated with improved procedural memory consolidation (Hornung et al., 2007). Although the mechanisms underpinning these changes are not fully understood, it is emerging that nocturnal cholinergic activity relates to sleep stage with higher levels of acetylcholine during REM phases, which may influence the synaptic consolidation of new memories in the cortex (for review, see Diekelmann and Born, 2010). Thus a greater appreciation of this cholinergic pattern of memory potentiation may have significant implications for the choice of cholinesterase inhibitors given the variable pharmacological profile of available preparations (Davis and Sadik, 2006). The use of cholinesterase inhibitors in PD has been generally restricted to those patients with established dementia. While showing some modest benefits on cognition, there is little data regarding their impact on sleep disturbance (Emre et al., 2004). Small studies exploring the use of cholinesterase inhibitors in LBD have suggested improvements in sleep as well as cognition (Grace et al., 2009; Maclean et al., 2001).

Nonpharmacological treatments For insomnia, it is likely that a combination of both pharmacological and nonpharmacological approaches is beneficial for short-term and longterm efficacy, respectively. However, further controlled studies are required in this area (see review by Bloom et al., 2009). The use of nonpharmacological approaches including cognitive behavior therapy, sleep restriction/sleep compression, and other behavioral approaches has been

42

studied and has largely been shown to be effective and sustainable for older people (Bloom et al., 2009; Morin et al., 1999). Napping may partly compensate for some of the changes in nocturnal sleep quality that occur with aging, and in broader age samples, napping has been shown to have similar benefits on learning as a night of sleep (Mednick et al., 2003). While further data are likely needed, there do not appear to be any negative cognitive effects of controlled napping. However, sleep inertia may emerge when awoken from SWS. Afternoon naps (between 1300 and 1600 h) appear to be preferred and are best studied. Although there is insufficient data to determine the optimal nap duration for older people, naps of 20–30 min have been mostly studied and may be less likely to interfere with nocturnal sleep. The use and duration of napping, however, has been cautioned as not being possible or even suitable for all older people, and those with hypersomnia versus insomnia may differentially benefit, with the former potentially alleviating sleep pressure through napping. There is also inconsistent data regarding the effects of napping on cognition and other waking functions. Thus, the use of napping should be tailored with respect to individuals’ profile and benefits (e.g., cognitive, mood, alertness) or side effects (e.g., sleep inertia) they may report (see review by Murphy and Campbell, 2010). Although complementary and alternative medicines tend to be popular, a recent overview indicated that there are limited data available supporting their efficacy for sleep–wake changes in older people. Examples include valerian, kava, acupuncture, relaxation techniques, yoga, tai chi, and music therapy. Exceptions to this are the use of melatonin, which has been extensively studied for its chronobiotic properties (described above), and relaxation therapy, where some data support its efficacy for insomnia (see review by Wider and Pittler, 2010). There is some evidence to suggest that multimodal approaches are particularly effective. These may incorporate increased daytime sunlight, social activity, less time in bed during the day, structured bed and wake times, and

nighttime noise reduction. Indeed, for healthy older people, daily social and physical activity has been shown to improve both SWS and neuropsychological functioning, particularly in the domain of memory (Naylor et al., 2000). For neurodegenerative conditions, however, data are generally lacking and inconsistent. Again, some multimodal treatment approaches have shown beneficial effects in terms of both sleep outcomes (actigraphy-defined nocturnal awakenings) and mood (McCurry et al., 2005), and it has been suggested that multimodal programs including caregivers may be optimal (Spijker et al., 2008). Due to their low costs, low risks, and acceptability, such programs may be recommended, but they are yet to be incorporated into clinical practice (Bloom et al., 2009). For the institutionalized elderly, external or environmental factors are likely to contribute to sleep–wake disruption and should be considered in management programs. This is certainly conceivable given that exposure to light in the daytime may be reduced, while nocturnal lighting and noise levels may be greater, and that there may be less opportunity for social interaction, physical exercise, and appropriately timed exposure to other zeitgebers (Harper et al., 2004; Wu and Swaab, 2007). Light therapy aims to increase exposure to bright light at appropriate times typically defined as the morning hours, and to reduce exposure at other times. Overall, most studies using light therapy have been positive although a few negative studies exist (reviewed by Wu and Swaab, 2007). Bright-light therapy in the evening may be beneficial for those with advanced sleep phase disorder in terms of delaying circadian rhythm and improving sleep (see reviews by Bloom et al., 2009). In patients with dementia, a four-week program of light therapy has been reported to decrease napping and increase nighttime sleep (see review by Bloom et al., 2009). It may also reduce agitation, increase circadian amplitude, and improve the consolidation of nighttime sleep. Thus, there is some evidence of efficacy in those with dementia and irregular sleep–wake patterns.

43

In an uncontrolled study of 12 patients with PD, beneficial effects of light therapy were evident on sleep, mood, and symptoms of bradykinesia and rigidity (Willis and Turner, 2007), while another randomized controlled trial study showed small benefits on mood, tremor, and other motor features (Paus et al., 2007). Trials are currently underway to examine the longer-term effects of twice-daily light therapy on patients with MCI, AD as well as those with only subjective memory complaints. These will measure multiple outcomes including depressive symptoms, sleep, and neuropsychological and endocrine function (Mst et al., 2010). In older patients with mild to moderate vascular dementia and insomnia (with no evidence of circadian phase shifting), a recent study (Mishima et al., 2005) successfully showed that passive body heating (immersion in 40 water for 30 min 2 h before bedtime) had sleep-promoting benefits possibly via altering the thermoregulatory and autonomic systems.

Treatments for parasomnias The current first-line treatment for RSBD is the administration of nocturnal clonazepam (at bedtime or 2 h prior to bedtime), where doses across studies range from 0.25 to 4 mg. Clonazepam has been shown to be effective in around 90% of patients (Olson et al., 2000; Schenck et al., 1993) though side effects usually include dizziness, somnolence, impotence, and nocturnal urinary incontinence. The action of this benzodiazepine is unknown, although it has been suggested that it may reduce muscle activity without restoring REM atonia. Other benzodiazepines have not been shown to be beneficial for RSBD (see Iranzo et al., 2009). However, melatonin (3–12 mg) has been shown to help in the treatment of RSBD across neurodegenerative diseases (Aurora et al., 2010), with beneficial effects within the first week of treatment. It may be particularly useful when clonazepam cannot be tolerated due to side effects or when clonazepam

is not indicated due to severe dementia or obstructive sleep apnea. Melatonin probably improves RSBD via restoration of the REM circadian rhythm. Other methods for improving symptoms of RSBD are minimization of b-blockers (which suppress melatonin secretion), antidepressants, or minimizing harm by behavioral strategies such as removal of furniture and placing the bed mattress on the floor. Summary It is clear that sleep–wake disturbances are a prominent feature of neurodegenerative diseases affecting older adults. They are associated with neuropsychiatric features, reduced daily functioning, disability, quality of life, caregiver burden, and the need for institutionalization (Naismith et al., 2010c). The estimates of prevalence of sleep–wake disturbances across these diseases are astoundingly high and encompass multiple symptom complexes including insomnia, hypersomnia, sleep-disordered breathing, circadian dysfunction, and RSBD. These symptoms are variously linked to changes in sleep architecture and the circadian system that overall suggest common involvement of underlying neural circuitry and their associated neuropharmacological and modulatory hormonal systems. While some data from healthy older adults suggest that sleep–wake disturbance may be a prognostic indicator for cognitive decline (Cricco et al., 2001), detailed analysis of the cross-sectional and longitudinal associations between these variables has yet to be performed in neurodegenerative diseases. Evidence to date that has linked sleep–wake disturbance with progression to dementia is most strongly supported by studies of RSBD, a feature that also appears to have particular prodromal and prognostic significance. Indeed, this feature is associated with a 40% risk of developing a neurodegenerative disease including PD, LBD, multiple system atrophy, and AD (Postuma et al., 2009) and is also

44

associated with accelerated decline. Unfortunately, characterization of circadian changes in diseases such as AD has often been conducted in those with moderate to severe stages of the diseases, although there is certainly some clinical evidence to suggest that this feature is associated with more rapid progression to dementia (Mortimer et al., 1992). Although the above data suggest an association between sleep–wake disturbances and cognitive decline, there is currently insufficient data available to conclusively support the notion that sleep–wake disturbance is an independent risk factor. Given the emerging literature linking sleep to neurogenesis and memory consolidation, it is certainly plausible that sleep disturbance could play a key etiological role in longitudinal cognitive and functional decline. Indeed, there are likely to be additive or synergistic links between the neurotoxic effects of the neurodegenerative diseases (e.g., b-amyloid, synucleinopathies, cerebrovascular disease, dysregulation of the hypothalamic–pituitary axis) and sleep–wake disturbances, with a multitude of factors influencing the capacity for neurogenesis, neuroprotection, and neuroplasticity, or conversely, impacting on accelerated decline. Further studies focusing on the prodromal periods of neurodegenerative diseases (e.g., MCI, late-life depression, RSBD) may shed light on these interrelationships and should aim to incorporate clinical, neuropsychological, neuroimaging, medical (e.g., inflammatory, vascular), and genetic (e.g., clock genes) data. What is additionally evident from the studies reviewed in this Chapter is that further intervention studies focusing on the link between the sleep–wake, mood, and cognitive systems are required in this population. There are currently little treatment data focusing specifically on neurodegenerative diseases, and although there appear to be studies demonstrating efficacy of the pharmacological and nonpharmacological interventions for healthy older adults, these studies have failed to consider many of the broader

effects on neuropsychological and neuropsychiatric features. Although some evidence suggests that sleep–wake disturbance may be a vulnerability marker or “biomarker” for cognitive decline or neurodegenerative disease, there has been little attempt to examine this relationship prospectively. Such studies may help to elucidate those most likely to progress to dementia, or to determine whether early interventions targeting sleep–wake disturbance have the capacity to delay or ameliorate the onset of cognitive decline or mood disturbance (Naismith et al., 2009a). With the rapidly aging population and subsequent increased rates of neurodegenerative diseases, research investigating whether sleep–wake disturbance is a “modifiable” risk factor for progression to dementia is urgently required. Acknowledgments The authors would like to thank Dr. Zoë Terpening and Dr. Louisa Norrie for their assistance in editing drafts of this chapter. Abbreviations AD DSM-IV

FTD LBD MCI NREM PD PSG REM RSBD SCN SWS VLPO WASO

Alzheimer’s disease Diagnostic and Statistical Manual of Mental Disorders—4th edition frontotemporal dementia Lewy body dementia mild cognitive impairment non-rapid eye movement sleep Parkinson’s disease polysomnography rapid eye movement sleep REM sleep behavior disorder suprachiasmatic nucleus slow-wave sleep ventrolateral preoptic nucleus wake after sleep onset

45

References Abbott, R. D., Ross, G. W., White, L. R., Tanner, C. M., Masaki, K. H., Nelson, J. S., et al. (2005). Excessive daytime sleepiness and subsequent development of Parkinson disease. Neurology, 65, 1442–1446. Alexopoulos, G. S., Young, R. C., & Meyers, B. S. (1993). Geriatric depression: Age of onset and dementia. Biological Psychiatry, 34, 141–145. Alves, G., Wentzel-Larsen, T., Aarsland, D., & Larsen, J. P. (2005). Progression of motor impairment and disability in Parkinson disease: A population-based study. Neurology, 65, 1436–1441. Anderson, K. N., Hatfield, C. F., Kipps, C., Hastings, M., & Hodges, J. R. (2008). Disrupted sleep and circadian patterns in frontotemporal dementia. European Journal of Neurology, 16, 317–323. Arendt, J., & Rajaratnam, S. M. (2008). Melatonin and its agonists: An update. The British Journal of Psychiatry, 193, 267–269. Aurora, R. N., Zak, R. S., Maganti, R. K., Auerbach, S. H., Casey, K. R., Chowdhuri, S., et al. (2010). Best practice guide for the treatment of REM sleep behavior disorder (RBD). Journal of Clinical Sleep Medicine, 6, 85–95. Autret, A., Lucas, B., Mondon, K., Hommet, C., Corcia, P., Saudeau, D., et al. (2001). Sleep and brain lesions: A critical review of the literature and additional new cases. Neurophysiologie Clinique, 31, 356–375. Banasr, M., Soumier, A., Hery, M., Mocaer, E., & Daszuta, A. (2006). Agomelatine, a new antidepressant, induces regional changes in hippocampal neurogenesis. Biological Psychiatry, 59, 1087–1096. Beaulieu-Bonneau, S., & Hudon, C. (2009). Sleep disturbances in older adults with mild cognitive impairment. International Psychogeriatrics, 21, 654–666. Benca, R. M., & Peterson, M. J. (2008). Insomnia and depression. Sleep Medicine, 9(Suppl. 1), S3–S9. Berlow, Y. A., Wells, W. M., Ellison, J. M., Sung, Y. H., Renshaw, P. F., & Harper, D. G. (2009). Neuropsychiatric correlates of white matter hyperintensities in Alzheimer’s disease. International Journal of Geriatric Psychiatry, 25, 780–788. Bliwise, D., Tinklenberg, J., Yesavage, J., Davies, H., Pursley, A., Petta, D., et al. (1989). REM latency in Alzheimer’s disease. Biological Psychiatry, 25, 320–328. Bliwise, D. L., Watts, R. L., Watts, N., Rye, D. B., Irbe, D., & Hughes, M. (1995). Disruptive nocturnal behavior in Parkinson’s disease and Alzheimer’s disease. Journal of Geriatric Psychiatry and Neurology, 8, 107–110. Bloom, H. G., Ahmed, I., Alessi, C. A., Ancoli-Israel, S., Buysse, D. J., Kryger, M. H., et al. (2009). Evidence-based recommendations for the assessment and management of sleep disorders in older persons. Journal of the American Geriatrics Society, 57, 761–789.

Bonanni, E., Maestri, M., Tognoni, G., Fabbrini, M., Nucciarone, B., Manca, M. L., et al. (2005). Daytime sleepiness in mild and moderate Alzheimer’s disease and its relationship with cognitive impairment. Journal of Sleep Research, 14, 311–317. Bordet, R., Devos, D., Brique, S., Touitou, Y., Guieu, J. D., Libersa, C., et al. (2003). Study of circadian melatonin secretion pattern at different stages of Parkinson’s disease. Clinical Neuropharmacology, 26, 65–72. Born, J., Rasch, B., & Gais, S. (2006). Sleep to remember. The Neuroscientist, 12, 410–424. Bozner, P., Grishko, V., Ledoux, S. P., Wilson, G. L., Chyan, Y. C., & Pappolla, M. A. (1997). The amyloid beta protein induces oxidative damage of mitochondrial DNA. Journal of Neuropathology and Experimental Neurology, 56, 1356–1362. Braak, H., Alafuzoff, I., Arzberger, T., Kretzschmar, H., & Del Tredici, K. (2006a). Staging of Alzheimer diseaseassociated neurofibrillary pathology using paraffin sections and immunocytochemistry. Acta Neuropathologica, 112, 389–404. Braak, H., Muller, C. M., Rub, U., Ackermann, H., Bratzke, H., De Vos, R. A., et al. (2006b). Pathology associated with sporadic Parkinson’s disease—Where does it end? The Journal of Neural Transmission, (Suppl. No 70), 89–97. Braak, H., Rub, U., & Del Tredici, K. (2006c). Cognitive decline correlates with neuropathological stage in Parkinson’s disease. Journal of the Neurological Sciences, 248, 255–258. Branchey, L., Weinberg, U., & Branchey, M. (1982). Simultaneous study of 24-hour patterns of melatonin and cortisol secretion in depressed patients. Neuropsychobiology, 8, 225. Brankack, J., Platt, B., & Riedel, G. (2009). Sleep and hippocampus: Do we search for the right things? Progress in Neuropsychopharmacology & Biological Psychiatry, 33, 806–812. Brusco, L. I., Marquez, M., & Cardinali, D. P. (2000). Melatonin treatment stabilizes chronobiologic and cognitive symptoms in Alzheimer’s disease. Neuroendocrinology Letters, 21, 39–42. Brzezinski, A., Vangel, M. G., Wurtman, R. J., Norrie, G., Zhdanova, I., Ben-Shushan, A., et al. (2005). Effects of exogenous melatonin on sleep: A meta-analysis. Sleep Medicine Reviews, 9, 41–50. Buysse, D. J. (2004). Insomnia, depression and aging. Assessing sleep and mood interactions in older adults. Geriatrics, 59, 47–51 Quiz 52. Buysse, D. J., Reynolds, C. F., 3rd, Houck, P. R., Stack, J., & Kupfer, D. J. (1988). Age of illness onset and sleep EEG variables in elderly depressives. Biological Psychiatry, 24, 355–359.

46 Carvalho, L., Gorenstein, C., Moreno, R., & Markus, R. (2006). Melatonin levels in drug-free patients with major depression from the southern hemisphere. Psychoneuroendocrinology, 31, 761–768. Cho, H., Lavretsky, H., Olmstead, R., Levin, M., Oxman, M., & Irwin, M. (2008). Sleep disturbance and depression recurrence in community-dwelling older adults: A prospective study. The American Journal of Psychiatry, 165, 1543–1550. Claustrat, B. (1984). A chronobiological study of melatonin and cortisol secretion in depressed subjects: Plasma melatonin, a biochemical marker in major depression. Biological Psychiatry, 19, 1215–1228. Cohen-Mansfield, J., Garfinkel, D., & Lipson, S. (2000). Melatonin for treatment of sundowning in elderly persons with dementia—A preliminary study. Archives of Gerontology and Geriatrics, 31, 65–76. Compta, Y., Santamaria, J., Ratti, L., Tolosa, E., Iranzo, A., Munoz, E., et al. (2009). Cerebrospinal hypocretin, daytime sleepiness and sleep architecture in Parkinson’s disease dementia. Brain, 132, 3308–3317. Cooke, J. R., Liu, L., Natarajan, L., He, F., Marler, M., Loredo, J. S., et al. (2006). The effect of sleep-disordered breathing on stages of sleep in patients with Alzheimer’s disease. Behavioral Sleep Medicine, 4, 219–227. Copeland, J., Beekman, A., Braam, A., Dewey, M., Delespaul, P., Fuhrer, R., et al. (2004). Depression among older people in Europe: The EURODEP studies. World Psychiatry, 3, 45–49. Cricco, M., Simonsick, E. M., & Foley, D. J. (2001). The impact of insomnia on cognitive functioning in older adults. Journal of the American Geriatrics Society, 49, 1185–1189. Davis, B., & Sadik, K. (2006). Circadian cholinergic rhythms: Implications for cholinesterase inhibitor therapy. Dementia and Geriatric Cognitive Disorders, 21, 120–129. De Gennaro, L., & Ferrara, M. (2003). Sleep spindles: An overview. Sleep Medicine Reviews, 7, 423–440. De Rijk, M. C., Breteler, M. M., Graveland, G. A., Ott, A., Grobbee, D. E., Van Der Meche, F. G., et al. (1995). Prevalence of Parkinson’s disease in the elderly: The Rotterdam Study. Neurology, 45, 2143–2146. Dew, M. A., Reynolds, C. F., 3rd, Buysse, D. J., Houck, P. R., Hoch, C. C., Monk, T. H., et al. (1996). Electroencephalographic sleep profiles during depression. Effects of episode duration and other clinical and psychosocial factors in older adults. Archives of General Psychiatry, 53, 148–156. Diederich, N. J., Vaillant, M., Mancuso, G., Lyen, P., & Tiete, J. (2005). Progressive sleep ‘destructuring’ in Parkinson’s disease. A polysomnographic study in 46 patients. Sleep Medicine, 6, 313–318. Diekelmann, S., & Born, J. (2010). The memory function of sleep. Nature Reviews Neuroscience, 11, 114–126. Dinges, D. F. (2005). Can habitual sleep duration harbor sleep debt? Sleep, 28, 1209–1210.

Dombrovski, A. Y., Cyranowski, J. M., Mulsant, B. H., Houck, P. R., & Buysse, D. J. (2008). Which symptoms predict recurrence of depression in women treated with maintenance interpersonal psychotherapy? Depression and Anxiety, 25, 1060–1066 (1091–4269). Dorrego, M. F., Sabe, L., Garcia Cuerva, A., Kuzis, G., Tiberti, C., Boller, F., et al. (1999). Remote memory in Alzheimer’s disease. The Journal of Neuropsychiatry and Clinical Neurosciences, 11, 490–497. Dowling, G. A., Burr, R. L., Van Someren, E. J., Hubbard, E. M., Luxenberg, J. S., Mastick, J., et al. (2008). Melatonin and bright-light treatment for rest-activity disruption in institutionalized patients with Alzheimer’s disease. Journal of the American Geriatrics Society, 56, 239–246. Dowling, G. A., Mastick, J., Colling, E., Carter, J. H., Singer, C. M., & Aminoff, M. J. (2005). Melatonin for sleep disturbances in Parkinson’s disease. Sleep Medicine, 6, 459–466. Dresler, M., Kluge, M., Genzel, L., Schussler, P., & Steiger, A. (2010). Impaired off-line memory consolidation in depression. European Neuropsychopharmacology, 20, 553–561. Emre, M., Aarsland, D., Albanese, A., Byrne, E. J., Deuschl, G., De Deyn, P. P., et al. (2004). Rivastigmine for dementia associated with Parkinson’s disease. The New England Journal of Medicine, 351, 2509–2518. Fava, M., Thase, M. E., Debattista, C., Doghramji, K., Arora, S., & Hughes, R. J. (2007). Modafinil augmentation of selective serotonin reuptake inhibitor therapy in MDD partial responders with persistent fatigue and sleepiness. Annals of Clinical Psychiatry, 19, 153–159. Fernandez Martinez, M., Castro Flores, J., Perez De Las Heras, S., Mandaluniz Lekumberri, A., Gordejuela Menocal, M., & Zarranz Imirizaldu, J. J. (2008). Prevalence of neuropsychiatric symptoms in elderly patients with dementia in Mungialde County (Basque Country, Spain). Dementia and Geriatric Cognitive Disorders, 25, 103–108. Ferrara, M., Iaria, G., Tempesta, D., Curcio, G., Moroni, F., Marzano, C., et al. (2008). Sleep to find your way: The role of sleep in the consolidation of memory for navigation in humans. Hippocampus, 18, 844–851. Fertl, E., Auff, E., Doppelbauer, A., & Waldhauser, F. (1991). Circadian secretion pattern of melatonin in Parkinson’s disease. Journal of Neural Transmission Parkinson’s Disease and Dementia Section, 3, 41–47. Fertl, E., Auff, E., Doppelbauer, A., & Waldhauser, F. (1993). Circadian secretion pattern of melatonin in de novo Parkinsonian patients: Evidence for phase-shifting properties of Ldopa. Journal of Neural Transmission Parkinson’s Disease and Dementia Section, 5, 227–234. Floyd, J. A., Medler, S. M., Ager, J. W., & Janisse, J. J. (2000). Age-related changes in initiation and maintenance of sleep: A meta-analysis. Research in Nursing and Health, 23, 106–117.

47 Fronczek, R., Overeem, S., Lee, S. Y., Hegeman, I. M., Van Pelt, J., Van Duinen, S. G., et al. (2007). Hypocretin (orexin) loss in Parkinson’s disease. Brain, 130, 1577–1585. Fuller, P. M., Gooley, J. J., & Saper, C. B. (2006). Neurobiology of the sleep-wake cycle: Sleep architecture, circadian regulation, and regulatory feedback. Journal of Biological Rhythms, 21, 482–493. Furio, A. M., Brusco, L. I., & Cardinali, D. P. (2007). Possible therapeutic value of melatonin in mild cognitive impairment: A retrospective study. Journal of Pineal Research, 43, 404–409. Gagnon, J. F., Petit, D., Latreille, V., & Montplaisir, J. (2008). Neurobiology of sleep disturbances in neurodegenerative disorders. Current Pharmaceutical Design, 14, 3430–3445. Gallagher-Thompson, D., Brooks, J. O., 3rd, Bliwise, D., Leader, J., & Yesavage, J. A. (1992). The relations among caregiver stress, "sundowning" symptoms, and cognitive decline in Alzheimer’s disease. Journal of the American Geriatrics Society, 40, 807–810. Garcia-Borreguero, D., Larrosa, O., & Bravo, M. (2003). Parkinson’s disease and sleep. Sleep Medicine Reviews, 7, 115–129. Gauthier, S., Reisberg, B., Zaudig, M., Petersen, R. C., Ritchie, K., Broich, K., et al. (2006). Mild cognitive impairment. Lancet, 367, 1262–1270. Gehrman, P. R., Connor, D. J., Martin, J. L., Shochat, T., Corey-Bloom, J., & Ancoli-Israel, S. (2009). Melatonin fails to improve sleep or agitation in double-blind randomized placebo-controlled trial of institutionalized patients with Alzheimer disease. The American Journal of Geriatric Psychiatry, 17, 166–169. Gehrman, P. R., & Gooneratne, N. S. (2010). Non-pharmacological treatment of insomnia in the elderly: Cognitive behavioral therapies. In S. R. Pandi-Perumal, J. M. Monti & A. A. Monjan (Eds.), Principles and practice of geriatric sleep medicine. United Kingdom: Cambridge University Press. Giubilei, F., Patacchioli, F. R., Antonini, G., Sepe Monti, M., Tisei, P., Bastianello, S., et al. (2001). Altered circadian cortisol secretion in Alzheimer’s disease: Clinical and neuroradiological aspects. Journal of Neuroscience Research, 66, 262–265. Giuditta, A., Ambrosini, M. V., Montagnese, P., Mandile, P., Cotugno, M., Grassi Zucconi, G., et al. (1995). The sequential hypothesis of the function of sleep. Behavioural Brain Research, 69, 157–166. Glass, J., Lanctot, K. L., Herrmann, N., Sproule, B. A., & Busto, U. E. (2005). Sedative hypnotics in older people with insomnia: Meta-analysis of risks and benefits. British Medical Journal, 331, 1169. Grace, J., Amick, M. M., & Friedman, J. H. (2009). A doubleblind comparison of galantamine hydrobromide ER and placebo in Parkinson disease. Journal of Neurology, Neurosurgery and Psychiatry, 80, 18–23.

Grace, J. B., Walker, M. P., & Mckeith, I. G. (2000). A comparison of sleep profiles in patients with dementia with Lewy bodies and Alzheimer’s disease. International Journal of Geriatric Psychiatry, 15, 1028–1033. Gressens, P., Schwendimann, L., Husson, I., Sarkozy, G., Mocaer, E., Vamecq, J., et al. (2008). Agomelatine, a melatonin receptor agonist with 5-HT(2C) receptor antagonist properties, protects the developing murine white matter against excitotoxicity. European Journal of Pharmacology, 588, 58–63. Grunhaus, L., Shipley, J. E., Eiser, A., Pande, A. C., Tandon, R., Remen, A., et al. (1996). Polysomnographic studies in patients referred for ECT: Pre-ECT studies. Convulsive Therapy, 12, 224–231. Gunn, D., Naismith, S. L., & Lewis, S. J. G. (2010). Sleep disturbances in Parkinson’s disease and their potential role in heterogeneity. Journal of Geriatric Psychiatry and Neurology, 23, 123–130. Gutchess, A. H., Welsh, R. C., Hedden, T., Bangert, A., Minear, M., Liu, L. L., et al. (2005). Aging and the neural correlates of successful picture encoding: Frontal activations compensate for decreased medial-temporal activity. Journal of Cognitive Neuroscience, 17, 84–96. Halliday, G. (2008). Clarifying the pathological progression of Parkinson’s disease. Acta Neuropathologica, 115, 377–378. Happe, S., & Berger, K. (2002). The association between caregiver burden and sleep disturbances in partners of patients with Parkinson’s disease. Age and Ageing, 31, 349–354. Harper, D. G., Stopa, E. G., Mckee, A. C., Satlin, A., Fish, D., & Volicer, L. (2004). Dementia severity and Lewy bodies affect circadian rhythms in Alzheimer disease. Neurobiology of Aging, 25, 771–781. Harper, D. G., Stopa, E. G., Mckee, A. C., Satlin, A., Harlan, P. C., Goldstein, R., et al. (2001). Differential circadian rhythm disturbances in men with Alzheimer disease and frontotemporal degeneration. Archives of General Psychiatry, 58, 353–360. Harper, D. G., Volicer, L., Stopa, E. G., Mckee, A. C., Nitta, M., & Satlin, A. (2005). Disturbance of endogenous circadian rhythm in aging and Alzheimer disease. The American Journal of Geriatric Psychiatry, 13, 359–368. Hatfield, C. F., Herbert, J., Van Someren, E. J., Hodges, J. R., & Hastings, M. H. (2004). Disrupted daily activity/rest cycles in relation to daily cortisol rhythms of home-dwelling patients with early Alzheimer’s dementia. Brain, 127, 1061–1074. Hely, M. A., Reid, W. G., Adena, M. A., Halliday, G. M., & Morris, J. G. (2008). The Sydney multicenter study of Parkinson’s disease: The inevitability of dementia at 20 years. Movement Disorders, 23, 837–844. Hickie, I., Naismith, S., Ward, P. B., Turner, K., Scott, E., Mitchell, P., et al. (2005). Reduced hippocampal volumes and memory loss in patients with early- and late-onset depression. The British Journal of Psychiatry, 186, 197–202.

48 Hickie, I., Scott, E., Mitchell, P., Wilhem, K., Austin, M., & Bennett, B. (1995). Subcortical hyperintensities on magnetic resonance imaging: Clinical correlates and prognostic significance in patients with severe depression. Biological Psychiatry, 37, 151–160. Hickie, I., Scott, E., Wilhelm, K., & Brodaty, H. (1997). Subcortical hyperintensities on magnetic resonance imaging in patients with severe depression–a longitudinal evaluation. Biological Psychiatry, 42, 367–374. Hobson, D. E., Lang, A. E., Martin, W. R., Razmy, A., Rivest, J., & Fleming, J. (2002). Excessive daytime sleepiness and sudden-onset sleep in Parkinson disease: A survey by the Canadian Movement Disorders Group. Journal of American Medical Association, 287, 455–463. Hornung, O. P., Regen, F., Danker-Hopfe, H., Schredl, M., & Heuser, I. (2007). The relationship between REM sleep and memory consolidation in old age and effects of cholinergic medication. Biological Psychiatry, 61, 750–757. Iranzo, A., Santamaria, J., & Tolosa, E. (2009). The clinical and pathophysiological relevance of REM sleep behavior disorder in neurodegenerative diseases. Sleep Medicine Reviews, 13, 385–401. Jean-Louis, G., Von Gizycki, H., & Zizi, F. (1998). Melatonin effects on sleep, mood, and cognition in elderly with mild cognitive impairment. Journal of Pineal Research, 25, 177–183. Ji, D., & Wilson, M. A. (2007). Coordinated memory replay in the visual cortex and hippocampus during sleep. Nature Neuroscience, 10, 100–107. Kasper, S., Hajak, G., Wulff, K., Hoogendijk, W. J., Montejo, A. L., Smeraldi, E., et al. (2010). Efficacy of the novel antidepressant agomelatine on the circadian restactivity cycle and depressive and anxiety symptoms in patients with major depressive disorder: A randomized, double-blind comparison with sertraline. The Journal of Clinical Psychiatry, 71, 109–120. Kawinska, A., Dumont, M., Selmaoui, B., Paquet, J., & Carrier, J. (2005). Are modifications of melatonin circadian rhythm in the middle years of life related to habitual patterns of light exposure? Journal of Biological Rhythms, 20, 451. Kennaway, D. J., Lushington, K., Dawson, D., Lack, L., Van Den Heuvel, C., & Rogers, N. (1999). Urinary 6-sulfatoxymelatonin excretion and aging: New results and a critical review of the literature. Journal of Pineal Research, 27, 210. Krystal, A. D. (2010). Geriatric psychopharmacology: An overview. In S. R. Pandi-Perumal, J. M. Monti & A. A. Monjan (Eds.), Geriatric sleep medicine. United Kingdom: Cambridge University Press. Kumar, V. M. (2008). Sleep and sleep disorders. The Indian Journal of Chest Diseases & Allied Sciences, 50, 129–135. Kumru, H., Santamaria, J., Tolosa, E., & Iranzo, A. (2007). Relation between subtype of Parkinson’s disease and REM sleep behavior disorder. Sleep Medicine, 8, 779–783.

Landolt, H., & Gillin, J. (2000). GABAA1a receptors: Involvement in sleep regulation and potential of selective agonists in the treatment of insomnia. Central Nervous System, 13, 185–199. Langston, J. W., & Forno, L. S. (1978). The hypothalamus in Parkinson disease. Annals of Neurology, 3, 129–133. Lauer, C. J., Riemann, D., Wiegand, M., & Berger, M. (1991). From early to late adulthood changes in EEG sleep of depressed patients and healthy volunteers. Biological Psychiatry, 29, 979–993. Lee, J. H., Bliwise, D. L., Ansari, F. P., Goldstein, F. C., Cellar, J. S., Lah, J. J., et al. (2007). Daytime sleepiness and functional impairment in Alzheimer disease. The American Journal of Geriatric Psychiatry, 15, 620–626. Lewis, S. J., Dove, A., Robbins, T. W., Barker, R. A., & Owen, A. M. (2003). Cognitive impairments in early Parkinson’s disease are accompanied by reductions in activity in frontostriatal neural circuitry. The Journal of Neuroscience, 23, 6351–6356. Lewy, A. (2002). Circadian phase sleep and mood disorders. In Philadelphia: Lippincott Williams and Wilkins, pp. 1880–1893. Lezak, M. D. (1995). Neuropsychological assessment. New York: Oxford University Press. Lokk, J. (2010). Daytime sleepiness in elderly Parkinson’s disease patients and treatment with the psychostimulant modafinil: A preliminary study. Neuropsychiatric Disease Treatment, 6, 93–97. Luboshitzky, R., Shen-Orr, Z., Tzischichinsky, O., Maldonado, M., Herer, P., & Lavie, P. (2001). Actigraphic sleep-wake patterns and urinary 6-sulfatoxymelatonin excretion in patients with Alzheimer’s disease. Chronobiology International, 18, 513. Maclean, L. E., Collins, C. C., & Byrne, E. J. (2001). Dementia with Lewy bodies treated with rivastigmine: Effects on cognition, neuropsychiatric symptoms, and sleep. International Psychogeriatrics, 13, 277–288. Maetzler, W., Liepelt, I., & Berg, D. (2009). Progression of Parkinson’s disease in the clinical phase: Potential markers. Lancet Neurology, 8, 1158–1171. Mallon, L., Broman, J.-E., & Hetta, J. (2000). Relationship between insomnia, depression, and mortality: A 12-year follow-up of older adults in the community. International Psychogeriatrics, 12, 295–306. Marion, M. H., Qurashi, M., Marshall, G., & Foster, O. (2008). Is REM sleep behaviour disorder (RBD) a risk factor of dementia in idiopathic Parkinson’s disease? Journal of Neurology, 255, 192–196. Marshall, L., & Born, J. (2007). The contribution of sleep to hippocampus-dependent memory consolidation. Trends in Cognitive Sciences, 11, 442–450. Mccurry, S. M., Gibbons, L. E., Logsdon, R. G., Vitiello, M. V., & Teri, L. (2005). Nighttime insomnia treatment and education for Alzheimer’s disease: A randomized, controlled trial. Journal of the American Geriatrics Society, 53, 793–802.

49 Medeiros, C. A., Carvalhedo De Bruin, P. F., Lopes, L. A., Magalhaes, M. C., De Lourdes Seabra, M., & De Bruin, V. M. (2007). Effect of exogenous melatonin on sleep and motor dysfunction in Parkinson’s disease. A randomized, double blind, placebo-controlled study. Journal of Neurology, 254, 459–464. Mednick, S., Nakayama, K., & Stickgold, R. (2003). Sleepdependent learning: A nap is as good as a night. Nature Neuroscience, 6, 697–698. Meerlo, P., Mistlberger, R. E., Jacobs, B. L., Heller, H. C., & Mcginty, D. (2009). New neurons in the adult brain: The role of sleep and consequences of sleep loss. Sleep Medicine Reviews, 13, 187–194. Mills, P. J., Ancoli-Israel, S., Von Kanel, R., Mausbach, B. T., Aschbacher, K., Patterson, T. L., et al. (2009). Effects of gender and dementia severity on Alzheimer’s disease caregivers’ sleep and biomarkers of coagulation and inflammation. Brain, Behavior, and Immunity, 23, 605–610. Mishima, Y., Hozumi, S., Shimizu, T., Hishikawa, Y., & Mishima, K. (2005). Passive body heating ameliorates sleep disturbances in patients with vascular dementia without circadian phase-shifting. The American Journal of Geriatric Psychiatry, 13, 369–376. Mishima, K., Okawa, M., Hozumi, S., & Hishikawa, Y. (2000). Supplementary administration of artificial bright light and melatonin as potent treatment for disorganized circadian rest-activity and dysfunctional autonomic and neuroendocrine systems in institutionalized demented elderly persons. Chronobiology International, 17, 419–432. Mizuno, S., Kameda, A., Inagaki, T., & Horiguchi, J. (2004). Effects of donepezil on Alzheimer’s disease: The relationship between cognitive function and rapid eye movement sleep. Psychiatry and Clinical Neurosciences, 58, 660–665. Montplaisir, J., Petit, D., Lorrain, D., Gauthier, S., & Nielsen, T. (1995). Sleep in Alzheimer’s disease: Further considerations on the role of brainstem and forebrain cholinergic populations in sleep-wake mechanisms. Sleep, 18, 145–148. Moraes Wdos, S., Poyares, D. R., Guilleminault, C., Ramos, L. R., Bertolucci, P. H., & Tufik, S. (2006). The effect of donepezil on sleep and REM sleep EEG in patients with Alzheimer disease: A double-blind placebo-controlled study. Sleep, 29, 199–205. Moran, M., Lynch, C. A., Walsh, C., Coen, R., Coakley, D., & Lawlor, B. A. (2005). Sleep disturbance in mild to moderate Alzheimer’s disease. Sleep Medicine, 6, 347–352. Morin, C. M., Colecchi, C., Stone, J., Sood, R., & Brink, D. (1999). Behavioral and pharmacological therapies for latelife insomnia: A randomized controlled trial. Journal of American Medical Association, 281, 991–999. Morris, G. O., Williams, H. L., & Lubin, A. (1960). Misperception and disorientation during sleep deprivation. Archives of General Psychiatry, 2, 247–254.

Mortimer, J. A., Ebbitt, B., Jun, S. P., & Finch, M. D. (1992). Predictors of cognitive and functional progression in patients with probable Alzheimer’s disease. Neurology, 42, 1689–1696. Mst, E. I., Scheltens, P., & Van Someren, E. J. (2010). Prevention of depression and sleep disturbances in elderly with memory-problems by activation of the biological clock with light–a randomized clinical trial. Trials, 11, 19. Murphy, P. J., & Campbell, S. S. (2010). Therapeutic benefits of napping in the elderly. In S. R. Pandi-Perumal, J. M. Monti & A. A. Monjan (Eds.), Principles and practice of geriatric sleep medicine. United Kingdom: Cambridge University Press. Murphy, E., Smith, R., Lindesay, J., & Slattery, J. (1998). Increased mortality rates in late-life depression. The British Journal of Psychiatry, 152, 347–353. Naismith, S. L., Glozier, N., Burke, D., Carter, P., Scott, E., & Hickie, I. (2009a). Early intervention for cognitive decline: Is there a role for multiple medical or behavioural interventions? Early Intervention in Psychiatry, 3, 19–27. Naismith, S. L., Norrie, L., Lewis, S. J., Rogers, N. L., Scott, E. M., & Hickie, I. B. (2009b). Does sleep disturbance mediate neuropsychological functioning in older people with depression? Journal of Affective Disorders, 116, 139–143. Naismith, S. L., Hickie, I. B., Turner, K., Little, C. L., Winter, V., Ward, P. B., et al. (2003). Neuropsychological performance in patients with depression is associated with clinical, etiological and genetic risk factors. Journal of Clinical and Experimental Neuropsychology, 25, 866–877. Naismith, S., Hickie, I., Ward, P. B., Turner, K., Scott, E., Little, C., et al. (2002). Caudate nucleus volumes and genetic determinants of homocysteine metabolism in the prediction of psychomotor speed in older persons with depression. The American Journal of Psychiatry, 159, 2096–2098. Naismith, S. L., Rogers, N. L., Hickie, I. B., Mackenzie, J., Norrie, L. M., & Lewis, S. J. (2010a). Sleep well, think well: Sleep-wake disturbance in mild cognitive impairment. Journal of Geriatric Psychiatry and Neurology, 23, 123–130. Naismith, S. L., Rogers, N. L., Mackenzie, J., Hickie, I. B., & Lewis, S. J. G. (2010b). The relationship between actigraphically-defined sleep disturbance and REM Sleep Behaviour Disorder in Parkinson’s Disease. Clinical Neurology and Neurosurgery, 112, 420–423. Naismith, S. L., Hickie, I. B., & Lewis, S. J. G. (2010c). The role of mild depression in sleep disturbance and quality of life in Parkinson’s Disease. The Journal of Neuropsychiatry and Clinical Neurosciences, 22, 384–389. Naismith, S. L., Lagopoulos, J., Ward, P. B., Davey, C., & Hickie, I. B. (2010d). Fronto-striatal correlates of impaired implicit sequence learning in major depression: An fMRI study. Journal of Affective Disorders, 125, 256–261. Naismith, S. L., Rogers, N.L., Lewis, S. J. G., Terpening, Z., Ip, T., Diamond, K., et al. (2011). Sleep disturbance relates

50 to neuropsychological functioning in late-life depression. Journal of Affective Disorders. Mar 23. [Epub ahead of print]. Naismith, S., Winter, V., Gotsopoulos, H., Hickie, I., & Cistulli, P. (2004). Neurobehavioral functioning in obstructive sleep apnea: Differential effects of sleep quality, hypoxemia and subjective sleepiness. Journal of Clinical and Experimental Neuropsychology, 26, 43–54. Naylor, E., Penev, P. D., Orbeta, L., Janssen, I., Ortiz, R., Colecchia, E. F., et al. (2000). Daily social and physical activity increases slow-wave sleep and daytime neuropsychological performance in the elderly. Sleep, 23, 87–95. Nebes, R. D., Buysse, D. J., Halligan, E. M., Houck, P. R., & Monk, T. H. (2009). Self-reported sleep quality predicts poor cognitive performance in healthy older adults. The Journals of Gerontology Series B: Psychological Sciences and Social Sciences, 64, 180–187. Nelson, J. C., Clary, C. M., Leon, A. C., Schneider, L. S., Nelson, J. C., Clary, C. M., et al. (2005). Symptoms of latelife depression: Frequency and change during treatment. The American Journal of Geriatric Psychiatry, 13, 520–526. Nissen, C., Nofzinger, E. A., Feige, B., Waldheim, B., Radosa, M. P., Riemann, D., et al. (2006). Differential effects of the muscarinic M1 receptor agonist RS-86 and the acetylcholine-esterase inhibitor donepezil on REM sleep regulation in healthy volunteers. Neuropsychopharmacology, 31, 1294–1300. Olson, E. J., Boeve, B. F., & Silber, M. H. (2000). Rapid eye movement sleep behaviour disorder: Demographic, clinical and laboratory findings in 93 cases. Brain, 123(Pt 2), 331–339. Ondo, W. G., Dat Vuong, K., Khan, H., Atassi, F., Kwak, C., & Jankovic, J. (2001). Daytime sleepiness and other sleep disorders in Parkinson’s disease. Neurology, 57, 1392–1396. Oosterman, J. M., Van Someren, E. J., Vogels, R. L., Van Harten, B., & Scherder, E. J. (2009). Fragmentation of the rest-activity rhythm correlates with age-related cognitive deficits. Journal of Sleep Research, 18, 129–135. Overeem, S., Van Hilten, J. J., Ripley, B., Mignot, E., Nishino, S., & Lammers, G. J. (2002). Normal hypocretin-1 levels in Parkinson’s disease patients with excessive daytime sleepiness. Neurology, 58, 498–499. Pacchierotti, C. (2001). Melatonin in psychiatric disorders: A review on the melatonin involvement in psychiatry. Frontiers in Neuroendocrinology, 22, 18–32. Pace-Schott, E. F., & Hobson, J. A. (2002). The neurobiology of sleep: Genetics, cellular physiology and subcortical networks. Nature Reviews Neuroscience, 3, 591–605. Pandi-Perumal, S. R., Srinivasan, V., Maestroni, G. J., Cardinali, D. P., Poeggeler, B., & Hardeland, R. (2006). Melatonin: Nature’s most versatile biological signal? The FEBS Journal, 273, 2813–2838. Pandi-Perumal, S. R., Zisapel, N., Srinivasan, V., & Cardinali, D. P. (2005). Melatonin and sleep in aging population. Experimental Gerontology, 40, 911–925.

Parry, B., & Newton, P. (2001). Chronobiological basis of female-specific mood disorders. Neuropsychopharmacology, 25, S102–S108. Paus, S., Schmitz-Hubsch, T., Wullner, U., Vogel, A., Klockgether, T., & Abele, M. (2007). Bright light therapy in Parkinson’s disease: A pilot study. Movement Disorders, 22, 1495–1498. Peigneux, P., Laureys, S., Fuchs, S., Collette, F., Perrin, F., Reggers, J., et al. (2004). Are spatial memories strengthened in the human hippocampus during slow wave sleep? Neuron, 44, 535–545. Perlis, M. L., Smith, L. J., Lyness, J. M., Matteson, S. R., Pigeon, W. R., Jungquist, C. R., et al. (2006). Insomnia as a risk factor for onset of depression in the elderly. Behavioral Sleep Medicine, 4, 104–113. Peters, K. R., Ray, L., Smith, V., & Smith, C. (2008). Changes in the density of stage 2 sleep spindles following motor learning in young and older adults. Journal of Sleep Research, 17, 23–33. Petersen, R. C., & Morris, J. C. (2005). Mild cognitive impairment as a clinical entity and treatment target. Archives of Neurology, 62, 1160–1163, discussion. 1167. Pigeon, W. R., Hegel, M., Unutzer, J., Fan, M. Y., Sateia, M. J., Lyness, J. M., et al. (2008). Is insomnia a perpetuating factor for late-life depression in the IMPACT cohort? Sleep, 31, 481–488. Politis, M., Piccini, P., Pavese, N., Koh, S. B., & Brooks, D. J. (2008). Evidence of dopamine dysfunction in the hypothalamus of patients with Parkinson’s disease: An in vivo 11Craclopride PET study. Experimental Neurology, 214, 112–116. Pollak, C. P., & Perlick, D. (1991). Sleep problems and institutionalization of the elderly. Journal of Geriatric Psychiatry and Neurology, 4, 204–210. Postuma, R. B., Gagnon, J. F., Vendette, M., Fantini, M. L., Massicotte-Marquez, J., & Montplaisir, J. (2009). Quantifying the risk of neurodegenerative disease in idiopathic REM sleep behavior disorder. Neurology, 72, 1296–1300. Quera Salva, M. A., Vanier, B., Laredo, J., Hartley, S., Chapotot, F., Moulin, C., et al. (2007). Major depressive disorder, sleep EEG and agomelatine: An open-label study. The International Journal of Neuropsychopharmacology, 10, 691–696. Rasch, B., Pommer, J., Diekelmann, S., & Born, J. (2009). Pharmacological REM sleep suppression paradoxically improves rather than impairs skill memory. Nature Neuroscience, 12, 396–397. Rasetti, R., Mattay, V. S., Stankevich, B., Skjei, K., Blasi, G., Sambataro, F., et al. (2010). Modulatory effects of modafinil on neural circuits regulating emotion and cognition. Neuropsychopharmacology, 35(10), 2101–2109. Ratnavalli, E., Brayne, C., Dawson, K., & Hodges, J. R. (2002). The prevalence of frontotemporal dementia. Neurology, 58, 1615–1621.

51 Rauchs, G., Bertran, F., Guillery-Girard, B., Desgranges, B., Kerrouche, N., Denise, P., et al. (2004). Consolidation of strictly episodic memories mainly requires rapid eye movement sleep. Sleep, 27, 395–401. Reiter, R. J. (1998). Oxidative damage in the central nervous system: Protection by melatonin. Progress in Neurobiology, 56, 359–384. Rogers, N. L., & Dinges, D. F. (2008). Interaction of chronic sleep restriction and circadian system in humans. Journal of Sleep Research, 17, 406–411. Rogers, N., Dinges, D., Kennaway, D., & Dawson, D. (2003). Potential action of melatonin in insomnia. Sleep, 26, 1058–1105. Rogers, N. L., Dorrian, J., & Dinges, D. F. (2003). Sleep, waking and neurobehavioural performance. Frontiers in Bioscience, 8, s1056–s1067. Rogers, N. L., Kennaway, D. J., & Dawson, D. (2003). Neurobehavioural performance effects of daytime melatonin and temazepam administration. Journal of Sleep Research, 12, 207–212. Rogers, N. L., & Kloss, J. D. (2005). Medical conditions and diseases. In C. A. Kushida (Ed.), Sleep deprivation: Clinical issues, pharmacology and sleep loss effects. New York: Marcel Dekker, Inc.. Rongve, A., Boeve, B. F., & Aarsland, D. (2010). Frequency and correlates of caregiver-reported sleep disturbances in a sample of persons with early dementia. Journal of the American Geriatrics Society, 58, 480–486. Rubin, R. T. (1992). Neuroendocrine aspects of primary endogenous-depression: Serum melatonin measures in patients and matched control subjects. Archives of General Psychiatry, 49, 558–567. Rye, D. B., Bliwise, D. L., Dihenia, B., & Gurecki, P. (2000). FAST TRACK: Daytime sleepiness in Parkinson’s disease. Journal of Sleep Research, 9, 63–69. Saper, C. B., Scammell, T. E., & Lu, J. (2005). Hypothalamic regulation of sleep and circadian rhythms. Nature, 437, 1257–1263. Schenck, C. H., Hurwitz, T. D., & Mahowald, M. W. (1993). Symposium: Normal and abnormal REM sleep regulation: REM sleep behaviour disorder: An update on a series of 96 patients and a review of the world literature. Journal of Sleep Research, 2, 224–231. Schredl, M., Weber, B., Leins, M. L., & Heuser, I. (2001). Donepezil-induced REM sleep augmentation enhances memory performance in elderly, healthy persons. Experimental Gerontology, 36, 353–361. Sekula, L. K., Lucke, J. F., Heist, E. K., Czambel, R. K., & Rubin, R. T. (1997). Neuroendocrine aspects of primary endogenous depression: Mathematical modeling of nocturnal melatonin secretion in major depressives and normal controls. Psychiatry Research, 69, 143–153. Simic, G., Stanic, G., Mladinov, M., Jovanov-Milosevic, N., Kostovic, I., & Hof, P. R. (2009). Does Alzheimer’s disease

begin in the brainstem? Neuropathology and Applied Neurobiology, 35, 532–554. Singer, C., Tractenberg, R. E., Kaye, J., Schafer, K., Gamst, A., Grundman, M., et al. (2003). A multicenter, placebo-controlled trial of melatonin for sleep disturbance in Alzheimer’s disease. Sleep, 26, 893–901. Smith, C. (2001). Sleep states and memory processes in humans: Procedural versus declarative memory systems. Sleep Medicine Reviews, 5, 491–506. Spijker, A., Vernooij-Dassen, M., Vasse, E., Adang, E., Wollersheim, H., Grol, R., et al. (2008). Effectiveness of nonpharmacological interventions in delaying the institutionalization of patients with dementia: A meta-analysis. Journal of the American Geriatrics Society, 56, 1116–1128. Srinivasan, V., Smits, M., Spence, W., Lowe, A. D., Kayumov, L., Pandi-Perumal, S. R., et al. (2006). Melatonin in mood disorders. The World Journal of Biological Psychiatry, 7, 138–151. Stickgold, R. (2005). Sleep-dependent memory consolidation. Nature, 437, 1272–1278. Stopa, E. G., Volicer, L., Kuo-Leblanc, V., Harper, D., Lathi, D., Tate, B., et al. (1999). Pathologic evaluation of the human suprachiasmatic nucleus in severe dementia. Journal of Neuropathology and Experimental Neurology, 58, 29–39. Sunderajan, P., Gaynes, B. N., Wisniewski, S. R., Miyahara, S., Fava, M., Akingbala, F., et al. (2010). Insomnia in patients with depression: A STAR*D report. CNS Spectrums, 15, 394–404. Swaab, D. F., Grundke-Iqbal, I., Iqbal, K., Kremer, H. P., Ravid, R., & Van De Nes, J. A. (1992). Tau and ubiquitin in the human hypothalamus in aging and Alzheimer’s disease. Brain Research, 590, 239–249. Thannickal, T. C., Lai, Y. Y., & Siegel, J. M. (2007). Hypocretin (orexin) cell loss in Parkinson’s disease. Brain, 130, 1586–1595. Touitou, Y. (2001). Human aging and melatonin. Clinical relevance. Experimental Gerontology, 36, 1083–1100. Troster, A. I. (2008). Neuropsychological characteristics of dementia with Lewy bodies and Parkinson’s disease with dementia: Differentiation, early detection, and implications for "mild cognitive impairment" and biomarkers. Neuropsychology Review, 18, 103–119. Tyne, H. L., Taylor, J., Baker, G. A., & Steiger, M. J. (2010). Modafinil for Parkinson’s disease fatigue. Journal of Neurology, 257, 452–456. Uzun, S., Kozumplik, O., Jakovljevic, M., & Sedic, B. (2010). Side effects of treatment with benzodiazepines. Psychiatria Danubina, 22, 90–93. Van Coevorden, A., Mockel, J., Laurent, E., Kerkhofs, M., L’hermite-Baleriaux, M., Decoster, C., et al. (1991). Neuroendocrine rhythms and sleep in aging men. American Journal of Physiology. Endocrinology and Metabolism, 260, E651.

52 Van der Werf, Y. D., Altena, E., Schoonheim, M. M., SanzArigita, E. J., Vis, J. C., De Rijke, W., et al. (2009). Sleep benefits subsequent hippocampal functioning. Nature Neuroscience, 12, 122–123. Van Someren, E. J. W. (2000). Circadian and sleep disturbances in the elderly. Experimental Gerontology, 35, 1229–1237. Vendette, M., Gagnon, J. F., Decary, A., Massicotte-Marquez, J., Postuma, R. B., Doyon, J., et al. (2007). REM sleep behavior disorder predicts cognitive impairment in Parkinson disease without dementia. Neurology, 69, 1843–1849. Verbaan, D., Van Rooden, S. M., Visser, M., Marinus, J., & Van Hilten, J. J. (2008). Nighttime sleep problems and daytime sleepiness in Parkinson’s disease. Movement Disorders, 23, 35–41. Walker, M. P. (2010). Neuropsychology and neuroimaging of sleep-dependent memory processing: Implications for aging. In S. R. Pandi-Perumal, J. M. Monti & A. A. Monjan (Eds.), Principles and practice of geriatric sleep medicine. United Kingdom: Cambridge University Press. Walker, M. P., Stickgold, R., Alsop, D., Gaab, N., & Schlaug, G. (2005). Sleep-dependent motor memory plasticity in the human brain. Neuroscience, 133, 911–917. Wehr, T. A., Rosenthal, N. E., Sack, D. A., & Gillin, J. C. (1985). Antidepressant effects of sleep deprivation in bright and dim light. Acta Psychiatrica Scandinavica, 72, 161–165. Wider, B., & Pittler, M. H. (2010). Complementary and alternative medicine for sleep disturbance in the elderly. In S. R. Pandi-Perumal, J. M. Monti & A. A. Monjan (Eds.), Principles and practice of geriatric sleep medicine. United Kingdom: Cambridge University Press. Willis, G. L., & Turner, E. J. (2007). Primary and secondary features of Parkinson’s disease improve with strategic exposure to bright light: A case series study. Chronobiology International, 24, 521–537. Wilson, M. A., & Mcnaughton, B. L. (1994). Reactivation of hippocampal ensemble memories during sleep. Science, 265, 676–679. Wong, K. K., Marshall, N. S., Grunstein, R. R., Dodd, M. J., & Rogers, N. L. (2008). Comparing the neurocognitive effects of 40 h sustained wakefulness in patients with untreated OSA and healthy controls. Journal of Sleep Research, 17, 322–330.

Wright, K. J., & Rogers, N. (2008). Endogenous versus exogenous effects of melatonin. In S. Pandi-Perumal & D. Cardinali (Eds.), Melatonin: From molecules to therapy. New York: Nova Science Publishers Inc. Wu, Z. L., Ciallella, J. R., Flood, D. G., O’kane, T. M., Bozyczko-Coyne, D., & Savage, M. J. (2006). Comparative analysis of cortical gene expression in mouse models of Alzheimer’s disease. Neurobiology of Aging, 27, 377–386. Wu, Y. H., Feenstra, M. G., Zhou, J. N., Liu, R. Y., Torano, J. S., Van Kan, H. J., et al. (2003). Molecular changes underlying reduced pineal melatonin levels in Alzheimer disease: Alterations in preclinical and clinical stages. The Journal of Clinical Endocrinology and Metabolism, 88, 5898–5906. Wu, Y. H., Fischer, D. F., Kalsbeek, A., Garidou-Boof, M. L., Van Der Vliet, J., Van Heijningen, C., et al. (2006). Pineal clock gene oscillation is disturbed in Alzheimer’s disease, due to functional disconnection from the "master clock". The FASEB Journal, 20, 1874–1876. Wu, Y. H., & Swaab, D. F. (2007). Disturbance and strategies for reactivation of the circadian rhythm system in aging and Alzheimer’s disease. Sleep Medicine, 8, 623–636. Wulff, K., Gatti, S., Wettstein, J. G., & Foster, R. G. (2010). Sleep and circadian rhythm disruption in psychiatric and neurodegenerative disease. Nature Reviews Neuroscience, 11, 589–599. Yoo, S. S., Hu, P. T., Gujar, N., Jolesz, F. A., & Walker, M. P. (2007). A deficit in the ability to form new human memories without sleep. Nature Neuroscience, 10, 385–392. Zaccai, J., Mccracken, C., & Brayne, C. (2005). A systematic review of prevalence and incidence studies of dementia with Lewy bodies. Age and Ageing, 34, 561–566. Zeitzer, J. M., Daniels, J. E., Duffy, J. F., Klerman, E. B., Shanahan, T. L., Dijk, D. J., et al. (1999). Do plasma melatonin concentrations decline with age? The American Journal of Medicine, 107, 432. Zhu, L. Q., Wang, S. H., Ling, Z. Q., Wang, D. L., & Wang, J. Z. (2004). Effect of inhibiting melatonin biosynthesis on spatial memory retention and tau phosphorylation in rat. Journal of Pineal Research, 37, 71–77.

H. P. A. Van Dongen and G. A. Kerkhof (Eds.) Progress in Brain Research, Vol. 190 ISSN: 0079-6123 Copyright Ó 2011 Elsevier B.V. All rights reserved.

CHAPTER 3

Cognition and daytime functioning in sleep-related breathing disorders Melinda L. Jackson{,*, Mark E. Howard{ and Maree Barnes{ {

Sleep and Performance Research Center, Washington State University, Spokane, WA, USA Institute for Breathing and Sleep, Austin Health, Heidelberg, Melbourne, VIC, Australia

{

Abstract: Sleep-related breathing disorders encompass a range of disorders in which abnormal ventilation occurs during sleep as a result of partial or complete obstruction of the upper airway, altered respiratory drive, abnormal chest wall movement, or respiratory muscle function. The most common of these is obstructive sleep apnea (OSA), occurring in both adults and children, and causing significant cognitive and daytime dysfunction and reduced quality of life. OSA patients experience repetitive brief cessation of breathing throughout the night, which causes intermittent hypoxemia (reductions in hemoglobin oxygen levels) and fragmented sleep patterns. These nocturnal events result in excessive daytime sleepiness, and changes in mood and cognition. Chronic excessive sleepiness during the day is a common symptom of sleep-related breathing disorders, which is assessed in sleep clinics both subjectively (questionnaire) and objectively (sleep latency tests). Mood changes are often reported by patients, including irritability, fatigue, depression, and anxiety. A wide range of cognitive deficits have been identified in untreated OSA patients, from attentional and vigilance, to memory and executive functions, and more complex tasks such as simulated driving. These changes are reflected in patient reports of difficulty in concentrating, increased forgetfulness, an inability to make decisions, and falling asleep at the wheel of a motor vehicle. These cognitive changes can also have significant downstream effects on daily functioning. Moderate to severe cases of the disorder are at a higher risk of having a motor vehicle accident, and may also have difficulties at work or school. A number of comorbidities may also influence the cognitive changes in OSA patients, including hypertension, diabetes, and stroke. These diseases can cause changes to neural vasculature and result in neural damage, leading to cognitive impairments. Examination of OSA patients using neuroimaging techniques such as structural magnetic resonance imaging and proton magnetic resonance spectroscopy has observed significant changes to brain structure and metabolism. The downstream effects of neural,

*Corresponding author. Tel.: þ 1 509 358 5514; Fax: þ 1 509 358 7810 E-mail: [email protected] DOI: 10.1016/B978-0-444-53817-8.00003-7

53

54

cognitive, and daytime functional impairments can be significant if left untreated. A better understanding of the cognitive effects of these disorders, and development of more effective assessment tools for diagnosis, will aid early intervention and improve quality of life of the patient. Keywords: sleep apnea; mood; sleepiness; attention; memory; executive function.

Introduction Sleep-disordered breathing encompasses disorders in which abnormal ventilation occurs during sleep as a result of partial or complete obstruction of the upper airway, altered respiratory drive, abnormal chest wall movement, or respiratory muscle function. There is a range of severity, from primary snoring and upper airway resistance syndrome through to sleep apnea and obesity hypoventilation syndrome (OHS). Obstructive sleep apnea (OSA) is characterized by obstruction of the upper airway and the presence of continued respiratory effort with normal central nervous system drive for respiration during sleep (Patil et al., 2007). Central sleep apnea (CSA) is characterized by unstable ventilatory control with short periods of absent or reduced central nervous system drive to breath. This results in periods of reduced or absent respiratory muscle activity and hence reduced or absent respiration (Eckert et al., 2007). Obstructive and CSA may occur in the same individual (mixed apnea). CSA and mixed apnea, like OSA, also cause arousals from sleep and hypoxemia (decreased hemoglobin oxygen levels), resulting in impaired daytime function. In OHS, there is reduced ventilation during sleep and subsequently during wakefulness as a result of an external restrictive effect due to mass loading of the chest wall and abdomen, an increased incidence of peripheral airflow obstruction (Schachter et al., 2001) and narrowing of the upper airway, which is worsened during sleep. This causes persistent hypoxemia and hypercapnia (increased hemoglobin carbon dioxide levels) during sleep, which persists into wake. CSA makes up less than 10% of people presenting with sleep apnea, and CSA and OHS

are relatively understudied; therefore, this chapter primarily focuses on OSA. OSA is found in 24% adult males and 9% adult females, with 4% and 2%, respectively, being symptomatic of this disorder (Young et al., 1993). Children and adolescents may also be affected. Age is a risk factor, with the proportion of OSA patients averaging 30% after the age of 65. Anatomical upper airway abnormalities, such as a longer soft palate, occur more often in men, contributing to the higher incidence of OSA in the male population (Patil et al., 2007). Ethnicity also influences the risk of OSA (Kripke et al., 1997). The world wide increase in the prevalence of obesity (McLellan, 2002) is likely to result in a rise in the prevalence of OSA, as it is closely linked to the occurrence of this disease. The most common reason for an individual seeking a diagnosis is snoring, which is usually reported by the bed partner, often in association with the partner witnessing episodes of apnea during sleep. Patients suffering from OSA also may exhibit other nocturnal symptoms such as excessive sweating, motor restlessness, enuresis, and frequent awakenings with gasping or choking (Patil et al., 2007). There are also a number of daytime symptoms reported, such as excessive daytime sleepiness (EDS), difficulties concentrating, and changes in mood and cognition, which combine to reduce patients’ enjoyment of and participation in life (Barnes et al., 2004). Mechanisms of cognitive impairment in OSA What is unique about OSA compared to other sleep disorders is that at least two primary nocturnal physiological abnormalities occur, which may

55

and intermittent hypoxemia also occur in CSA and OHS, and are proposed causes of daytime dysfunction in untreated patients (Colt et al., 1991).

be the underlying cause of cognitive daytime impairments: hypoxemia and sleep fragmentation. Repetitive brief upper airway obstruction occurs during sleep, resulting in apneas (complete cessation of breathing) or hypopneas (a reduction in breathing; Fig. 1). The major diagnostic criterion for OSA is frequent periods of respiratory arrest of 10 s or more during sleep. The apnea/ hypopnea index (AHI; number of events per hour of sleep) must be at least five to make a diagnosis of OSA (AASM, 1999). Severe patients may experience apneas that last over 3 min, or occur hundreds of times per night. Apneas and hypopneas often terminate with arousal from sleep, with resultant fragmentation of sleep and reduction in total sleep time. Additionally, hypoxemia occurs during an apneic event, causing a disruption in the biochemical and hemodynamic state of the central nervous system (Patil et al., 2007). As discussed in the section Neural impairment, some studies support the hypothesis that hypoxemia ultimately results in structural neuronal damage. Patients diagnosed with severe OSA may record arterial hemoglobin oxygen saturation (SaO2) reductions to below 50% of preapneic levels. Multiple awakenings from sleep

Daytime function Sleepiness Chronic EDS is one of the hallmark features of OSA, CSA, and OHS, and appears to result predominantly from frequent arousal from sleep (Colt et al., 1991). Its clinical features are a strong feeling of abnormal daytime tiredness, and reduced wakefulness, and vigilance (Sauter et al., 2000). The association between sleepiness and sleep-related breathing disorders has been demonstrated using both subjective and objective assessment of sleepiness. Subjective sleepiness The Epworth Sleepiness Scale (ESS) is a simple, widely used subjective measure of sleepiness, which asks about the likelihood of falling asleep

100

SaO2 % 50

THOR x1

ABDO

Ob.A

Ob.A

Ob.A

Ob.A

Ob.A

Ob.A

Ob.A

Ob.A

Ob.A

Ob.A

Ob.A

Ob.A

x1

AIRFLOW x1

EEG/DiEMG 62.5 mV

SOUND x1

Fig. 1. Polysomnography trace of an individual with OSA. Blue squares indicate where an apnea has occurred. During an apneic event, there is a reduction or cessation of respiratory drive (in ABDO and air flow traces), which causes a reduction in oxygen saturations (SaO2). At the end of the apneic event, there is an arousal from sleep (in the EEG trace), and a gasping sound as respiratory drive begins. This cycle continues constantly across the entire night, causing frequent intermitted hypoxemia and arousals from sleep. EMG, electromyogram; EEG, electroencephalogram; SaO2, oxygen saturations; THOR, thoracic band; ABDO, abdominal band.

56

in a variety of situations (Johns, 1991). Average scores of 9.5, 11.5, and 16.0 were recorded in patients with mild, moderate, and severe OSA from a clinic population, compared to 5.9 in a control group (potential range 0–24; Johns, 1991). Moderate to severe sleepiness was found in a study of 54 patients with OHS, who had a mean ESS of 16 (Perez de Llano et al., 2005). Increased sleepiness on the ESS has also been found in general population studies of OSA. In the Sleep Heart Health study, the mean ESS scores were 7.2, 7.8, 8.3, and 9.3 for subjects with normal polysomnography, mild, moderate, and severe OSA, respectively (Gottlieb et al., 1999). The lesser sleepiness in the latter study may relate to the different study populations (clinic vs. general population). Other methods assessing the propensity to fall asleep in different situations have also identified increased sleepiness in OSA subjects. In the Wisconsin sleep cohort, subjects were described as excessively sleepy if they "woke unrefreshed regardless of how long they had slept," "felt excessively sleepy during the day," and had "uncontrollable daytime sleepiness that interfered with daytime living" at least 2 days a week. Sixteen percent of men with OSA had all three symptoms compared to only 3% of men without OSA (Young et al., 1993). The functional outcomes of sleep questionnaire (FOSQ) is a sleep-specific quality of life tool, with subscales related to activity level, vigilance, general productivity, social outcome, and intimate and sexual relationships (Weaver et al., 1997). Subjects with moderate OSA indicated significantly more difficulty with sleepiness on all subscales of the FOSQ and on the FOSQ global score. Increased sleepiness on both the ESS and FOSQ is related to elevated road crash risk (Howard et al., 2004), thus signifying real-world relevance.

Objective sleepiness Increased objectively measured sleepiness has also been demonstrated in subjects with OSA.

Current recommended indices of chronic objective sleepiness measure sleep latency (the time taken to fall asleep as measured by brain activity on electroencephalography (EEG)) while in the laboratory. In the multiple sleep latency test (MSLT), sleep latency is measured while subjects are lying down and told not to resist falling asleep. In the maintenance of wakefulness test (MWT), sleep latency is measured while subjects are sitting passively in a comfortable chair in a quiet dark room, but asked to try and remain awake. In standard testing, four sleep latency measurements are recorded at intervals during the same day and the average sleep latency is recorded. Sleep latency for both tests varies in relation to sleep apnea severity and sleep restriction (Banks et al., 2004; Chervin and Aldrich, 1998). In a group of patients with predominantly severe OSA, mean sleep latency was markedly reduced (2.6 min) compared to a matched control group (12.9 min) on the MSLT (Roth et al., 1980). Reduced sleep latency has also been observed in subjects with severe OSA using the MWT (Mazza et al., 2002). In patients with a range of severity of OSA, Chervin and Aldrich found that disease severity explained only 11% of the variance of the mean sleep latency, indicating that other factors have a significant influence on sleep latency (Chervin and Aldrich, 1998). Other studies, assessing the relationship between mild OSA and objective sleepiness, have found variable results. One study found no difference in mean sleep latency between a group of patients with mild OSA and a control group (Redline et al., 1997). However, a more recent study did find a weak relationship between sleep latency on the MWT and AHI in subjects with mild to moderate OSA (Banks et al., 2004). The inclusion of subjects with moderate OSA may have led to the finding in the latter study. Increased objective sleepiness has been demonstrated in professional drivers with OSA. Mean sleep latency on the MSLT was related to severity of OSA in a study of American truck drivers (Dinges, 1998). Mean sleep latency was 5.76 and

57

4.36 min in drivers with moderate or severe OSA, respectively, compared to 7.9 min in drivers without OSA. Drivers with mild OSA were not different from those without OSA. Hakkanen et al. found a shorter sleep latency and increased blink duration while driving in a group of bus drivers with mild OSA compared to a control group (Hakkanen et al., 1999). In summary, the majority of studies support a relationship between OSA and sleepiness, particularly for subjects with moderate to severe disease. Increased objective and subjective sleepiness has been demonstrated, including subjective measures that have been shown to relate to road crash risk. These changes have been identified in both clinical and general populations, with several studies also identifying increased sleepiness in transport drivers with sleep disorders.

Mood Psychological and personality alterations are commonly reported in OSA patients, and often first noted by their family members. These changes generally stem from fragmented sleep patterns and increased sleepiness levels. Patients with significant daytime sleepiness may feel unmotivated, lack energy, and report less enjoyment from daily activities. They may experience irritability, impatience, fatigue, moodiness, depression, anxiety and in more severe cases, psychosis, paranoia, and irrational behavior (Guilleminault et al., 1978). In children with OSA, mood changes may manifest as shyness and social withdrawal, aggressiveness, or hyperactivity which may reflect a profile of attention deficit hyperactivity disorder. In particular, there is considerable evidence suggesting a high prevalence of depression and anxiety symptomatology in OSA. The prevalence of depression in clinical samples of OSA is as high as 40% (Schroder and O'Hara, 2005); this is compared to general population estimates of 6% of males (American Psychological Association, 1994). Studies of

clinical samples have generally found higher rates of depression in OSA when compared to population surveys. Variation in the prevalence of depression in OSA may be due to different methods and measurements for scoring depression and different diagnostic criteria for selecting disease severity. Thus, it is difficult to compare these studies and draw direct conclusions as to the comorbid prevalence of these disorders. Another confounding factor is gender. Depression is more commonly reported in females in the general population, and therefore the rates of depression in OSA samples may depend on the proportion of women included. Correlational studies have reported mixed findings as to the underlying mechanisms of depression in OSA. Although the specific impact of hypoxia versus sleep fragmentation on depressive symptoms in OSA patients is unclear, it appears that daytime sleepiness resulting from fragmented sleep is the primary candidate. Interestingly, OSA patients with depression have been reported to have a higher rate of disordered breathing events compared to their nondepressed counterparts (Millman et al., 1989) suggesting a causal relationship. However, changes in mood have also been observed in primary snorers thus the relationship between disease severity is not clear cut (Aikens and Mendelson, 1999; Jackson et al., 2010). This suggests that other aspects of the disorder may also contribute to depression, such as fatigue (Bardwell et al., 2007). Reports of anxiety in OSA patients are less common than depression, but are nonetheless not unusual. As with depression, there is still debate as to the prevalence of anxiety in OSA patients, and its relationship to the nocturnal symptoms of OSA. The association between OSA and anxiety may be mediated by a patient's quality of life, rather than the direct influence of sleep variables (Sanchez et al., 2001). The relationship between mood and OSA is far from clear; it is yet to be resolved whether depression in OSA is a primary consequence, or if it occurs as a secondary disorder associated with

58

OSA-related symptoms (e.g., sleepiness, fatigue, and social withdrawal). Utilization of clinical scales of fatigue and depression should be included in the clinical interview for patients presenting to sleep physicians. Although some mood assessment tools have been validated for patient populations, none have been validated for use in OSA patients in whom many of the daytime features also overlap with depression symptoms (e.g., lethargy, lack of energy, and fatigue). Depressive symptoms may also impact upon compliance of treatment for OSA, such as continuous positive airway pressure (CPAP) or weight loss. These findings also have important clinical implications for mental health professionals reviewing patients with depression who may have a concomitant sleep disorder. Difficulties with sleep could exacerbate depressive symptoms, and potentially affect the efficacy of treatment.

Cognition Changes in cognition are one of the hallmark features of OSA. For those patients who misperceive their daytime sleepiness, cognitive complaints of impaired concentration and forgetfulness, as well as falling asleep at the wheel of a motor vehicle, may be the key initial signs that they have the disorder. Cognition in OSA patients has typically been examined in four clinically relevant domains: attention and vigilance, memory and learning, executive functions, and simulated driving. There have been a number of clinical population studies examining these areas of cognition and some theories and speculations of the underlying mechanisms of these deficits have been proposed, including sleep fragmentation and recurrent cyclical nocturnal hypoxia–reoxygenation.

Attention and vigilance Attention is a multifaceted domain, which involves focused attention (or concentration),

divided attention, sustained attention (or vigilance), and alertness. An inability to maintain concentration and attention is a commonly reported symptom of untreated OSA patients, which can have significant detrimental consequences for occupational performance, motor vehicle safety, and daily functioning. As a result, attention is one of the most researched cognitive domains in this population (Ayalon et al., 2009; Bedard et al., 1991a; Redline et al., 1997). Deficits in vigilance, attention, and psychomotor speed in untreated patients appear to be the most consistently reported cognitive domains. Electrophysiological evidence supports the behavioral findings, demonstrating increased P300 latency of eventrelated potentials—an index or attentional allocation and processing speed—in OSA patients (Kotterba et al., 1998). Performance on attentional tasks in OSA patients compared to a control group appears to reflect a dose–response relationship with disease severity (as measured by the AHI), with more severe OSA patients displaying poorer attention performance. For example, studies of mild OSA patients report minor to no deficits in attention performance (Redline et al., 1997), whereas studies examining sustained attention performance in severe OSA patients using the psychomotor vigilance test (PVT) and four choice reaction time task report deficits in this particular attentional domain (Barbe et al., 1998; Bedard et al., 1991b). Studies examining focused and divided attention performance in moderate OSA patients have yielded mixed results; significant differences between patients and controls have been reported in some studies (Greenberg et al., 1987; Naegele et al., 1995) but not others (Lee et al., 1999). Most studies of moderate OSA patients, however, do find impaired performance on at least one measure of attention (Greenberg et al., 1987; Kotterba et al., 1998; Naegele et al., 1995). However, since the attentional deficits are not common across all patients, it is possible that impairments in attention may occur in some, but not all patients with moderate forms of the disorder.

59

The underlying mechanisms of attentional deficits in OSA patients have primarily been linked to EDS associated with sleep disruption and arousals (Ayalon et al., 2009; Naismith et al., 2004). Evidence from functional magnetic resonance imaging (MRI) also supports this contention—OSA patients have reduced brain activation while performing the PVT compared to healthy controls, and this activation pattern is related to the number of arousals during sleep and slower reaction times (Ayalon et al., 2009). However, hypoxia may also play a role in attention deficits to some degree (Bedard et al., 1991a; Greenberg et al., 1987). Greenberg assessed neuropsychological function in a group of subjects with sleep-disordered breathing, a matched group with excessive sleepiness but without sleep-disordered breathing and a control group (Greenberg et al., 1987). The sleep-disordered breathing group performed worse on most tasks compared to both other groups and the degree of hypoxemia was related to motor performance. Deficits in vigilance and attention may also mediate performance impairments in other cognitive domains. For example, reduced vigilance has been shown to contribute significantly to the verbal memory impairment in untreated patients (Naegele et al., 1995).

Memory and learning Memory is a complex cognitive function, aspects of which have been shown to be impaired in OSA patients. Short-term memory is usually described in terms of visual or verbal memory. Long-term memory can be divided into two components—procedural memory and declarative memory. The former refers to unconscious memories of how to do things (e.g., tie a shoe lace). In contrast, declarative memory refers to memories that are consciously recalled, either facts (semantic memory, e.g., the names of different types of roses) or events (episodic memory, e.g., the first day of school). Thus, episodic

memory involves recall of personal events whereas semantic memory involves recollection of facts that may have been learnt at that event. OSA patients have deficits in memory processes, reflected in clinical complaints such as forgetting names and phone numbers, and learning difficulties are seen in children with OSA. Despite extensive research in this area, it remains unclear which aspects of memory are affected by the disorder. Mild impairments in short-term verbal (Naegele et al., 1995, 2006; Twigg et al., 2010) and visual (Ferini-Strambi et al., 2003; Naegele et al., 1995) memory have been observed in some studies but not others (Greenberg et al., 1987; Kim et al., 1997). Longterm semantic memory impairment has also been reported (Ferini-Strambi et al., 2003; Salorio et al., 2002), but again the findings are not consistent on all tests (Bedard et al., 1991b; Greenberg et al., 1987; Naegele et al., 1995). Where procedural memory has been investigated, this too has produced inconsistent results (Naegele et al., 2006; Rouleau et al., 2002). Taken together, these studies suggest that memory impairments are mild, and do not affect all aspects of memory. Discrepancies in the field with regard to the specific memory impairments associated with OSA may be due to the influence of other cognitive factors such as vigilance on memory performance. Recent studies that have utilized extensive memory test batteries, while controlling for vigilance and sleepiness, have reported that verbal episodic memory is the primary memory deficit seen in OSA patients (Naegele et al., 2006; Twigg et al., 2010). This suggests that memory deficits are related to impairment of retrieval and not of learning or recognition. Thus, OSA patients have a reduced capacity to acquire new information but no difficulty in retaining previously learned memories. Such memory impairments may affect a patients’ ability to take in important information, for example, relating to the correct use of their treatment. Since visual memory seems to be preserved, the use of visual information in training OSA patients in the use of their treatment may be more effective.

60

Unlike attentional impairments, which stem from sleep fragmentation and daytime sleepiness, memory impairments appear to be related to intermittent hypoxemia. The known vulnerability of the hippocampus—a region associated with memory processing—to intermittent hypoxia (Gozal et al., 2001) has led to speculation that neural cell loss associated with intermittent hypoxemia may underlie aspects of memory impairment (this is further discussed in the section Neural impairment). Establishing a link between aspects of memory impairment and any single factor, such as intermittent hypoxemia, is nonetheless hampered by the corelationship of indices such as sleep fragmentation and daytime sleepiness that are also likely to contribute to memory deficits in OSA patients. Memory impairment is not always a common presenting complaint of patients with OSA; the decline is often gradual in onset and deterioration is slow, so compensatory mechanisms may mask the memory loss. For example, OSA patients with intact verbal encoding ability display increased activation in prefrontal, temporal, and parietal regions and the cerebellum on functional MRI (Ayalon et al., 2006). Neuroimaging studies have also revealed that during encoding of episodic information, healthy sleep-deprived subjects display significant reduction in activation of the hippocampus, associated with reduced encoding ability (Yoo et al., 2007). The findings suggest that adequate sleep prior to learning is needed for the proper functioning of brain regions involved in encoding new memories. It could be argued that this response may also occur in OSA patients who experience fragmented sleep patterns. Interestingly in healthy subjects, negative emotional stimuli seem to be relatively resistant to encoding impairment from prior sleep loss, as compared to positive stimuli (Walker and Stickgold, 2006). As a consequence, sleep insufficiency may produce a negative learning bias, which may explain the higher rates of comorbid mood disorder in OSA patients (Benca et al., 1992).

Executive functions Executive functioning is a loosely defined collection of brain processes responsible for planning, problem-solving, mental flexibility, abstract thinking, rule acquisition, initiating appropriate actions, inhibiting inappropriate actions, and selecting relevant sensory information. Working memory is the temporary storage and manipulation of information, and is also part of the executive functioning of the brain. By the application of these processes, people are able to function and react appropriately in novel and changing situations, selecting previously acquired information, and manipulating, modifying, and applying it to a new environment. OSA patients may report decision-making difficulties, increased errors, and poor judgment, which can potentially affect their job performance or educational pursuits. Other aspects of executive functioning performance, such as set shifting, initiating new responses, planning and inhibition of automatic responses have also been reported in untreated sleep clinic populations (Bedard et al., 1991b; Naegele et al., 1995). Some reports of working memory impairment in OSA patients include maintenance and manipulation of information (Naegele et al., 2006; Redline et al., 1997) but not dual task performance (Naegele et al., 2006). Inconsistency in working memory impairment between some studies (e.g. Naegele et al., 2006; Twigg et al., 2010) may be due to the influence of vigilance and concentration impairments on working memory performance. The disruption in executive function seen in both children and adults with OSA has been proposed to be due to a combination of sleep deprivation due to sleep fragmentation and irreversible damage due to the hypoxemia–reoxygenation cycles (Bedard et al., 1991a; Naegele et al., 1995). The pattern of deficits observed in OSA is consistent with damage to the prefrontal and frontal lobe areas of the brain. An elegant model linking upper airway collapse to daytime functional impairment has been

61

proposed with the anatomical locus for the decrement in executive function being the prefrontal cortex (Beebe and Gozal, 2002). Prolonged latencies of the P300 component of event-related potentials, which are not reversed with CPAP treatment, have also been demonstrated in OSA patients (Kotterba et al., 1998). The P300 component is believed to be generated by subcortical structures and the prefrontal cortex; thus, abnormalities in this brain response may reflect damage to these regions in OSA patients. Support for this model also arises from functional MRI studies, which have reported impaired performance on executive type functions (response inhibition and working memory) is associated with decreased activation of the prefrontal cortex (Ayalon et al., 2009; Thomas et al., 2005).

Simulated and on-road driving Many patients with OSA have difficulty in driving, and occasionally report falling asleep at the wheel. Laboratory studies using simulated driving or tracking tasks confirm that patients with untreated or undiagnosed OSA have impaired driving performance (George et al., 1996). These impairments include a reduced ability to avoid obstacles; more errors for steering, signaling, braking, and accelerating; greater tracking error; and slower reaction times (George et al., 1996, Risser et al., 2000). These and other studies of driving simulation have involved patients with moderate to severe OSA compared to normal controls. Performance impairments have been shown to be worse than that of control subjects who had ingested alcohol (mean blood-alcohol concentration 0.09%; George et al., 1996). Other factors that impair driving performance appear to be more detrimental in OSA patients than in normal subjects. In a study that evaluated the effects of alcohol and sleep deprivation, OSA patients had greater deterioration in steering performance and more crashes than control subjects following these interventions (Vakulin et al.,

2007). Examination of the relationship between driving performance and OSA in professional drivers, a particularly high-risk population, found that driving impairment was related to severity of OSA, but performance deterioration was only evident for drivers with moderate to severe disease (Dinges, 1998). These findings suggest that untreated patients may be at a higher risk of having a motor vehicle accident compared to the general population. Indeed, epidemiological evidence supports this contention—people with OSA are at two to seven times increased risk of road accidents (TeranSantos et al., 1999; Young et al., 1997) and a two to three times increased risk of industrial accidents (Lindberg et al., 2001) as a result of sleepiness and altered cognitive function. A recent meta-analysis found an increased accident rate of 2.7 in those with OSA (Tregear et al., 2009). It has been estimated that 800,000 OSArelated motor vehicle accidents occur annually in the United States at a cost of U.S. $15.9 billion and 1400 lives lost (Sassani et al., 2004). In middle aged (30–60 years) drivers, OSA is likely to be a prominent cause of sleepiness-related crashes. Findley et al. (1988) published the first controlled data demonstrating the relationship between OSA and crash risk (Findley et al., 1988), comparing state accident records from 29 patients with OSA to those of matched control subjects and the average accident statistics for the state of Virginia (Table 1). There was a sevenfold increased accident rate in patients compared to controls and two and a half times increase in the rate compared to the state average. Three subsequent case–control studies found similar associations between sleep-disordered breathing and accidents (Barbe et al., 1998; George and Smiley, 1999; Teran-Santos et al., 1999). One large prospective, cohort study also supports the association between sleep-disordered breathing and accident risk found in the case–control studies (Young et al., 1997). This study was performed in a general population of employed

62 Table 1. Sleep-disordered breathing and accident risk Author/Year

Study design

Population

Accident risk

Findley et al. (1988) Aldrich (1989) Wu et al. (1996) Young et al. (1997) Barbe et al. (1998) Teran-Santos et al. (1999) Horstmann et al. (2000) George (2001) Shiomi et al. (2002) Howard et al. (2004) Tregear et al. (2009)

Case–control Case–control Cross-sectional Prospective cohort Case–control Case–control Case–control Case–control Case–control Cross-sectional Meta-analysis

Sleep clinic Sleep clinic Sleep clinic General population Sleep clinic Drivers from traffic accidents Sleep clinic Sleep clinic Sleep clinic Commercial drivers –

7.0 1.0 3.0 4.2 2.3 6.3 8.7 3.0 2.3 1.3 2.7

RR OR OR OR OR OR RR RR RR OR RR

RR, rate ratio; OR, odds ratio.

adults. The odds ratio for having an accident was 4.2 in men with mild sleep-disordered breathing and 3.4 for those with moderate to severe disease, although the difference between the groups was not statistically different. However, this suggests that while the presence of sleep apnea is related to accident risk, increasing severity of sleep apnea (as measured by the AHI) is not predictive of crash risk. Other factors, such as the degree of associated sleepiness or cognitive impairment may mediate the increase in risk. Although there was no increased risk of having an accident in women with sleep-disordered breathing, there was a marked increased risk of having multiple accidents in both women and men with sleepdisordered breathing, which has also been observed in other studies (Barbe et al., 1998). Most studies have not found a relationship between the severity of sleep-disordered breathing and accident risk (Teran-Santos et al., 1999; Young et al., 1997). However, a few studies have suggested that it is predominantly patients with severe OSA patients who have an increased accident risk. George found that the average accident rate was 0.12 per year in those with severe disease compared to 0.07 per year in control subjects (George and Smiley, 1999). There was only a minor increase in accident rate (0.08) in patients with lesser degrees of sleep-disordered breathing.

In summary, moderate to severe OSA is associated with impaired vigilance and driving performance. The degree of impairment is similar to that at high blood-alcohol levels, and is associated with increased road accident risk. Studies in both clinical populations with OSA and subjects with OSA from general populations have shown an increased risk of having a road accident, which improves following treatment (George, 2001; refer to Chapter 4 for more details). Patients are also more likely to have multiple accidents. While some studies have found a dose relationship between severity of sleepdisordered breathing and accident risk, this is not a universal finding. Other factors associated with cognitive decline in sleep-related breathing disorders Comorbidities There are several conditions that are strongly associated with OSA and OHS that may also influence cognitive function, including obesity, diabetes mellitus, hypertension, and cardiovascular disease including stroke (Hajjar et al., 2010; Shahar et al., 2001). This raises the possibility that some of the cognitive impairment evident in OSA

63

is due to these confounding comorbidities. However, OSA may in fact be causal in some of these conditions by contributing to oxidative stress, systemic inflammation, and sympathetic activation through intermittent hypoxia and sleep fragmentation (Pack and Gislason, 2009). Recent cross-sectional data suggest that endothelial dysfunction and elevation of inflammatory biomarkers are greater in OSA than controls matched for obesity, and even higher in OHS (Punjabi and Beamer, 2007). These abnormalities are related to the presence of nocturnal hypoxia and may explain the emergence and progression of atherosclerosis, hypertension, and other cardiovascular and metabolic disorders in these conditions (Assmann et al., 2002). There is a reduction in the normal nocturnal dip in blood pressure in OSA patients (Suzuki et al., 1996), and recently this has been associated with brain atrophy (Hajjar et al., 2010). In summary, there are several comorbidities related to OSA and OHS that may contribute to cognitive impairment. Conversely, OSA and OHS may also increase the risk of some of these conditions, causing a vicious cycle of poor health outcomes resulting from the disorder, leading to unhealthy behaviors, which lead to worsening of symptoms (Spruyt et al., 2010).

Neural impairment The cognitive and functional impairment observed in OSA patients suggests that there may be some underlying changes in brain structure associated with the disease. In a model of OSA in rats, intermittent hypoxemia caused increased neuronal death in the hippocampus, and this was associated with impaired learning and memory in aging rats (Gozal et al., 2001). This has led to speculation that neural cell loss associated with intermittent hypoxia may underlie aspects of memory impairment in OSA in humans. Deficits in attention, vigilance, and verbal episodic memory in hypoxic, compared to nonhypoxic patients, support this contention (Findley et al., 1986).

In the past decade, a number of studies have emerged describing neural cell loss and changes in neurochemical levels of untreated OSA patients, although there are inconsistencies in the findings between studies. Some studies report quantitative regional gray matter loss in the hippocampus (Morrell et al., 2003), cerebellum, frontal and parietal cortex, and the anterior cingulate gyrus (Macey et al., 2002; Yaouhi et al., 2009), whereas others report no difference in structural volumes (O'Donoghue et al., 2005). These studies had small sample sizes, included patients with known comorbidities, and used different statistical thresholds which may explain the discrepancies in findings between studies. A recent study attempted to overcome these issues by examining a relatively large sample of OSA patients and controls, using standardized improved methods of structural analysis (Morrell et al., 2010). OSA patients, relative to controls, had significant reductions in the right middle temporal gyrus and left cerebellum—brain regions involved in the motor regulation of the upper airways, as well as in cognitive processing (Fig. 2). Studies utilizing magnetic resonance spectroscopy (MRS), a neuroimaging technique useful for measuring brain metabolism, in patients with untreated OSA have found metabolic abnormalities in frontal white matter, associated with the severity of OSA (AHI; Kamba et al., 2001) and cognitive deficits (Bartlett et al., 2004). These abnormalities appear to occur before any structural changes are evident, and are reversible with treatment. For instance, children with OSA have no sign of structural neuronal damage, but have reduced neural metabolites in hippocampus and right frontal cortex, and exhibit impairments in memory, learning, and executive functions (Halbower et al., 2006). Conclusion The influence of sleep-related breathing disorders on cognition and daytime functioning is complex,

64 (a)

5 4 3 2 1 0

(b)

4 3

SPM{T101}

2 1 0

Fig. 2. (a) Images show reductions in gray matter in right middle temporal gyrus of OSA patients compared with controls. Left panel shows “glass brain” view, and right panel shows gray matter loss superimposed upon an MR template. Views, in neurological orientation (left is left), from top left clockwise are, sagittal, coronal, and transverse. The voxel of maximum significance is marked as arrow head in the left panel. On the right panel, the cross hairs (t ¼ 4.05—indicated by colored bar) are located on the right at x ¼ 52, y ¼ 4, z ¼ 22 mm (Montreal Neurological Institute (MNI) coordinates relative to anterior commissure). The images are thresholded to include clusters that survived the topological false detection rate threshold of p < 0.05. (b) Images show statistically significant reductions in cerebellar gray matter in a subset of OSA patients compared with controls using a spatially unbiased infratentorial template (SUIT). Left panel shows “glass brain” view, and right panel shows gray matter loss superimposed upon the SUIT template. The image is displayed with a threshold of p < 0.05 uncorrected. The voxel of maximum significance survives correction at p < 0.05 FWE corrected (t ¼ 4.62—indicated by colored bar) and is located on the left at x ¼ 14, y ¼ 51, z ¼ 48 mm (MNI coordinates relative to anterior commissure). Two further maxima were located on the inferior edge of the cerebellum (t ¼ 4.18, MNI: 50, 58, 55; t ¼ 3.99, MNI:  35, 53, 62). Note: the location of the cerebellum and the associated gray matter reduction extends below the standard glass brain reference grid. Adapted from Morrell et al. (2010).

involving a number of overlapping physiological and psychological processes. The downstream effects of daytime functional impairments due to

vigilance and other cognitive impairment on-road and occupational accidents can be significant if the disorder is left untreated.

65

The conflicting results of studies of neurocognitive decrements in OSA are in part related to the vast array of tests used. A number of neurocognitive test batteries designed specifically for OSA patients have been proposed, which would improve interpretation of the results of such studies. There is also the issue of the validity of the neurocognitive tasks and mood assessment tools used in OSA patient studies, since most of these were originally developed for head trauma patients. Research in this area would also benefit from standardized research diagnostic criteria. Normal cognitive decline in older OSA patients may mask or overwhelm decline that is due to sleep apnea (Mathieu et al., 2008). Further to this, when testing higher cognitive functions such as memory and executive functions, it may be difficult to delineate between the impairment due to reduced vigilance, sleepiness, or poor motivation. It is therefore essential that all these elements are adequately measured and confounding factors are accounted for, including age, educational level, intelligence, and other medical illnesses may contribute to neurocognitive deficits. With the obesity epidemic growing throughout the world (McLellan, 2002), the prevalence of OSA is likely to increase, and with it a significant public health burden. A better understanding of the sequelae of OSA, as well as improved screening and diagnostic tools for assessment of these daytime and cognitive impairments, will facilitate early intervention and treatment to prevent permanent neurological damage, thereby reducing the public health and economic consequences of this disorder. References American Academy of Sleep Medicine Task Force (AASM) (1999). Sleep-related breathing disorders in adults: Recommendations for syndrome definition and measurement techniques in clinical research. Sleep, 22, 667–689. Aikens, J., & Mendelson, W. (1999). A matched comparison of mmpi responses in patients with primary snoring or obstructive sleep apnea. Sleep, 22, 355–359.

Aldrich, M. (1989). Automobile accidents in patients with sleep disorders. Sleep, 12, 487–494. American Psychological Association. (1994). Diagnostic and statistical manual of mental disorders. Washington, DC: American Psychiatric Association. Assmann, G., Cullen, P., & Schulte, H. (2002). Simple scoring scheme for calculating the risk of acute coronary events based on the 10-year follow-up of the prospective cardiovascular munster (procam) study. Circulation, 105, 310–315. Ayalon, L. F., Ancoli-Israel, S., Aka, A. F., McKenna, B., & Drummond, S. P. (2009). Relationship between obstructive sleep apnea severity and brain activation during a sustained attention task. Sleep, 32, 373–381. 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. Neuroimage, 31, 1817–1825. Banks, S., Barnes, M., Tarquinio, N., Pierce, R. J., Lack, L. C., & Doug McEvoy, R. (2004). Factors associated with maintenance of wakefulness test mean sleep latency in patients with mild to moderate obstructive sleep apnoea and normal subjects. Journal of Sleep Research, 13, 71–78. Barbe, F., Pericas, J., Munoz, A., Findley, L., Anto, J. M., & Agusti, A. G. (1998). Automobile accidents in patients with sleep apnea syndrome. An epidemiological and mechanistic study. American Journal of Respiratory and Critical Care Medicine, 158, 18–22. Bardwell, W. A., Ancoli-Israel, S., & Dimsdale, J. E. (2007). 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, 97, 181–186. Barnes, M., McEvoy, R. D., Banks, S., Tarquinio, N., Murray, C. G., Vowles, N., et al. (2004). Efficacy of positive airways pressure and oral appliance in mild to moderate obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine, 170, 656–664. 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. Bedard, M. A., Montplaisir, J., Richer, F., & Malo, J. (1991). Nocturnal hypoxemia as a determinant of vigilance impairment in sleep apnea syndrome. Chest, 100, 367–370. Bedard, M. A., Montplaisir, J., Richer, F., Rouleau, I., & Malo, J. (1991). Obstructive sleep apnea syndrome: Pathogenesis of neuropsychological deficits. Journal of Clinical and Experimental Neuropsychology, 13, 950–964. 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.

66 Benca, R. M., Obermeyer, W. H., Thisted, R. A., & Gillin, J. C. (1992). Sleep and psychiatric disorders: A meta-analysis. Archives of General Psychiatry, 49, 651–668. Chervin, R. D., & Aldrich, M. S. (1998). The relation between multiple sleep latency test findings and the frequency of apneic events in rem and non-rem sleep. Chest, 113, 980–984. Colt, H. G., Haas, H., & Rich, G. B. (1991). Hypoxemia vs sleep fragmentation as cause of excessive daytime sleepiness in obstructive sleep apnea. Chest, 100, 1542–1548. Dinges, D. (1998). Sleepiness and neurobehavioral functioning in relation to apnea severity in a cohort of commercial motor vehicle operators. Sleep, 21, 83. Eckert, D., Jordan, A., Merchia, P., & Malhotra, A. (2007). Central sleep apnea: Pathophysiology and treatment. Chest, 131, 595–607. Ferini-Strambi, L., Baietto, C., Di Gioia, 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. Findley, L., Barth, J., Powers, D., Wilhoit, S., Boyd, D., & Suratt, P. (1986). Cognitive impairment in patients with obstructive sleep apnea and associated hypoxemia. Chest, 90, 686–690. Findley, L. J., Unverzagt, M. E., & Suratt, P. M. (1988). Automobile accidents involving patients with obstructive sleep apnea. The American Review of Respiratory Disease, 138, 337–340. George, C. F. (2001). Reduction in motor vehicle collisions following treatment of sleep apnoea with nasal CPAP. Thorax, 56, 508–512. George, C. F., Boudreau, A. C., & Smiley, A. (1996). Simulated driving performance in patients with obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine, 154, 175–181. George, C., & Smiley, A. (1999). Sleep apnea and automobile accidents. Sleep, 22, 790–795. Gottlieb, D. J., Whitney, C. W., Bonekat, W. H., Iber, C., James, G. D., 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., Daniel, J. M., & Dohanich, G. P. (2001). Behavioral and anatomical correlates of chronic episodic hypoxia during sleep in the rat. The Journal of Neuroscience, 21, 2442–2450. Greenberg, G. D., Watson, R. K., & Deptula, D. (1987). Neuropsychological dysfunction in sleep apnea. Sleep, 10, 254–262. Guilleminault, C., Van Den Hoed, J., & Mitler, M. (1978). Clinical overview of the sleep apnea syndrome. In C. Guilleminult & W. Dement (Eds.), Sleep apnea syndromes (pp. 1–12). New York: Alan R. Liss. Hajjar, I., Zhao, P., Alsop, D., & Al, E. (2010). Association of blood pressure elevation and nocturnal dipping with brain

atrophy, perfusion and functional measures in stroke and nonstroke individuals. American Journal of Hypertension, 23, 17–23. Hakkanen, H., Summala, H., Partinen, M., Tiihonen, M., & Silvo, J. (1999). Blink duration as an indicator of driver sleepiness in professional bus drivers. Sleep, 22, 798–802. Halbower, A. C., Degaonkar, M., Barker, P. B., Earley, C. J., Marcus, C. L., Smith, P. L., et al. (2006). Childhood obstructive sleep apnea associates with neuropsychological deficits and neuronal brain injury. PLoS Medicine, 3, e301. Horstmann, S., Hess, C. W., Bassetti, C., Gugger, M., & Mathis, J. (2000). Sleepiness-related accidents in sleep apnea patients. Sleep, 23, 383–389. Howard, M. E., Desai, A. V., Grunstein, R. R., Hukins, C., Armstrong, J. G., Joffe, D., et al. (2004). Sleepiness, sleepdisordered breathing, and accident risk factors in commercial vehicle drivers. American Journal of Respiratory and Critical Care Medicine, 170, 1014–1021. Jackson, M., Stough, C., Howard, M., Spong, J., Downey, L., & Thompson, B. (2010). The contribution of fatigue and sleepiness to depression in patients attending the sleep laboratory for evaluation of obstructive sleep apnea. Sleep and Breathing, online first, 1–7. Johns, M. (1991). A new method for measuring daytime sleepiness: The Epworth Sleepiness Scale. Sleep, 14, 540–545. Kamba, M., Inoue, Y., Higami, S., Suto, Y., Ogawa, T., & Chen, W. (2001). Cerebral metabolic impairment in patients with obstructive sleep apnoea: An independent association of obstructive sleep apnoea with white matter change. Journal of Neurology, Neurosurgery and Psychiatry, 71, 334–339. Kim, H. C., Young, T., Matthews, C. G., Weber, S. M., Woodward, A. R., & Palta, M. (1997). Sleep-disordered breathing and neuropsychological deficits A populationbased study. American Journal of Respiratory and Critical Care Medicine, 156, 1813–1819. Kotterba, S., Rasche, K., Widdig, 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. Kripke, D. F., Ancoli-Israel, S., Klauber, M. R., Wingard, D. L., Mason, W. J., & Mullaney, D. J. (1997). Prevalence of sleep-disordered breathing in ages 40–64 years: A population-based survey. Sleep, 20, 65–76. Lee, M., Strauss, M., Adams, N., & Redline, S. (1999). Executive functions in persons with sleep apnea. Sleep and Breathing, 3, 13–16. Lindberg, E., Carter, N. G. T., & Janson, C. (2001). Role of snoring and daytime sleepiness in occupational accidents. American Journal of Respiratory and Critical Care Medicine, 164, 2031–2035. Macey, P. M., Henderson, L. A., Macey, K. E., Alger, J. R., Frysinger, R. C., Woo, M. A., et al. (2002). Brain morphology

67 associated with obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine, 166, 1382–1387. Mathieu, A., Mazza, S., Décary, A., Massicotte-Marquez, J., Petit, D., Gosselin, N., et al. (2008). Effects of obstructive sleep apnea on cognitive function: A comparison between younger and older OSAS patients. Sleep Medicine, 9, 112–120. Mazza, S., Pepin, J. L., Deschaux, C., Naegele, B., & Levy, P. (2002). Analysis of error profiles occurring during the Osler test: A sensitive mean of detecting fluctuations in vigilance in patients with obstructive sleep apnea syndrome. American Journal of Respiratory and Critical Care Medicine, 166, 474–478. McLellan, F. (2002). Obesity rising to alarming levels around the world. Lancet, 359, 1412. Millman, R., Fogel, B., Mcnamara, M., & Carlisle, C. (1989). Depression as a manifestation of obstructive sleep apnea: Reversal with nasal continuous positive airway pressure. The Journal of Clinical Psychiatry, 50, 348–351. Morrell, M. J., Jackson, M. L., Twigg, G. L., Ghiassi, R., McRobbie, D. W., Quest, R. A., et al. (2010). Changes in brain morphology in patients with obstructive sleep apnoea. Thorax, 65, 908–914. 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. Naegele, B., Launois, S., Mazza, S., Feuerstein, C., Pepin, J., & 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., Thouvard, V., Pepin, J., Levy, P., Bonnet, C., Perret, J., et al. (1995). Deficits of cognitive executive functions in patients with sleep apnea syndrome. Sleep, 18, 43–52. Naismith, S., Winter, V., Gotsopoulos, H., Hickie, I., & Cistulli, P. (2004). Neurobehavioral functioning in obstructive sleep apnea: Differential effects of sleep quality, hypoxemia and subjective sleepiness. Journal of Clinical and Experimental Neuropsychology, 26, 43–54. 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 and Critical Care Medicine, 171, 1185–1190. Pack, A., & Gislason, T. (2009). Obstructive sleep apnea and cardiovascular disease: A perspective and future directions. Progress in Cardiovascular Diseases, 5, 434–451. Patil, S., Schneider, H., Schwartz, A., & Smith, P. (2007). Adult obstructive sleep apnea: Pathophysiology and diagnosis. Chest, 132, 325–337. Perez De Llano, L., Golpe, R., Ortiz Piquer, M., & Veres Racamonde, A. (2005). Short-term and long-term effects of nasal intermittent positive pressure ventilation in patients with obesity-hypoventilation syndrome. Chest, 128, 587–594.

Punjabi, N., & Beamer, B. (2007). C-reactive protein is associated with sleep disordered breathing independent of adiposity. Sleep, 30, 29–34. Redline, S., Strauss, M. E., Adams, N., Winters, M., Roebuck, T., Spry, K., et al. (1997). Neuropsychological function in mild sleep-disordered breathing. Sleep, 20, 160–167. Risser, M. R., Ware, J. C., & Freeman, F. G. (2000). Driving simulation with EEG monitoring in normal and obstructive sleep apnea patients. Sleep, 23, 393–398. Roth, T., Hartse, K. M., Zorick, F., & Conway, W. (1980). Multiple naps and the evaluation of daytime sleepiness in patients with upper airway sleep apnea. Sleep, 3, 425–439. Rouleau, I., Décary, A., Chicoine, A., & Montplaisir, J. (2002). Procedural skill learning in obstructive sleep apnea syndrome. Sleep, 25, 404–411. 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, 93–100. Sanchez, A., Buela-Casal, G., Bermudez, M., & CasasMaldonado, F. (2001). The effects of continuous positive air pressure treatment on anxiety and depression levels in apnea patients. Psychiatry and Clinical Neurosciences, 55, 641–646. Sassani, A., Findley, L., Kryger, M. H., Goldlust, E., George, C., & Davidson, T. (2004). Reducing motor-vehicle collisions, costs, and fatalities by treating obstructive sleep apnea syndrome. Sleep, 27, 453–458. Sauter, C., Asenbaum, S., Popovic, R., Bauer, H., Lamm, C., Klösch, G., et al. (2000). Excessive daytime sleepiness in patients suffering from different levels of obstructive sleep apnoea syndrome. Journal of Sleep Research, 9, 293–301. Schachter, L., Salome, C., Peat, J., & Woolcock, A. (2001). Obesity is a risk for asthma and wheeze but not airway hyperresponsiveness. Thorax, 56, 4–8. Schroder, C., & O'Hara, R. (2005). Depression and obstructive sleep apnea (OSA). Annals of General Psychiatry, 4, 13. Shahar, E., Whitney, C. W., Redline, S., Lee, E. T., Newman, A. B., Javier Nieto, F., et al. (2001). Sleep-disordered breathing and cardiovascular disease: Cross-sectional results of the sleep heart health study. American Journal of Respiratory and Critical Care Medicine, 163, 19–25. Shiomi, T., Arita, A. T., Sasanabe, R., Banno, K., Yamakawa, H., Hasegawa, R., Ozeki, K., Okada, M., & Ito, A. (2002). Falling asleep while driving and automobile accidents among patients with obstructive sleep apneahypopnea syndrome. Psychiatry & Clinical Neurosciences, 56, 333–334. Spruyt, K., Capdevila, O. S., Serpero, L. D., KheirandishGozal, L., & Gozal, D. (2010). Dietary and physical activity patterns in children with obstructive sleep apnea. Journal of Pediatrics, 156, 724–730.

68 Suzuki, M., Guilleminault, C., Otsuka, K., & Shoimi, T. (1996). Blood pressure “dipping” and “non-dipping” in obstructive sleep apnea syndrome patients. Sleep, 19, 382–387. Teran-Santos, J., Jimenez-Gomez, A., & Cordero-Guevara, J. (1999). The association between sleep apnea and the risk of traffic accidents Cooperative group burgos-santander. The New England Journal of Medicine, 340, 847–851. Thomas, R., Rosen, B., Stern, C., Weiss, J., & Kwong, K. (2005). Functional imaging of working memory in obstructive sleep-disordered breathing. Journal of Applied Physiology, 98, 2226–2234. Tregear, S., Reston, J., Schoelles, K., & Phillips, B. (2009). Obstructive sleep apnea and risk of motor vehicle crash: Systematic review and meta-analysis. Journal of Clinical Sleep Medicine, 5, 573–581. Twigg, G. L., Papaioannou, I., Jackson, M., Ghiassi, R., Shaikh, Z., Jaye, J., et al. (2010). Obstructive sleep apnoea syndrome is associated with deficits in verbal but not visual memory. American Journal of Respiratory and Critical Care Medicine, 182, 98–103. Vakulin, A., Baulk, S. D., Catcheside, P. G., Anderson, R., Van Den Heuvel, C. J., Banks, S., et al. (2007). Effects of moderate sleep deprivation and low-dose alcohol on driving simulator performance and perception in young men. Sleep, 30, 1327–1333.

Walker, M., & Stickgold, R. (2006). Sleep, memory, and plasticity. Annal Review of Psychology, 57, 139–166. Weaver, T., Laizner, A., Evans, L., Maislin, G., Chugh, D., Lyon, K., et al. (1997). An instrument to measure functional status outcomes for disorders of excessive sleepiness. Sleep, 20, 835–843. Wu, H., & Yan-Go, F. (1996). Self-reported automobile accidents involving patients with obstructive sleep apnea. Neurology, 46, 1254–1257. Yaouhi, K., Bertran, F., Clochon, P., Mézenge, F., Denise, P., Foret, J., et al. (2009). A combined neuropsychological and brain imaging study of obstructive sleep apnea. Journal of Sleep Research, 18, 36–48. Yoo, S., Hu, P., Gujar, N., Jolesz, F., & Walker, M. (2007). A deficit in the ability to form new human memories without sleep. Nature Neuroscience, 10, 385–392. Young, T., Blustein, J., Finn, L., & Palta, M. (1997). Sleepdisordered breathing and motor vehicle accidents in a population-based sample of employed adults. Sleep, 20, 608–613. 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.

H. P. A. Van Dongen and G. A. Kerkhof (Eds.) Progress in Brain Research, Vol. 190 ISSN: 0079-6123 Copyright Ó 2011 Elsevier B.V. All rights reserved.

CHAPTER 4

Cognitive recovery following positive airway pressure (PAP) in sleep apnea Ellyn E. Matthews{ and Mark S. Aloia{,* {

{ College of Nursing, University of Colorado, Aurora, CO, USA Department of Medicine, National Jewish Health, Denver, CO, USA

Abstract: Obstructive sleep apnea–hypopnea syndrome (OSAHS) is a common sleep disorder that is characterized by repeated episodes of complete or partial cessation of breathing while sleeping. These recurrent breathing events result in fragmented sleep and recurrent hypoxemia. Distressing daytime sequelae reported by OSAHS patients include excessive daytime sleepiness, self-reported changes in mood, and cognitive problems. It has been well established that OSAHS can negatively impact functioning in multiple cognitive domains, such as attention and memory. In this chapter, neurobehavioral deficits in OSAHS are discussed, and proposed models of cognitive dysfunction are summarized. Current studies examining cognitive recovery with positive airway pressure treatment are presented. It appears that the cognitive dysfunction of OSAHS is not likely to be due to a single mediating mechanism, nor is it pervasive across all patients. Future research should attempt to identify these moderators for cognitive dysfunction in OSAHS and to highlight the mechanisms of dysfunction by cognitive domain. Keywords: sleep apnea; cognition; daytime sleepiness; neuropsychology; positive airway pressure.

Introduction

OSAHS is characterized by repeated complete (apnea) or partial (hypopnea) cessations of breathing most typically caused by a narrowing at various potential sites along the upper airway. During these breathing events, arterial oxygen saturation can drop to dangerously low levels (desaturation), resulting in increased respiratory effort and arousals from sleep to resume breathing. Recurrent hypoxemia, hypercapnia, and fragmented sleep are direct consequences of

Population-based epidemiologic studies have underscored the high prevalence of obstructive sleep apnea–hypopnea syndrome (OSAHS), occurring in 5% of the general population (Young et al., 2002). The pathophysiology of *Corresponding author. Tel.: þ 303 270 2386; Fax: þ 303 270 2115 E-mail: [email protected] DOI: 10.1016/B978-0-444-53817-8.00004-9

71

72

OSAHS that can impact neurobehavioral performance (Dempsey et al., 2010). The primary daytime sequelae of OSAHS include excessive daytime sleepiness, selfreported changes in mood, and cognitive problems. Decreased cognitive acuity can have significant negative consequences for occupational performance, driving safety, educational pursuits, and global functioning. For example, the risk of being involved in a motor vehicle accident is substantially higher as the severity of OSAHS increases (Tregear et al., 2009). Moreover, OSAHS has been associated with an increased risk of comorbid medical illnesses, particularly vascular diseases, such as hypertension, heart disease, and stroke (Dempsey et al., 2010). Psychiatric disorders including depression and anxiety have also been linked to OSAHS (Saunamaki and Jehkonen, 2007) with accompanying functional consequences that can be severe. OSAHS treatment Although surgical and behavioral (e.g., weight loss) options exist for the clinical management of OSAHS, positive airway pressure (PAP) is considered the treatment of choice (Giles et al., 2006). PAP consists of a nasal mask attached to a pneumatic pump by a flexible air tube, and held in place by a harness that fits over the head. The pump supplies positive air pressure to the upper airway, preventing it from collapsing during sleep. Once the appropriate pressure is determined, PAP generally eliminates most or all nocturnal breathing disturbances (American Thoracic Society, 1994). With proper use, PAP has been found to diminish sleep fragmentation, increase nocturnal oxygen saturation, and improve cognitive functioning (Giles et al., 2006). Despite these benefits, PAP is not a cure, but provides only a “pneumatic splint” to prevent collapse of the upper airway each night. Unfortunately, research has revealed that long-term adherence to PAP is less than optimal for reasons that remain to be

fully elucidated, but which include barriers associated with the equipment (e.g., mask discomfort, nasal congestion, nasal dryness) and psychological barriers to behavior change (Aloia et al., 2005; Matthews and Aloia, 2009). Adherence may indeed have implications for the magnitude of cognitive improvements with treatment. Neurobehavioral deficits in OSAHS Neurobehavioral functioning is a broad term that includes several specific cognitive functions (Beebe, 2005). Numerous studies of OSAHS have examined cognitive deficits, a broad term that describes cognitive performance outside of expected normal values in both global and specific cognitive domains. The cognitive domains will be summarized later in this chapter. Neurobehavioral testing is common in studies of OSAHS; however, cognitive sequelae remain difficult to interpret given the wide diversity of tests, and lack of standardized characterization of disease severity (Aloia et al., 2004; Beebe et al., 2003). Until recently, many investigations focused on selected cognitive deficits, and few studies employed comprehensive neurobehavioral test batteries. Fortunately, the number of studies examining cognitive deficits is expanding and advanced neuroimaging methods are improving our understanding of brain structure and function in OSAHS (Zimmerman and Aloia, 2006). Studies of cognitive deficits in adult OSAHS participants typically fall into three main areas: pretreatment group comparisons of OSAHS patients compared to healthy controls; OSAHS patients compared on levels of disease severity; and cognitive impairments before and after treatment, most often PAP. Apnea patients may exhibit fewer global cognitive impairments when compared to normal controls because the various components of the global score may mask specific deficits. Studies assessing specific cognitive domains have been more revealing, although cognitive domains are

73

not mutually exclusive. Cognitive abilities most frequently found to be affected by OSAHS are selective attention (concentration), sustained attention (vigilance), short-term or working memory, executive functioning, and motor functioning, with negligible impact on intellectual and language functioning (Beebe et al., 2003). Cognitive domains and OSAHS The most commonly investigated cognitive domains in OSAHS have been vigilance and attention. Although vigilance and attention are different constructs, apnea researchers do not always differentiate between them, and tend to favor tests of vigilance. Deficits in vigilance include sustained attention, controlled attention, information processing, and response time over long periods. This appears to be the domain that is most consistently affected in OSAHS patients and it has been suggested that vigilance and attention deficits are central to all aspects of higher order cognitive functioning (Verstraeten et al., 2004). Attention deficits in OSAHS patients may affect the ability to both remain awake in monotonous situations and manage information in more stimulating conditions (Mazza et al., 2005). This has implications for clinical practice and highlights the importance of conducting an adequate panel of vigilance tests. Recent studies have suggested that attention and alertness in OSAHS patients may be moderated by intelligence and age. For example, one study revealed that OSAHS patients with normal intelligence had greater deficits in selective and permanent attention compared with controls matched for age and intelligence, and these deficits were fully corrected after 1 year of PAP. High-intelligence OSAHS patients, however, had no greater attention deficits compared with controls of corresponding age and intelligence even before PAP treatment (Alchanatis et al., 2005). Measures of memory and executive function in OSAHS have been less thoroughly studied, due

in part to the challenging complexity of these domains. Memory is a broad domain comprised of long- and short-term memory functions that are associated with different memory processes. For example, long-term memory includes an episodic memory component (recollection of specific experiences) and a procedural memory component (learning skills). The construct of a short-term memory has evolved to include a multicomponent “working memory” subdomain. Working memory involves temporary storage and management of limited information required to carry out complex cognitive tasks such as learning and reasoning. Although interrelated, memory subdomains can be measured by a variety of specific tests. Limitations in memory testing have been attributed to complications in initial learning, free recall, or long-term forgetfulness (Aloia et al., 2004). Sleep apnea researchers do not always identify the precise memory subdomain and processes that are affected by OSAHS, which has led to difficulty interpreting discrepant memory-test findings across studies (Beebe et al., 2003). Impaired memory has been found in patients with moderate and severe OSAHS as well as older adults with OSAHS, relative to healthy controls. Greater impairment of information retrieval, and verbal and visual episodic memory subdomains have been demonstrated in OSAHS patients compared to healthy controls (FeriniStrambi et al., 2003; Naegele et al., 2006). Deficits of working memory in OSAHS patients have been reported in some studies (Redline et al., 1997; Salorio et al., 2002; Thomas et al., 2005); however, other studies have failed to demonstrate this impairment in OSAHS patients (FeriniStrambi et al., 2003; Lee et al., 1999). In a large sample of moderate to severe OSAHS patients and closely matched healthy controls, Naegele and colleagues (2006) recently reported that OSAHS patients had mild but significant memory impairment affecting episodic, procedural, and working memory relative to controls. Specifically, OSAHS patients demonstrated a retrieval deficit of episodic memory but intact subdomains of

74

maintenance, recognition, and forgetfulness. Taken together, these studies underscore the importance of extensive testing of specific memory subdomains to fully understand deficits in OSAHS. Comprehensive testing may provide clues about the affected brain regions and guide functional brain imaging. Findings from a study of impaired performance and decreased brain activation in OSAHS patients during a working memory task relative to health controls suggest that compromised brain function in response to cognitive challenges may underlie some of the cognitive deficits seen in patients with OSAHS (Ayalon et al., 2009). Archbold and colleagues (2009) found that with greater OSAHS severity, neuronal activation during working memory tasks was increased in the right parietal lobe, but decreased in the cerebellar vermis indicating that the severity of OSA may correlate with neural activation during tasks of working memory (Archbold et al., 2009). Executive function refers to the ability to develop and sustain an organized, goal-directed, and flexible approach to problem solving using basic cognitive skills such as working memory, mental flexibility, planning, and, to some degree, core language skills. The broadness of the executive function construct makes it difficult to accurately describe the deficits, assign causation, and distinguish executive dysfunction from impaired attention. Executive function is often operationalized as a measure of working memory (e.g., Wisconsin Card Sorting Test), set shifting (e.g., Trails B), or verbal fluency (e.g., Controlled Oral Word Association), in addition to testing behavioral inhibition, mental flexibility, planning, organizations, and problem solving. Despite the broadness of the concept and its measurement, OSAHS patients consistently underperform on tests of executive function relative to healthy controls. This includes a decreased ability to initiate new mental processes and to inhibit automatic ones relative to controls. In one study, OSAHS patients exhibited less-efficient use of semantic clustering, and poorer use of

semantic cues compared to healthy controls. With the exception of letter fluency, deficits were not observed in general executive control, and retention of previously encoded information and recognition was intact (Salorio et al., 2002). Psychomotor functions are characterized by fine motor coordination and psychomotor speed. Although the exact mechanism is unclear, OSAHS patients appear to perform more poorly on tests of psychomotor skills compared to healthy controls (see Aloia et al., 2004 for review). The deficits seem to be linked to fine motor coordination abilities rather than to motor speed. There has been comparatively less discussion about psychomotor deficits compared to other cognitive domains in OSAHS, in part because of the potential overlap with daytime sleepiness; however, this does not account for the difference between tests of fine motor skills and motor speed. Cognitive performance has been linked to severity of OSAHS. Most studies have found that OSAHS severity as measured by the apnea–hypopnea index (AHI) is associated with prolonged reaction times, impaired sustained attention, and monitoring information. Other studies have examined specific cognitive domains in relation to the severity of sleep fragmentation and hypoxemia. Contrary to expectation that psychomotor functioning would be associated with hypoxemia, the findings have been inconsistent. Moreover, psychomotor functioning appears to be resistant to PAP treatment, suggesting the possibility of an irreversible central nervous system damage in severe OSAHS. Models of neuropsychological deficits in OSAHS Brain operations responsible for cognitive function occur primarily in the cerebral cortex (gray matter layer covering the frontal lobes of the brain) and its subcortical structures (e.g., the hippocampus and lenticular nuclei). Various discrete mechanisms of cognitive dysfunction in OSAHS

75

have been proposed, including the effects of sleep fragmentation, chronic sleep deprivation, neuronal cell loss due to hypoxemia, cerebral vascular compromise (Aloia et al., 2004; Beebe, 2005; Lanfranchi and Somers, 2001; Verstraeten and Cluydts, 2004), and, most recently, inflammatory processes (Haensel et al., 2009). Four models will be summarized below. Cumulative evidence in animal and human studies suggests that the frontal lobes of the brain are most affected by OSAHS. Beebe and Gozal (2002) have proposed a model with two primary mechanisms, sleep fragmentation, and hypoxemia. Sleep fragmentation is thought to preferentially affect the frontal lobes by disrupting the normal restorative processes of sleep, while hypoxemia results in cellular changes in the prefrontal cortex. This model is supported by basic and clinical studies, particularly studies of the executive domain. Limitations of this model include exclusion of brain regions other than the frontal lobes and little discussion of the nuances of executive functions. The authors are credited, however, with the early development of one of the first neurofunctional models of OSAHS (Beebe and Gozal, 2002). A second “hierarchical” model posits that knowledge of lower order deficits (e.g., attention) that underlie more complex deficits (e.g., executive functioning) can provide better understanding of the cognitive mechanisms in OSAHS (Verstraeten et al., 2004). Previous studies have demonstrated that sleep disruption has a profound effect on arousal, processing speed, and attentional ability. Verstraeten and Cluydts (2004) hypothesized that higher order cognitive dysfunction in OSAHS can be explained by impairment in both attention decrements and slowed mental processing due to sleep disruption. The authors suggest that theoretical frameworks are helpful in guiding both the choice and interpretation of higher order cognitive tests. They conducted a study which provides support for this model by controlling for basic attentional

performance when evaluating executive attention in patients with severe OSAHS. Attention, vigilance, and executive functions were assessed using a battery of established neuropsychological tests including Trail Making Test A,B; Stroop; and Symbol Digit Modalities Test. The authors concluded that cognitive performance was comparable to the decline found after sleep loss, but qualitatively different from patients with chronic obstructive pulmonary disease. Verstraeten and colleagues (2004) suggest that sleepiness is the primary factor in cognitive deficits in sleep apnea, without the need to assume prefrontal brain damage. It is suggested that future studies systematically control for lower order functions when assessing executive tasks in OSAHS patients with arousal and attention difficulties. Justification for a cognitive hierarchy is compelling; however, discussion of sleep fragmentation and hypoxemia is limited in this model. A third model, the microvascular theory, was put forth by Aloia and colleagues in 2004. This model is based on the work of Lanfranchi and Somers (2001), and extensive cardiovascular and hypoxia literature establishing a link between cardiovascular dysfunction and OSAHS. Thus, it was plausible to suggest (1) vascular compromise might also exist in the small vessels of the brain, and (2) intermittent hypoxemia of OSAHS would preferentially affect the regions of the brain that were metabolically active during hypoxemia events. Damage to the vulnerable small vessels may result in a predictable pattern of cognitive deficits of attention, mental processing, memory, executive abilities, motor speed, and coordination. With sufficient underpinning in the literature, Aloia et al. (2004) proposed that this pattern of cognitive dysfunction was present in OSAHS, and represented microvascular disease of the small vessels feeding the white matter of the brain. Several studies highlighting the involvement of subcortical white matter of the brain using magnetic resonance imaging (MRI) were presented to support the model. The authors

76

demonstrated evidence in a small sample that microvascular diseases could be seen on brain MRI in OSAHS. More recent studies have demonstrated a significant decrease in the brain's white-matter volume in OSAHS patients attributed to decreased myelin and a reduction in the number of axons which could induce alterations in mood and cognition (Macey et al., 2008), while others have failed to find an association between white matter ischemia and OSAHS. A strength of this model is the ability to merge an established mechanism in OSAHS, vascular compromise, with the cognitive aspects of sleep apnea. Similar to Verstraeten and Cluydts’ hierarchical model, however, sleep fragmentation and hypoxemia are not well integrated into the model. Further research is needed to test and expand this model and relate vascular compromise to complaints of fatigue and daytime sleepiness. Recently, Beebe (2005) developed a comprehensive heuristic model in which multiple factors are proposed to affect neurobehavioral functioning among individuals with OSAHS. Beebe proposed that the effects of sleep fragmentation and hypoxemia are intermingled and may even be synergistic. He hypothesized that these synergistic mechanisms interact with vulnerable brain regions (e.g., the hippocampus, prefrontal cortex, subcortical gray and white matter), suggesting the potential involvement of small vessels in the brain (Beebe, 2005). Addressing the complexity of higher order cognitive abilities, this model suggests that the mechanisms of executive functions may be dependent on task demands or the testing environment, and deficits may be specific to some tasks but not others. Finally, this model addresses extraneous variables that should be considered in the mechanisms of cognitive dysfunction of OSAHS: (1) genetic endowment, (2) prior testing experience, and (3) sociodemographic factors. Although testing of this model is needed, it is a useful guide for future research, and provides a comprehensive approach to understanding the wide variation OSAHS cognitive dysfunction, including moderating factors.

Alteration in brain morphology and cognitive function The pathophysiology of alterations in the brain due to OSAHS may involve both chronic and acute insults on the cerebral vascular structure and function. As proposed in the microvascular model of OSAHS-related cognitive dysfunction (Aloia et al., 2004), these vascular changes may lead to alterations in the structure and function of the brain. Studies utilizing neuroimaging in OSAHS are increasing, and may provide unique information about brain structures, function, and metabolic composition that are affected by OSAHS (Zimmerman et al., 2006). Several of these studies will be summarized below. Subcortical brain systems play an important role in the regulation of cognitive and emotional processes, and frontal–subcortical circuits are a major organizing neural network of the brain. Changes in subcortical white matter and deep gray matter nuclei, often noted in older adults, appear as foci of increase signal intensity on certain pulse sequences of MRI. Current data suggest subcortical hyperintensities may reflect a spectrum of structural changes resulting from hypoperfusion of these subcortical regions (Campbell and Coffey, 2001). In one study of OSAHS patients, Colrain and colleagues (2002) demonstrated a relationship between severity of subcortical white matter hyperintensities and level of hypoxemia (Colrain et al., 2002). In another study, older adults with severe OSAHS had more subcortical white matter hyperintensities on brain MRI (functional MRI, fMRI) relative to those with minimal apnea, and there was a trend for a negative association between subcortical hyperintensities and free recall of a word list (Aloia et al., 2001). These subcortical hyperintensities suggest the involvement of small vessel damage in regions where hypoperfusion is more prevalent, however, that damage to these small vessels can result in a number of cognitive problems without specificity of any explicit domain (Aloia et al., 2004).

77

Structural volume changes have been demonstrated in brain regions including areas that regulate memory, executive function, and affect (e.g., frontal cortex, anterior cingulate, and hippocampus). Several studies utilized MRI to assess gray matter volume, which decreases with atrophy and cell death, in regions of the brain involved in motor regulation of the upper airway and cognitive function. For example, one study compared gray matter in 21 males with OSAHS with 21 controls, and found significant reductions in gray matter in several brain regions (i.e., anterior cingulate, hippocampus, frontal, parietal, and temporal lobes) that correlated with OSAHS severity (Macey et al., 2002). In another study, gray matter loss was demonstrated in the hippocampus, a key area for cognitive processing (Morrell et al., 2003). In contrast, one study did not find a significant loss of gray matter in 27 males with OSAHS relative to 24 controls (O’Donoghue et al., 2005). These negative results may be due to recruitment of a younger, healthier sample. Regional brain metabolism may provide insight into abnormalities in neurochemical transmission that may reflect pathologic insults to brain integrity. One study using positron emission tomography (PET) in OSAHS patients with residual daytime sleepiness despite effective PAP treatment revealed five of seven participants had impaired glucose utilization in the frontal, temporal, and/or parietal cortex (Antczak et al., 2007). Magnetic resonance spectroscopy (MRS) provides a measure of metabolic change through the chemical activity of neurotransmitters and amino acids (e.g., N-acetylaspartate [NAA], choline [Cho], creatine) that may reflect neural injury. Kamba and colleagues conducted two studies using MRS. In the first study, they reported that metabolic changes occur in normal-appearing white matter of patients with moderate to severe OSA (Kamba et al., 1997). A follow-up study addressed the effects of potentially confounding comorbid conditions. Fifty-five patients with severe OSAHS underwent MRS. The NAA/Cho ratio in the

cerebral cortex decreased with age, which is indicative of cerebral metabolic injury. The severity of OSAHS was found to have a significant negative association with the NAA/Cho ratio for normalappearing white matter that was independent of age or comorbid condition, suggesting the severity of OSAHS may be associated with the degree of white matter metabolic impairment and this impairment may be due to cerebrovascular risk factors (Kamba et al., 2001). Previous studies have revealed that OSAHS might compromise the recruitment of task-related brain regions compared with controls. Using MRI signal in OSAHS patients while performing a cognitive challenge, both over- and underactivation of specific brain regions have been reported (Ayalon et al., 2009). Thomas and colleagues (2005) found a lack of task-related signal activity in the dorsolateral prefrontal cortex in 16 untreated OSAHS patients compared with matched normal controls during a working memory challenge. In another study using fMRI, OSAHS patients had an increased activation response in several brain regions involved in attention tasks compared with controls on a verbal learning task. Notably, OSAHS patients recruited greater brain volume in task-related areas (e.g., cingulate, frontal) and other regions, suggesting the need for compensatory resources (Ayalon et al., 2009a). In summary, the pathophysiology of OSAHSrelated cognitive deficits is controversial and multifactorial. OSAHS brings with it disturbances in sleep stages, blood oxygenation, and sympathetic nervous system regulation. Since OSAHS appears to affect multiple brain regions, brain changes may reflect damage resulting from several etiologies, including hypoxemia, small vessel damage, cerebral circulation oscillations, chronic inflammation, all of which can result in local ischemia. This damage suggests an ongoing series of accumulated injuries leading to gradual cognitive changes that may be unrecognized by patients and clinicians for some time.

78

Neurobehavioral recovery after PAP PAP normalizes both sleep disruption and oxygen desaturation and has been found to dramatically reduce morbidity and mortality in a variety of populations. Owing to these encouraging findings, the effect of PAP on cognitive function has been an area of interest for many apnea researchers. In the 1980s, a cadre of researchers provided early evidence of short-term effect of PAP on improvement in a variety of cognitive deficits. In a review of PAP treatment studies in peerreviewed journals from 1985 to 2002, Aloia et al. (2004) examined whether pretreatment cognitive impairments were permanent or if they remitted with PAP (Aloia et al., 2004). Overall, studies of OSAHS and PAP reported a positive association between treatment adherence and improved cognitive performance. The review revealed attention/vigilance improved in the majority of studies, but changes in global functioning, executive functioning, and memory improved in about half of the studies. Tests of psychomotor function and construction, however, failed to improve with PAP in most studies. The authors noted that results may be dependent on the selected neuropsychological tests, which may vary in sensitivity to the effects of treatment, just as some may be more sensitive to specific cognitive deficits. Since the publication of the review, additional studies have provided greater understanding of cognitive recovery after effective PAP treatment (see Table 1). Recent studies have concluded that adherence to PAP, as measured by hours per night, is an important factor when evaluating cognitive outcomes. Zimmerman and colleagues (2006) compared three groups of memory impaired OSAHS patients based on average PAP use (poor, moderate, optimal users) at 3 months. At baseline, groups were similar with regard to demographic variables and verbal memory performance. Moderate users (2–5 h of use nightly) were three times as likely to develop normal memory after 3 months of PAP compared to poor users (< 2 h of use nightly); however, the difference did not

reach significance. Optimal users (> 6 h of use nightly) were eight times more likely to exhibited normalization of memory function compared with poor users at 3 months. These findings suggest that memory performance may be reversible with optimal levels of PAP treatment, and OSAHS patients with memory deficits at baseline may need 6 h of use per night to experience meaningful benefit in memory abilities (Zimmerman et al., 2006). In a multisite study of adherence in 149 OSAHS patients, Weaver and colleagues (2007) demonstrated that subjective sleepiness as measured by the Epworth Sleepiness Scale can improve with as few as 4 h of PAP use per night. Objective sleepiness as measure by the Multiple Sleep Latency Test (MSLT) may require 6 h of use and functional outcomes associated with sleepiness may take over 7 h. These studies, among others, show that adherence as well as test sensitivity must be considered in the design of efficacy trials (Weaver et al., 2007). Changes in brain structure and function after PAP Growing evidence suggest the OSAHS-related changes in brain morphology may improve with PAP use. Neuroimaging studies performed during cognitive testing have provided insight into PAP’s effect on function of neuroanatomical circuits in the brain. A recent study of the effects of acute withdrawal of PAP on brain function in nine individuals with OSAHS revealed significant treatment effects on working memory and recruitment of task-related brain regions during fMRI (Aloia et al., 2009).This study is clinically relevant because it mimics the common treatment holidays of many PAP users. Although neuroradiography can provide important insights into the structural and functional differences associated with OSAHS, one of the challenges is to interpret the findings in light of comorbid conditions that also cause neural injury. A second challenge in cross-sectional studies is the timing of neural injury: whether it preceded OSAHS or vice versa (Dempsey et al., 2010).

Table 1. Cognitive recovery after PAP for OSAHS (published in the past decade) Sample Reference Alchanatis et al. (2005)

Total # (# men) groups

Age M (SD)

83 47 OSAHS (22 HI, 20 NI) 36 healthy controls (15 HI, 21 NI)

OSAHS: HI 47.3 (7.9) NI 50.3 (7.6) Controls: HI 49.3 (2.9) NI 48.4 (4.1)

OSAHS severity M (SD) AHI: OSAHS: HI 62.8 (22.6) NI 69.6 (22.2) Controls: HI 4.52 (1.4) NI 3.95 (1.72)

Study variables

Key findings

Attention (selective and permanent), alertness before PAP use and 1 year later









Aloia et al. (2003)

12

64.8 (4.5)

RDI: 51.2 (19.8) adherent 45.9 (21.7) nonadherent

Attention, constructional abilities, motor speed, memory, language, executive functioning before and 3 months after CPAP







The HI and NI OSAHS patient groups did not differ with regard to OSAHS severity or sleepiness HI patients showed the same attention/ alertness performance compared with HI controls suggesting a protective effect against OSAHS cognitive decline NI patients showed deficits of reaction time, and selective and permanent attention decline compared with NI controls At 1-year follow-up, neither patient group showed any differences regarding attention and alertness compared with the control groups RDI at baseline was associated with delayed verbal recall, while oxygen desaturation was associated with both delayed recall and constructional abilities Compliant use of CPAP at 3 months was associated with greater improvements in attention, psychomotor speed, executive functioning, and nonverbal delayed recall Attention measures predicted compliance at 3 months

(Continued)

79

80

Table 1. Cognitive recovery after PAP for OSAHS (published in the past decade) (Continued) Sample Reference Aloia et al. (2009)

Ancoli-Israel et al. (2008)

Total # (# men) groups

Age M (SD)

OSAHS severity M (SD)

9 (5 men)

51.1 (9.3)

AHI: 42.4 (28.2)

52 older adults with Alzheimer's disease (39 men) 27 CPAP 25 placebo

78.6 (6.8) CPAP 77.7 (7.7) Placebo

AHI: 29.8 (16.1) CPAP 26.9 (15.5) Placebo

Study variables

Key findings

Verbal working memory during repeated fMRI under conditions of PAP treatment (at least one consecutive week) or nontreatment (for two consecutive nights)



Attention/vigilance, psychomotor speed, verbal episodic memory, executive functioning pretreatment, at 3 and 6 weeks







Treatment effects on working memoryrelated brain activity were significant, with greater deactivation in the right posterior insula and overactivation in the right inferior parietal lobule The observed responses to PAP treatment withdrawal were more extreme in all regions of interest, such that memory-related activity increased and memory-related deactivation decreased further relative to the control task A comparison of pre- and posttreatment neuropsychological test scores after 3 weeks of therapeutic CPAP compared to placebo PAP in both groups showed a significant improvement in cognition Post hoc examination of change scores for individual tests suggested improvements in episodic verbal learning and memory and some aspects of executive functioning such as cognitive flexibility and mental processing speed

Barbe et al. (2001)

54 29 CPAP 25 Placebo

54 (2)

AHI: 54 (3) Adherence to CPAP 5.0 (0.4) h/day

Attention, vigilance, memory, information processing, visual–motor coordination Quality of life, objective sleepiness



After 6 weeks of CPAP or placebo, there were no significant changes in objective sleepiness, vigilance, attention, memory, information processing, or visual–motor coordination between the groups

Bardwell (2001)

36 20 PAP 16 placebo

PAP 47 (1.9) Placebo 48 (2.2)

RDI: Placebo 43.6 (6.4) PAP 56.8 (5.4)

Attention, constructional abilities, motor speed, memory, language, executive functioning



Only 1/22 cognitive test scores showed significant changes specific to PAP treatment: Digit Vigilance-Time (p ¼ 0.035). The PAP group improved their time (from 7.5 to 6.9 min. p ¼ 0.013) The rank-sum test revealed that the PAP group had significantly better overall cognitive functioning posttreatment than the placebo group (mean ranks of 17.8 vs. 20.2, respectively; p ¼ 0.022)



Castronovo et al. (2009)

28 (all men) 14 OSAHS patients 14 healthy controls

42.15 (6.64) patients 43.93 (7.78) controls

50.14 (24.84) patients 349.36 (34.15) min/ night adherence

Working memory, learning, recall, recognition memory; executive functions (inhibition, selective attention), vigilance, sleepiness







Compared to controls, never-treated OSAHS patients showed increased activations in the left frontal cortex, medial precuneus, and hippocampus, and decreased activations in the caudal pons OSAHS patients showed decreases in activation in the left inferior frontal gyrus and anterior cingulate cortex, and bilaterally in the hippocampus after PAP compared to baseline suggesting a neural compensation mechanism, which is reduced by effective treatment Except for the Stroop test, neurocognitive domains, impaired at baseline, showed significant improvement after 3 months of PAP

Cooke et al. (2009)

10 (7 men) 5 sustained PAP users (PAPþ) 5 discontinued PAP (PAP)

75.7 (5.9)

AHI: 1.6 (2.3) on PAP Sustained CPAP use 13.3 month (5.2)

Cognitive decline , depressive symptoms, and daytime somnolence



Compared to the PAP group, the PAPþ group showed less cognitive decline with sustained CPAP use, stabilization of depressive symptoms and daytime somnolence, and significant improvement in subjective sleep quality

Felver-Gant et al. (2007)

56 (39 men)

52.8 (11.2)

AHI: 41.4 (22.1) Average adherence ¼ 4 (2.22) h/night

Working memory, executive functioning, motor speed prior to PAP and at 3 months



High adherers showed small improvements in both tests of working memory (2-back: F46 ¼ 4.73, p < 0.04; PASAT: F46 ¼ 4.92, p < 0.04), whereas low adherers performed evidence a decline in scores over time There were no treatment effects for other cognitive measures of executive functioning, motor speed



Ferini-Strambi et al. (2003)

23 OSAHS patients (21 men) 23 controls

56.52 (6.13)

AHI Patient 54.9 (13.37)

Attention, vigilance, memory, learning, verbal ability, executive functions, motor, and constructional abilities, sleepiness, depressive symptoms



At baseline, OSAHS patients had a significant impairment, compared to controls, in tests of sustained attention, visuospatial learning, executive function, motor performance, and constructional abilities (Continued)

81

82

Table 1. Cognitive recovery after PAP for OSAHS (published in the past decade) (Continued) Sample Reference

Total # (# men) groups

Age M (SD)

OSAHS severity M (SD)

Study variables

Key findings 

 

Lim et al. (2007)

46 17 CPAP 14 placebo 15 oxygen

46.7 (2.4) PAP 48.9 (3.2) placebo 47.1 (2.3) oxygen

AHI: 63.5 (7.8), PAP 65.8 (8.2), Placebo 58.6 (8.3), Oxygen

Attention, vigilance, working memory, executive functions, psychomotor performance, speed of information processing, global cognitive score







After a 15-day PAP, attentive, visuospatial learning, and motor performances returned to normal levels in 16 patients with adherence 5 h/night A 4-month PAP did not result in any further improvement in cognitive tests Executive functions and constructional abilities were not affected by short- and long-term treatment with PAP Pretreatment, OSAHS patients showed diffuse impairments, particularly speed of information processing, attention, and working memory, executive functioning, learning, and memory, alertness, and sustained attention There was no significant Time  Treatment interaction for the global deficit score. When examining individual neuropsychological test scores, two thirds of them improved with time regardless of treatment, although only Digit Vigilance-Time (p ¼ 0.020) showed significant improvement specific to CPAP treatment 2 weeks of CPAP or oxygensupplementation treatment was insufficient to show overall beneficial cognitive effects, as compared with placebo-CPAP. However, 2 weeks of CPAP treatment might be helpful in terms of speed of information processing, vigilance, or sustained attention and alertness

Munoz et al. (2000)

160 patients (156 men) 80 OSAHS 80 healthy controls

49 (1) OSAHS 46 (1) control

AHI: 60 (2) OSAHS patients

Vigilance, reaction time, daytime sleepiness, depression, and anxiety at baseline and 12  1 months





O'Donoghue et al. (2005)

51 (all men) 27 OSAHS patients 24 controls

45.7 (10.1) OSAHS 43.3 (9.4) controls

AHI: 71.7 (17.0) OSAHS 5.9 (4.7) controls

T1-weighted brain imaging in a highresolution MRI pre-PAP, at 6 months





Sweet et al. (2010)

Thomas et al. (2005)

10 (6 men)

32 16 OSAHS 16 healthy controls

51.1 (9.3)

AHI: 42.4 (28.2)

Working memory, sleepiness fMRI following regular CPAP use, and after two nights of CPAP withdrawal



Verbal working memory task and fMRI to map cerebral activation





Before treatment, OSAHS patients were significantly more somnolent, anxious, and depressed and had a longer reaction time and poorer vigilance relative to controls The use of CPAP improved significantly the levels of somnolence (p < 0.0001) and vigilance (p < 0.01), but failed to modify anxiety and depression No areas of gray matter volume change were found in OSAHS patients relative to controls and no differences were seen in bilateral hippocampal, temporal lobe, or whole brain volumes, assessed by manual tracing of anatomical borders No longitudinal changes were seen in gray matter density or regional volumes after PAP treatment, but whole brain volume decreased slightly without focal changes after 6 months of continuous PAP Compared to the treatment adherent baseline, significant memory-related deactivation was observed during the PAP withdrawal condition The magnitude of deactivation during withdrawal was significantly associated with better working memory performance in the posterior cingulate and right postcentral gyrus, and greater sleepiness in the left and right medial frontal gyrus Working memory speed in OSAHS patients was significantly slower than in controls, and a group average map showed absence of dorsolateral prefrontal activation, regardless of nocturnal hypoxia (Continued)

83

84

Table 1. Cognitive recovery after PAP for OSAHS (published in the past decade) (Continued) Sample Reference

Total # (# men) groups

Age M (SD)

OSAHS severity M (SD)

Study variables

Key findings 



Weaver et al. (2007)

149 (87% men)

46.8 (8.8)

64.1 (29.1)

Sleepiness (subjective, objective), functional status Pretreatment, after 3 months of PAP use







After treatment, resolution of subjective sleepiness contrasted with no significant change in behavioral performance, persistent lack of prefrontal activation, and partial recovery of posterior parietal activation, suggesting that working memory may be impaired in OSAHS and that this impairment is associated with disproportionate impairment of function in the dorsolateral prefrontal cortex Nocturnal hypoxia may not be a necessary determinant of cognitive dysfunction, and sleep fragmentation may be sufficient There were significant differences in mean nightly PAP duration between treatment responders and nonresponders with regard to sleepiness and daily functioning Thresholds above which further improvements were less likely relative to nightly duration of CPAP were identified for Epworth Sleepiness Scale score (4 h), Multiple Sleep Latency Test (6 h), and Functional Outcomes associated with Sleepiness Questionnaire (7.5 h) A linear dose-response relationship (p < 0.01) between increased use and achieving normal levels was shown for daytime sleepiness, but only up to 7 h of PAP use for functional status

Zimmerman et al. (2006)

58 (49 men) PAP use: 14 poor 25 moderate 19 optimal users

48.1 (10.1)

46.1 (27.9)

Verbal memory prior to the initiation of PAP treatment and at 3 months of PAP







Logistic regression analyses revealed that the odds of optimal users exhibiting normalization of memory function following 3 months of PAP therapy were 7.9 times (p ¼ 0.01) the odds of poor users exhibiting normalization of memory abilities Overall, 21% of poor users, 44% of moderate users, and 68% of optimal users exhibited memory performance in the clinically normal range following 3 months of PAP use (w(2) ¼ 7.27; p ¼ 0.03) These preliminary findings indicate that impaired verbal memory performance in patients with OSA may be reversible with optimal levels of PAP treatment. OSA patients exhibiting verbal memory impairments may experience a clinically meaningful benefit in their memory abilities when they use PAP for at least 6 h/night

M (SD), means (standard deviation); OSAHS, obstructive sleep apnea and hypoxia syndrome; HI, high intelligence; NI, normal intelligence; AHI, apnea/hypopnea index; RDI, respiratory disturbance index; fMRI, functional magnetic resonance imaging.

85

86

Conclusions and future directions In this brief review, the mechanisms of neurobehavioral dysfunction in OSAHS, and remittance of these deficits through PAP treatments have been discussed. Recent publications suggest distinct but not mutually exclusive deficits in multiple cognitive domains. There have been exciting new models outlining the mechanisms of cognitive deficits in OSAHS, and findings based on advanced neuroimaging methods. It appears that the cognitive dysfunction of OSAHS is not likely to be due to a single mediating mechanism, nor is it pervasive across all patients. There has been growing evidence for the effect of potential moderating factors. For example, studies have demonstrated that high cognitive reserves may spare some OSAHS patients from developing cognitive problems (Alchanatis et al., 2005). Future research should attempt to identify these moderators for cognitive dysfunction in OSAHS and to highlight the mechanisms of dysfunction by cognitive domain. Finally, as our knowledge of the physiological consequences of OSAHS increases, cognitive research should try to develop parsimonious theories of cognitive dysfunction that incorporate these physiological mechanisms. References 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., Malloy, P., Salloway, S., & Rogg, J. (2001). MRI white matter hyperintensities in older adults with obstructive sleep apnea. Sleep, 24, A55. 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., Arnedt, J. T., Stepnowsky, C., Hecht, J., & Borrelli, B. (2005). Predicting treatment adherence in obstructive sleep apnea using principles of behavior change. Journal of Clinical Sleep Medicine, 1, 346–353.

Aloia, M. S., Ilniczky, N., Di, D. P., 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. Aloia, M. S., Sweet, L. H., Jerskey, B. A., Zimmerman, M., Arnedt, J. T., & Millman, R. P. (2009). Treatment effects on brain activity during a working memory task in obstructive sleep apnea. Journal of Sleep Research, 18, 404–410. American Thoracic Society (1994). Indications and standards for use of nasal continuous positive airway pressure (CPAP) in sleep apnea syndromes. American Journal of Respiratory and Critical Care Medicine, 150, 1738–1745. Ancoli-Israel, S., Palmer, B. W., Cooke, J. R., CoreyBloom, J., Fiorentino, L., Natarajan, L., et al. (2008). Cognitive effects of treating obstructive sleep apnea in Alzheimer’s disease: A randomized controlled study. Journal of the American Geriatrics Society, 56, 2076–2081. Antczak, J., Popp, R., Hajak, G., Zulley, J., Marienhagen, J., & Geisler, P. (2007). Positron emission tomography findings in obstructive sleep apnea patients with residual sleepiness treated with continuous positive airway pressure. Journal of Physiology and Pharmacology, 58(Suppl. 5), 25–35. Archbold, K. H., Borghesani, P. R., Mahurin, R. K., Kapur, V. K., & Landis, C. A. (2009). Neural activation patterns during working memory tasks and OSA disease severity: Preliminary findings. Journal of Clinical Sleep Medicine, 5, 21–27. Ayalon, L., Ancoli-Israel, S., & Drummond, S. P. (2009). Altered brain activation during response inhibition in obstructive sleep apnea. Journal of Sleep Research, 18, 204–208. Ayalon, L., Ancoli-Israel, S., Aka, A. A., McKenna, B. S., & Drummond, S. P. (2009a). Relationship between obstructive sleep apnea severity and brain activation during a sustained attention task. Sleep, 32, 373–381. Barbe, F., Mayoralas, L. R., Duran, J., Masa, J. F., Maimo, A., Montserrat, J. M., et al. (2001). Treatment with continuous positive airway pressure is not effective in patients with sleep apnea but no daytime sleepiness. A randomized, controlled trial. Annals of Internal Medicine, 134, 1015–1023. Bardwell, W. A., Ancoli-Israel, S., Berry, C. C., & Dimsdale, J. E. (2001). Neuropsychological effects of oneweek continuous positive airway pressure treatment in patients with obstructive sleep apnea: a placebo-controlled study. Psychosomatic Medicine, 63, 579–584. 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.

87 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, 298–307. Campbell, J. J., III, & Coffey, C. E. (2001). Neuropsychiatric significance of subcortical hyperintensity. The Journal of Neuropsychiatry and Clinical Neurosciences, 13, 261–288. Castronovo, V., Canessa, N., Strambi, L. F., Aloia, M. S., Consonni, M., Marelli, S., et al. (2009). Brain activation changes before and after PAP treatment in obstructive sleep apnea. Sleep, 32, 1161–1172. Colrain, I. M., Bliwise, D., DeCarli, C., & Carmelli, D. (2002). The contribution of oxygen desaturation to the development of white matter hyperintensities in elderly male twins. Sleep, 25, A3. Cooke, J. R., Ayalon, L., Palmer, B. W., Loredo, J. S., CoreyBloom, J., Natarajan, L., et al. (2009). Sustained use of CPAP slows deterioration of cognition, sleep, and mood in patients with Alzheimer's disease and obstructive sleep apnea: A preliminary study. Journal of Clinical Sleep Medicine, 5, 305–309. Dempsey, J. A., Veasey, S. C., Morgan, B. J., & O'Donnell, C. P. (2010). Pathophysiology of sleep apnea. Physiology Reviews, 90, 47–112. 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, 589–594. Ferini-Strambi, L., Baietto, C., Di Gioia, 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. Giles, T. L., Lasserson, T. J., Smith, B. H., White, J., Wright, J., & Cates, C. J. (2006). Continuous positive airways pressure for obstructive sleep apnoea in adults. Cochrane Database of Systematic Reviews, 3, CD001106. Haensel, A., Bardwell, W. A., Mills, P. J., Loredo, J. S., Ancoli-Israel, S., Morgan, E. E., et al. (2009). Relationship between inflammation and cognitive function in obstructive sleep apnea. Sleep & Breathing, 13, 35–41. Kamba, M., Inoue, Y., Higami, S., Suto, Y., Ogawa, T., & Chen, W. (2001). Cerebral metabolic impairment in patients with obstructive sleep apnoea: An independent association of obstructive sleep apnoea with white matter change. Journal of Neurology, Neurosurgery and Psychiatry, 71, 334–339. Kamba, M., Suto, Y., Ohta, Y., Inoue, Y., & Matsuda, E. (1997). Cerebral metabolism in sleep apnea. Evaluation by magnetic resonance spectroscopy. American Journal of Respiratory and Critical Care Medicine, 156, 296–298. Lanfranchi, P., & Somers, V. K. (2001). Obstructive sleep apnea and vascular disease. Respiratory Research, 2, 315–319.

Lee, M. M., Strauss, M. E., Adams, N., & Redline, S. (1999). Executive functions in persons with sleep apnea. Sleep & Breathing, 3, 13–16. Lim, W., Bardwell, W. A., Loredo, J. S., Kim, E. J., ncoliIsrael, S., Morgan, E. E., et al. (2007). Neuropsychological effects of 2-week continuous positive airway pressure treatment and supplemental oxygen in patients with obstructive sleep apnea: A randomized placebo-controlled study. Journal of Clinical Sleep Medicine, 3, 380–386. Macey, P. M., Henderson, L. A., Macey, K. E., Alger, J. R., Frysinger, R. C., Woo, M. A., et al. (2002). Brain morphology associated with obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine, 166, 1382–1387. Macey, P. M., Kumar, R., Woo, M. A., Valladares, E. M., YanGo, F. L., & Harper, R. M. (2008). Brain structural changes in obstructive sleep apnea. Sleep, 31, 967–977. Matthews, E. E., & Aloia, M. S. (2009). Continuous positive airway pressure treatment and adherence in obstructive sleep apnea. Sleep Medicine Clinics, 4, 473–485. Mazza, S., Pepin, J. L., Naegele, B., Plante, J., Deschaux, C., & Levy, P. (2005). Most obstructive sleep apnoea patients exhibit vigilance and attention deficits on an extended battery of tests. The European Respiratory Journal, 25, 75–80. Morrell, M. J., McRobbie, D. W., Quest, R. A., Cummin, A. R., Ghiassi, R., & Corfield, D. R. (2003). Changes in brain morphology associated with obstructive sleep apnea. Sleep Medicine, 4, 451–454. Munoz, A., Mayoralas, L. R., Barbe, F., Pericas, J., & Agusti, A. G. (2000). Long-term effects of CPAP on daytime functioning in patients with sleep apnoea syndrome. The European Respiratory Journal, 15, 676–681. 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. O'Donoghue, F. J., Briellmann, R. S., Rochford, P. D., Abbott, D. F., Pell, G. S., Chan, C. H., et al. (2005). Cerebral structural changes in severe obstructive sleep apnea. American Journal of Respiratory and Critical Care Medicine, 171, 1185–1190. Redline, S., Strauss, M. E., Adams, N., Winters, M., Roebuck, T., Spry, K., et al. (1997). Neuropsychological function in mild sleep-disordered breathing. Sleep, 20, 160–167. 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, 93–100. Saunamaki, T., & Jehkonen, M. (2007). Depression and anxiety in obstructive sleep apnea syndrome: A review. Acta Neurologica Scandinavica, 116, 277–288.

88 Sweet, L. H., Jerskey, B. A., & Aloia, M. S. (2010). Default network response to a working memory challenge after withdrawal of continuous positive airway pressure treatment for obstructive sleep apnea. Brain Imaging and Behavior, 4, 155–163. Thomas, R. J., Rosen, B. R., Stern, C. E., Weiss, J. W., & Kwong, K. K. (2005). Functional imaging of working memory in obstructive sleep-disordered breathing. Journal of Applied Physiology, 98, 2226–2234. Tregear, S., Reston, J., Schoelles, K., & Phillips, B. (2009). Obstructive sleep apnea and risk of motor vehicle crash: Systematic review and meta-analysis. Journal of Clinical Sleep Medicine, 5, 573–581. Verstraeten, E., & Cluydts, R. (2004). Executive control of attention in sleep apnea patients: Theoretical concepts and methodological considerations. Sleep Medicine Reviews, 8, 257–267. Verstraeten, E., Cluydts, R., Pevernagie, D., & Hoffmann, G. (2004). Executive function in sleep apnea: Controlling for

attentional capacity in assessing executive attention. Sleep, 27, 685–693. Weaver, T. E., Maislin, G., Dinges, D. F., Bloxham, T., George, C. F., Greenberg, H., et al. (2007). Relationship between hours of CPAP use and achieving normal levels of sleepiness and daily functioning. Sleep, 30, 711–719. 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. Zimmerman, M. E., & Aloia, M. S. (2006). A review of neuroimaging in obstructive sleep apnea. Journal of Clinical Sleep Medicine, 2, 461–471. Zimmerman, M. E., Arnedt, J. T., Stanchina, M., Millman, R. P., & Aloia, M. S. (2006). Normalization of memory performance and positive airway pressure adherence in memory-impaired patients with obstructive sleep apnea. Chest, 130, 1772–1778.

H. P. A. Van Dongen and G. A. Kerkhof (Eds.) Progress in Brain Research, Vol. 190 ISSN: 0079-6123 Copyright Ó 2011 Elsevier B.V. All rights reserved.

CHAPTER 5

Effects of the use of hypnotics on cognition Annemiek Vermeeren{,* and Anton M. L. Coenen{ {

Department of Neuropsychology and Psychopharmacology, Maastricht University, Maastricht, The Netherlands { Donders Centre for Cognition, Radboud University Nijmegen, Nijmegen, The Netherlands

Abstract: Hypnotic drugs are intended to induce sedation and promote sleep. As a result, they have deteriorating effects on cognitive performance following intake. Most hypnotics are benzodiazepine receptor agonists which can have effects on memory in addition to their sedative effects. Other sedating drugs, such as histamine H1 antagonists or melatonin agonists, may have less effect on memory and learning. Hypnotics with other mechanisms of action are currently being investigated for efficacy and safety. For patients using hypnotic drugs, the effects on cognition are relevant to the extent that a drug dose affects daytime performance. Use of benzodiazepine hypnotics is associated with increased risk of car accidents and falling. Therefore, most hypnotics are studied to determine whether they produce residual sedation and impairing effects on performance the morning after bedtime use. Experimental studies using a standardized driving test clearly show that some drugs and doses produce severe residual effects, whereas others seem to have no or only minor impairing effects on next-day performance. No hypnotic has been found yet to improve daytime performance. Studies on long-term use of benzodiazepine hypnotics suggest that effects on daytime performance may diminish over time due to tolerance. However, there are also studies showing that performance may improve after discontinuation of chronic benzodiazepine use, which suggests that tolerance may not be complete. Keywords: hypnotics; benzodiazepines; zolpidem; (es)zopiclone; zaleplon; cognition; memory; driving; acute effects; residual effects; long-term use.

*Corresponding author. Tel.: þ31 43 3881952; Fax: þ31 43 3884560 E-mail: [email protected] DOI: 10.1016/B978-0-444-53817-8.00005-0

89

90

Introduction Population surveys show that between 0.7% and 7% of all adults report current use of sleepenhancing medication, while approximately 20–30% of the adults complaining of poor sleep (mostly females and elderly) report using some form of sleep-enhancing medication (for a review, see Vermeeren, 2004). Although sleep-enhancing medication includes not only prescribed hypnotics but also anxiolytics, antidepressants, and nonprescription (over the counter, OTC) drugs, the most frequently mentioned drugs for the treatment of insomnia are still the benzodiazepines or benzodiazepine-like drugs, that is, GABA agonists. In spite of the recommendation and consensus among clinicians that hypnotics should be used at their lowest possible doses and for limited durations only, the majority of hypnotic users report using the drug for more than a year. A major problem associated with the use of hypnotics may be the residual daytime sleepiness and associated impairment of psychomotor and cognitive functioning during the day after bedtime administration, sometimes called “hangover” effects. Already in 1982, an extensive critical review of 52 placebo-controlled studies assessing residual effects of benzodiazepine hypnotics on performance concluded that although these drugs generally improved sleep, they did not improve the quality of daytime performance, as expected when adverse effects of poor sleep would be normalized by the use of hypnotics (Johnson and Chernik, 1982). In particular, at higher dose levels, all hypnotics available at that time were likely to be associated with residual effects. Tasks showing the largest decrements were those involving speeded performance and memory for information presented during the night, indicating that the drugs mainly produced psychomotor slowing and anterograde amnesia. Newer hypnotics with shorter half-lives generally have a more favorable safety profile with respect to patients’ daytime functioning, although there is still little evidence that their use has significant positive effects on patients’ performance.

The high prevalence of hypnotic use in the population constitutes a public health problem. It can worsen the clinical expression of early dementia, and the effects on motor functions and attention have been shown to constitute risk factors for falls and accidents at home, at work, or on the road. A number of epidemiological studies have shown that the use of benzodiazepines is associated with increased risk of injurious car accidents (e.g., Barbone et al., 1998; Neutel, 1995, 1998) and falling and hip or femur fractures (c.f., Vermeeren, 2004). In the elderly, in particular, ataxia and impaired motor coordination may increase risk of falling and hip fractures, which is of concern, since hip fractures constitute a major cause for referrals to nursing homes. Hypnotics The majority of available prescription hypnotics are benzodiazepine receptor agonists (BzRAs) that act by enhancing the effectiveness of sleeppromoting GABAergic pathways. They act on the GABA-A receptor complex that regulates the chloride channels of the cell membrane. GABA-A receptors consist of five subunits of which the subunits are three principal families that are designated alpha (a), beta (b), and gamma (g). Whereas GABA binds at the interface between a and b subunits, most hypnotics act via the so-called benzodiazepine-binding site, located at the interface between a and g subunits (c.f., Winsky-Sommerer, 2009). The binding of GABA to its receptor induces a conformational change that opens the chloride ion channel, whereupon chloride ions can enter the cell producing a slight, short lasting hyperpolarization causing reduced excitability of the neuron. BzRA hypnotics increase the affinity of the receptors for GABA and enhance the effects of GABA on the neuron. Since 1987, many different subtypes of a, b, and g subunits have been identified, which has significantly improved insights into the mechanism of

91

action of hypnotics (c.f., Winsky-Sommerer 2009). Each of the subunits display several isoforms (a1–a6, b1–b3, g1–g3), but only certain combinations of subtypes seem to occur, and these appear to be localized in specific cell types and at different locations of the neurons. The most common receptor subtypes are a1b2g2, a2b3g2, and a3b3g2 (Nutt and Stahl, 2010). The majority of GABA-A receptors containing a1, a2, and a3 subunits are localized synaptically, while receptors containing a5 and d subunits combined with a4 and a6 appear to be predominantly peri- and/or extrasynaptic. Benzodiazepines act at GABA-A receptors containing a1, a2, a3, and a5 subunits (in combination with a g2 subunit), but not on receptors containing a4 or a6 subunits. The newer BzRAs (zopiclone, zolpidem, zaleplon, and eszopiclone) are structurally different from benzodiazepines and show differential-binding affinities and potencies for a1, a2, a3, and a5 subunits. They all bind to GABAA receptors containing a1, a2, and a3 subunits, but their affinity for a1 subunits seems relatively higher than for a2 and a3 subunits. In addition, zopiclone and eszopiclone act on a5-containing receptors, whereas zolpidem and zaleplon do not (Nutt and Stahl, 2010). BzRAs have multiple actions. The most prominent central effects are sedation, sleep induction, anxiety reduction, muscle relaxation, anterograde amnesia, and anticonvulsant activity. The sedative effects seem to be linked to the a1 subunit, which is the most common of the GABA-A subunits and present throughout the brain, particularly on hippocampal and cortical interneurons. This subunit does not seem to mediate changes in sleep EEG. GABA-A receptors containing a2 and a3 subunits located in the thalamus, hypothalamus, and cortex seem to be mediating EEG delta and theta activity, and thalamo-cortical oscillations during sleep. Receptors in the limbic system are assumed to mediate the anxiolytic effects. GABA-A receptors containing a5 subunits are highly expressed in the hippocampus suggesting a role for this subunit in effects of

benzodiazepine on learning and memory (Nutt and Stahl, 2010). Benzodiazepines mainly shorten sleep latency and diminish the number and duration of awakenings during sleep, thus increasing the total time spent asleep during the night. Yet, the extra time asleep is mostly spent in stage 2 or light sleep. Compared to normal sleep, the percentage of time spent in the (putatively) most restorative stages of sleep, that is, deep sleep (stages 3 and 4) and REM sleep, is decreased following administration of a benzodiazepine. With continued use of benzodiazepines, tolerance seems to develop to their effects on sleep stages, although rebound occurs when such use is discontinued. During the first nights after discontinuation, the increase in REM sleep may be especially prominent. Nonbenzodiazepine BzRAs show overall similar actions on sleep as benzodiazepines, including a dose-dependent reduction in REM sleep and the spectral EEG signature of benzodiazepines (i.e., reduction of EEG components below 10 Hz, and an increase in EEG power in the spindle frequency range during NREM sleep; WinskySommerer, 2009). All hypnotics have a rapid onset of action (between 30 and 90 min), whereas the duration of their action differs considerably. Both characteristics are dose dependent. Onset of action is largely determined by the pharmaceutical formulation and the rate of absorption of the drug from the gastrointestinal tract after oral administration. For rapidly absorbed benzodiazepines, such as diazepam and flurazepam, time to peak plasma concentration (tmax) is often taken to indicate the onset of action. In case of the slowly absorbed loprazolam, however, tmax is delayed with respect to the onset of action. The pharmaceutical preparation (formulation) can influence the rate of absorption; for example, temazepam is much more slowly absorbed from hard gelatin capsules than from soft capsules. Once reaching the blood, all benzodiazepines quickly reach their site of action, since all are lipophilic substances that easily traverse the blood–brain barrier.

92

BzRA hypnotics are often divided into categories based on their elimination half-life (t½), as short-acting (t½ less than 6 h), intermediate-acting (t½ between 6 and 24 h), or long-acting (t½ more than 24 h) drugs. As these categories imply, duration of action is often equated to elimination half-life. However, a drug's action may be terminated by at least three mechanisms: disappearance from the receptor site by redistribution from the brain to peripheral tissue, biotransformation by the liver to inactive metabolites, and acute tolerance of the receptors. In addition, dose is considered one of the most important determinants of a drug's duration of action. It will take longer for drug concentrations to drop below effective levels after administration of twice the recommended dose, and shorter after only half the recommended dose. The relation between half-life and duration of action is therefore not straightforward. Acute effects Most studies of the cognitive effects of hypnotics have focused on the acute adverse effects of benzodiazepines. Best known and detrimental are the effects on memory, in particular, on explicit (declarative, episodic) memory, whereas the other types of memory (implicit, procedural) are relatively unaffected (Curran, 1999). Several studies in human subjects have provided evidence that benzodiazepines have detrimental effects on memory processing, that is, benzodiazepines induce severe anterograde amnesia (Coenen et al., 1989; Curran and Birch, 1991; Lister, 1985). The administration of benzodiazepines before learning of a list of words such as the classic 15-words list impairs the recall of these words in a delayed recall test. In a series of experiments, Gorissen et al. (1995, 1997, 1998) and Gorissen and Eling (1998) tried to specify the benzodiazepineinduced amnesia more precisely. Effects of diazepam on the chain of encoding operations were investigated, such as on activation of memory

processes, on spreading of activation, on semantic encoding, on organizational processes, as well as on retrieval processes. The main conclusion from these studies was that diazepam impairs memory processing by slowing down and reducing cognitive processes. Under the influence of diazepam, subjects do not seem to benefit optimally from opportunities to organize the learning material adequately in an early stage of the acquisition of information. The adequate organization of the material in relevant memory chains appears affected, and this forms the basis for the observed anterograde amnesia. In a recent review paper, Beracochea (2006) extends this conclusion to the benzodiazepines in general and states that benzodiazepine-induced anterograde amnesia can be explained by an impairment of acquisition processes, due to a disruption of the ability to build new associations between pieces of information. A number of studies investigated whether the effects of benzodiazepine on memory and learning are due to a direct effect on memory processes or a by-product of their sedative effects. Results show that these effects can be dissociated and can act independently. For example, Curran and Birch (1991) found that the benzodiazepine antagonist flumazenil was able to reverse the midazolam-induced sedation effects, but could not counteract the amnesic effects of this benzodiazepine. Also, Gorissen et al. (1997) found that 24 h of sleep deprivation, as a form of nonpharmacological sedation, reduced subjective alertness and slowed reaction times more than diazepam 15 mg, but only diazepam, and not sleep deprivation, impaired the delayed recall of a word list task. This shows that the reduction in alertness could not account for the benzodiazepine-induced memory impairments. The different effects of benzodiazepines are supported by findings that GABA-A receptor subtypes demonstrate distinct regional distribution patterns in the brain. As described above, the amnestic effects of benzodiazepines are thought to be mediated by GABA-A receptors containing a5 subunits which are highly expressed in the

93

hippocampus and related regions (Nutt and Stahl, 2010), regions known to contribute to performance on episodic memory tests (Squire, 1992). Alertness-reducing effects, however, seem to be mediated by GABA-A receptors containing a1, a2, and a3 subunits in regions of the brain regulating and controlling wakefulness, such as the hypothalamic nuclei and areas of the brainstem. In contrast to the general finding that benzodiazepines impair memory, Hinrichs et al. (1984) first presented evidence for a remarkable, facilitating effect of benzodiazepines on memory. They showed that information provided just before the intake of benzodiazepines was recalled better than without the administration of benzodiazepines. Coenen and van Luijtelaar (1997) also demonstrated this retrograde facilitation effect on word recall for diazepam and flunitrazepam in healthy volunteers. These authors argued that the positive effect of the benzodiazepines on memory might be analogous to the positive effect of sleep on memory. Compared to wakefulness, sleep is associated with less interference during consolidation of previously learned information. Benzodiazepines induce sedation with alertness-reducing effects, which might imply that, in analogy to sleep, the encoding of new information is reduced. Hence, interference with the consolidation of previously obtained material is weaker, resulting in better recall of previously learned information. Thus, retrograde facilitation seems to be caused by reduced interference, due to the sedating effects of benzodiazepines. Benzodiazepine-induced sedation and effects on psychomotor performance have been found to differ from those of alcohol, which also acts as a modulator of the GABA-A receptor. Tiplady et al. (2003) found comparable impairing effects of temazepam and ethanol on information processing capacity and on long-term memory formation, but a dissociation between the effects on speed and accuracy of psychomotor performance. Temazepam caused a reduction in psychomotor speed with few changes in accuracy, while ethanol

was associated with a substantial increase in errors and only little effect on speed. Perhaps the differential action of these drugs could be explained by the broader effects of alcohol on other neurotransmitters, both on excitatory and inhibitory ones. Benzodiazepines have also been found to affect oculomotor behavior. In particular, they slow down saccadic eye movements, which may be related to effects on attention (Fafrowicz et al., 1995; Van Leeuwen et al., 1994). Performance in vigilance tasks was found to be very sensitive to the deleterious effects of benzodiazepines. For example, Van Leeuwen et al. (1992) evaluated the effects of the benzodiazepines bromazepam and oxazepam on performance in a long-duration visual vigilance test. The expected decrement in performance was found as a reduction in accuracy. In the case of bromazepam, however, this was associated with improved speed, suggesting that subjects were less cautious with bromazepam (van Leeuwen et al., 1992). The effects of benzodiazepines on eventrelated potentials in the EEG have also been studied to unravel to their effects on cognitive processes. Event-related potentials are useful as a means to gain more insight into the covert processes underlying changes in performance and in executive functions. Results from these studies suggest that effects of benzodiazepines are already manifest in an early stage of information processing, as reflected by effects on the N1 amplitude (Abduljawad et al., 2001; Van Leeuwen et al., 1992). Effects on N1 indicate a deterioration of stimulus detection ability (attention–detection), which is thought to be responsible for the slowing down and adverse effects on later information processing. Subjects under the influence of benzodiazepines probably gather less signal information for further processing due to dampening of the input signal. Thus, administration of benzodiazepines deteriorates the ability of a subject to detect relevant information in an early stage of acquisition (van Leeuwen et al., 1992). In line with this,

94

benzodiazepines have also been found to reduce the amplitude of P3 in an oddball task (Unrug et al., 1997a,b). Generally, the electrical positivity of P3 reflects cognitive processing of incoming information. The reduction of P3 by a benzodiazepine is therefore interpreted as a reduction of cognitive processing of relevant stimuli. This reduction was also found to be associated with more omissions in target detection and longer reaction times. The changes in electrical brain activity produced by benzodiazepines correspond well with their deteriorating effects on cognitive functions. All in all, anterograde amnesia associated with benzodiazepine use most likely results primarily from less thorough cognitive processing of relevant information. In addition to their effects on evoked potentials, benzodiazepines also have nonspecific effects on the EEG, in particular, on the background EEG. Since benzodiazepines have sedative properties, an increase in low frequencies is expected. In contrast, however, benzodiazepines have been found to increase high frequency beta activity in the background EEG. This phenomenon is known as “pharmacological dissociation” because the effects on behavior are not in agreement with the effects on background EEG (Coenen and van Luijtelaar, 1991). So, whereas the effects of benzodiazepines on event-related potentials reflect their effects on behavior, their effects on background EEG do not reflect their effects on behavior. Although there may be some slight pharmacodynamic differences, the mechanism of action and the behavioral effects of nonbenzodiazpine BzRAs, such as (es)zopiclone, zolpidem, and zaleplon, are largely comparable to those of benzodiazepines. Similar to benzodiazepines, they produce sedation and associated impairment of attention and psychomotor performance. They also impair memory functions (e.g., Leufkens et al., 2009a,b; Leufkens and Vermeeren 2009; Vermeeren et al., 1998). In line with this, flumazenil, a benzodiazepine receptor antagonist which is used for antagonizing a benzodiazepine

overdose, was found to be able to reverse the zolpidem's sedative and memory impairing effects (Quaglio et al., 2005). It might be argued that the effects of nonbenzodiazepine BzRAs are less severe and of shorter duration than those of most benzodiazepines, but that can also be explained by pharmacokinetic differences. Still, not all GABA agonists seem to have similar effects on cognitive performance. For example, gaboxadol, a hypnotic that has been withdrawn from development in 2007, was found to have residual effects on car driving and psychomotor performance, but not on memory (Leufkens et al., 2009a). In contrast to BzRAs, gaboxadol does not interact with synaptically located GABA-A receptors, but exerts its effects via extrasynaptic GABA-A receptors containing a4 and a6 subunits. These receptors have been found to mediate persistent tonic inhibition, which is assumed to have a different functional role compared to phasic postsynaptic responses (WinskySommerer 2009). Residual effects Hypnotics all have largely comparable pharmacodynamic effects, but they differ substantially in their pharmacokinetic profiles. As a consequence, the cognitive problems associated with their use in clinical practice differ mainly in quantity (i.e., magnitude and duration), not in quality. Hypnotics are intended to be taken at bedtime and rapidly induce sleepiness and sedation. Ideally, these effects should continue throughout the sleep period but should no longer be present after awakening in the morning. Unfortunately, that is not the case for many drugs (for a review, see Vermeeren, 2004). Results from experimental performance studies show that hypnotics can have deteriorating effects on psychomotor performance, attention, and memory the day after bedtime use. Epidemiological studies confirm that these effects reduce patients’ quality of life and increase the risk of becoming involved in

95

accidents, such as falling, hip fractures, and traffic accidents. Both experimental and epidemiological studies show that increased risk varies with treatment-related factors, such as drug, dose, time after dosing, and frequency of dosing, and with patient-related factors such as age and gender. Several hypnotics are available in doses that have no or minimal next-day residual effects. So, the most effective way of minimizing the accident risks associated with use of hypnotics is to prescribe a safe alternative. However, if use of a hypnotic without clinically relevant residual effects is not possible, patients should at least be adequately informed about the duration and severity of the residual effects in order to be able to adjust their behavior appropriately. Information on the duration and severity of residual effects is best derived from experimental studies using objective performance tests that validly measure druginduced changes in daytime functioning. Many studies have been conducted to determine whether a particular drug dose has effects on performance that differ from placebo. The differences in methodologies used make comparison between studies very difficult, however. A few performance tests have been applied consistently for several years, providing comparable data on the effects of a variety of drugs and doses. One such test is a highway driving test, which was standardized in the early 1980s (O'Hanlon, 1984) and subsequently used in over 75 studies. The test evolved from studies on driver fatigue conducted in the United States during the early 1970s. It involves subjects driving a specially instrumented car over a 100-km (61 mile) primary highway circuit while maintaining a constant speed and a steady lateral position between the boundaries of the slower traffic lane (Fig. 1). Subjects are supervised by a licensed driving instructor, having access to dual controls. Speed and lateral position relative to the lane delineation are continuously recorded during the 1-h drive by apparatus aboard the vehicle. After completion of the test, several parameters are derived from the data, including the primary performance

parameter, standard deviation of lateral position (SDLP, in centimeters). SDLP can be interpreted as an index of weaving or road tracking error. It is a reliable index of individual driving performance (the test–retest correlation ranges from 0.7 to 0.9) and has proven sensitive to many sedating drugs (c.f., Ramaekers, 2003; Theunissen et al., 2009; Vermeeren 2004, Vermeeren et al., 2009). The test calibrated for the effects of alcohol in a closed circuit study wherein 24 social drinkers were tested sober and after controlled drinking, raising blood alcohol concentrations (BACs) in steps of 0.3 mg/ml to a maximum of 1.2 mg/ml (Louwerens et al., 1987). In line with the established relation between BAC and accident risk, the relation between BAC and SDLP was shown to be an exponential function. Based on this relation, BACs of 0.5, 0.8, and 1.0 mg/ml were associated with mean changes in SDLP of 2.4, 4.2, and 5.1 cm. Mean changes in driving performance under the influence of hypnotic drugs can thus be compared to those associated with BACs at various limits considered legal in different countries. Figure 2 shows the results from 12 studies employing comparable procedures for assessing the residual effects of hypnotic drugs on driving performance the next morning. Five studies assessed residual effects after two nights of treatment with hypnotics in women complaining of insomnia and who previously used hypnotics. In the other studies, testing occurred with subjects who were healthy and included both genders, or after a single night of treatment. In all studies, driving tests were undertaken in the morning between 10 and 11 h after intake. In six studies, a second driving test was performed in the afternoon, between 16 and 17 h after bedtime administration. The effects of zolpidem, zopiclone, and zaleplon were also assessed after administration of these drugs in the middle of the night, that is, 4 or 5 h before testing. The most severe residual effects were found for flurazepam 30 mg and loprazolam 2 mg. The average degrees of impairment were worse than

96 (a)

(b)

Lateral position8 (c) Placebo e.g. SDLP = 18 cm

Sedating drug e.g. SDLP = 35 cm Fig. 1. Highway driving test used in experimental studies for assessing residual effects of hypnotics on driving performance. (a) Subjects drive a specially instrumented vehicle over a 100-km primary highway in normal traffic, accompanied by a driving instructor having access to dual controls. They are instructed to drive as straight as possible in the middle of the slower (right) traffic lane with a constant speed of 95 km/h. (b) A camera on top of the car continuously registers the lateral position of the car on the road with respect to the left lane delineation. (c) The standard deviation of lateral position (SDLP in centimeters) is an index of road tracking error or “weaving.” It is a highly reliable variable of individual driving performance (mean test–retest correlation is 0.85) and it has proven sensitive to the effects of many sedating drugs including low doses of alcohol.

those associated with a BAC of 1.0 mg/ml in the morning and equivalent to 0.8 mg/ml in the afternoon. Drugs that had residual effects in the morning equivalent to BACs between 0.5 and 0.8 mg/ ml were nitrazepam 10 mg, flunitrazepam 2 mg, zopiclone 7.5 mg, oxazepam 50 mg, and lormetazepam 2 mg (capsules). The residual effects dissipated rapidly over time for hypnotics with short and intermediate half-lives (zopiclone 7.5 mg, lormetazepam 2 mg, and oxazepam 50 mg) but remained significant or even increased for hypnotics with long half-lives (flunitrazepam 2 mg and nitrazepam 10 mg, respectively). Drugs that had no significant residual effects in the morning and afternoon were zaleplon 10 and 20 mg, zolpidem 10 mg, lormetazepam 1 mg (capsules), temazepam 20 mg (soft gelatin capsules), and nitrazepam 5 mg. Table 1 summarizes the effects of hypnotics doses at different times after administration, categorized as unlikely, minor (comparable to BACs < 0.5 mg/ml), moderate (comparable

to BACs between 0.5 and 0.8 mg/ml), and severe (comparable to BACs > 0.8 mg/ml). For more details, see Vermeeren (2004). Not shown in the figure are the results of a study comparing the residual effects of triazolam 0.5 mg, midazolam 15 mg, and temazepam 20 mg (in soft gelatin capsules) after daytime sleep in shift workers. Results are not comparable to those of the other studies because the procedures differed: the driving test was performed in the afternoon, between 7.5 and 8.5 h after morning ingestion of drugs or placebo. Nonetheless, results confirmed previous findings suggesting that temazepam 20 mg is unlikely to produce residual effects on driving. In contrast, triazolam 0.5 mg produced residual impairment equivalent to a BAC over 1.0 mg/ml after the first treatment and equivalent to a BAC of 0.8 mg/ml after the fifth consecutive treatment. Midazolam 15 mg had minor effects on the fifth day of treatment, but none on the first day.

97

*

*

*

FLU 30

Equivalent effects of alcohol while BAC is

LOP 2

8

6 1.0 mg/ml

*

0.8 mg/ml 4

* *

ZOP 7.5

SEC 200

FLU 15

ZOP 7.5

FLN 2

LMT 2 (caps)

OXA 50

* * *

ZOP 7.5

NIT 10

GBX 15

TEM 20

NIT 5

LMT 1 (tabs)

–2

ZPD 10

ZAL 10

0

LMT 1 (caps)

*

*

ZOP 7.5

2

*

LOP 1

0.5 mg/ml

* *

Fig. 2. Residual effects of hypnotics on performance in a standardized highway driving test between 10 and 11 h after bedtime administration. Effects on SDLP (in centimeters) are presented as mean changes from placebo. Indicated are the hypnotic doses (in milligrams) and formulation (caps, capsules; tabs, tablets) for flunitrazepam (FLN), flurazepam (FLU), gaboxadol (GBX), lormetazepam (LMT), loprazolam (LOP), nitrazepam (NIT), oxazepam (OXA), secobarbital (SEC), temazepam (TEM), zaleplon (ZAL), zolpidem (ZPD), and zopiclone (ZOP). Asterisks indicate significant (p < 0.05) differences from placebo. The dotted lines indicate the equivalent effects of alcohol on SDLP while blood alcohol concentrations are 0.5, 0.8, and 1.0 mg/ml, as measured by Louwerens et al. (1987).

Effects in elderly and insomnia patients have recently been studied by (Leufkens et al., 2009b,c; Leufkens and Vermeeren, 2009). Using the same test and methods, they found that the severity of residual effects in middle-aged and older-age groups (up to 75 years) is comparable to those found in young adults. Temazepam 20 mg was found to have no residual effects on driving in the elderly, whereas the effects of zopiclone 7.5 mg in the same group were comparable to those found in young volunteers (Leufkens and Vermeeren, 2009). In addition, these investigators found that zopiclone 7.5 mg had moderately severe residual effects on driving of insomnia patients who did not use hypnotics on a regular basis (Leufkens et al., 2009b). Effects in insomniacs chronically using hypnotics were found to be slightly reduced, but still significant. This suggests that the residual effects of hypnotics are not compensated by therapeutic

effects on sleep in patients, and that tolerance to these effects is not complete in chronic users. Long-term effects In spite of recommendations to limit the use of hypnotics to a few weeks, a large number of patients use them chronically. This raises the question whether the adverse effects on cognition diminish with prolonged use or even disappear, due to the development of tolerance. To answer this, a number of studies have compared cognitive performance of long-term benzodiazepine users to that of controls using cross-sectional designs. According to a metaanalytic evaluation of 13 studies published between 1980 and 2000, long-term benzodiazepine users show impairment over a wide range of cognitive functions (Barker et al., 2004a). The authors

98 Table 1. Categorization of the residual effects of hypnotics Drug

Dose (mg)

Time after bedtime administration 4-8 h (2nd half of the night)

8-12 h (morning)

12-16 h (afternoon)

16-22 h (evening)

zaleplon zaleplon

10 20

Unlikely Unlikely

Unlikely Unlikely

Unlikely Unlikely

Unlikely Unlikely

zolpidem midazolam temazepam SGC triazolam lormetazepam capsules

10 7.5 20 0.125 1

Moderate Moderate Moderate Moderate Moderate

Unlikely Unlikely Unlikely Unlikely Unlikely

Unlikely Unlikely Unlikely Unlikely Unlikely

Unlikely Unlikely Unlikely Unlikely Unlikely

lormetazepam tablets midazolam temazepam HGC triazolam zolpidem

1 15 30 0.25 20

Severe Severe Severe Severe Severe

Minor Minor Minor Minor Minor

Unlikely Unlikely Unlikely Unlikely Unlikely

Unlikely Unlikely Unlikely Unlikely Unlikely

lormetazepam capsules loprazolam flunitrazepam triazolam zopiclone

2 1 1 0.5 7.5

Severe Severe Severe Severe Severe

Moderate Moderate Moderate Moderate Moderate

Unlikely Unlikely Unlikely Unlikely Unlikely

Unlikely Unlikely Unlikely Unlikely Unlikely

nitrazepam

5

Severe

Minor

Minor

Unlikely or minor

flunitrazepam

2

Severe

Moderate

Minor to moderate

Minor

nitrazepam flurazepam

10 15

Severe Severe

Moderate Moderate

Moderate Moderate

Moderate Moderate

flurazepam loprazolam

30 2

Severe Severe

Severe Severe

Severe Severe

Moderate Moderate

Source: Vermeeren (2004).

categorized the neuropsychological tests used in these studies into 12 cognitive domains and calculated effect sizes for differences found in each domain. Results showed that long-term benzodiazepine users were consistently more impaired than controls across all cognitive categories examined, with weighted effect sizes ranging between  1.30 for sensory processing and  0.42 for verbal reasoning. The most frequently used test in these studies was the digit symbol substitution test (DSST, category psychomotor speed, effect size  0.99), and the most frequently measured category was verbal memory (effect size  0.58). This clearly shows that the negative effects of benzodiazepines on cognition

do not disappear due to development of tolerance. This is of particular concern for elderly patients, many of whom use these drugs chronically and show age-related memory decline. Additional druginduced cognitive dysfunction may induce a state of clinical dementia. It should be noted that most studies included in the meta-analysis did not differentiate between benzodiazepines used as hypnotics and benzodiazepines used as anxiolytics. This may be relevant, however, because hypnotics should be effective during the night with minimal residual effects during the day, whereas anxiolytics should be effective at daytime. Van Steveninck et al.

99

(1997) showed that development of tolerance to the effects of a hypnotic (temazepam) was less than to the effects of an anxiolytic (lorazepam). Chronic users of lorazepam showed no objective impairment compared to controls, despite high plasma concentrations. In addition, their response to a subsequent acute dose of lorazepam was much smaller than in controls. In contrast, chronic users of temazepam and controls responded similarly to an acute dose of temazepam. Chronic users of lorazepam had higher baseline drug concentrations than users of temazepam, which suggests that exposure to lorazepam was more prolonged than to temazepam. This may have led to a difference in the development of tolerance. It could be argued that there is development of tolerance, but that there are premorbid differences in cognitive functioning between benzodiazepine users and controls that explain the differences between the groups. Longitudinal studies are therefore needed to reveal differential changes in cognition of users and nonusers over time. A longitudinal populationbased study by Paterniti et al. (2002) examined cognitive performance of 1176 elderly (aged 60–70 years) at 2-year intervals over a 4-year period. Results showed that those reporting to take benzodiazepines on all of these examinations did not differ significantly from nonusers at the start, but did show an accelerated decline of cognitive performance compared with nonusers. This indicates that long-term use of benzodiazepines is a risk factor for cognitive decline in elderly. The development of new hypnotics and increasing awareness of the risks associated with use of older benzodiazepines has changed the drugs prescribed. The majority of prescriptions nowadays are for relatively low doses of short half-life hypnotics, which have minor residual effects on daytime performance. In line with this, a recent study by Puustinen et al. (2007) found no difference in cognitive functioning between a group of elderly chronic users of zopiclone, temazepam, and oxazepam, and a group of nonusers. The authors examined 164 elderly patients admitted to acute hospital wards in Finland, using

interviews, the Mini Mental State Examination (MMSE), and analyses of serum concentrations of hypnotics. Only temazepam serum concentrations correlated negatively with decreases in MMSE scores, which was probably due to the relatively high doses used in this frail population. Similarly, Leufkens et al. (2009b,c) recently failed to find differences in cognitive performance and car driving ability between chronic users of hypnotics and a group of controls, which may have been due to the fact that the majority of patients in this study used hypnotics in doses that were not expected to have residual effects on performance. The same subjects participated in a subsequent double-blind placebo-controlled crossover study assessing the residual effects of a hypnotic known to produce nextday impairment (zopiclone 7.5 mg). Results of this study showed that chronic users were still sensitive to the adverse residual effects on driving and cognitive performance of zopiclone. It was also found that the drug–placebo differences in performance were smaller in chronic users than in controls, suggesting that chronic users were partially tolerant. However, performance after use of placebo was also worsened compared to baseline in chronic users, suggesting that their relatively small drug–placebo difference may in part be due to the adverse effects of withdrawal. Although discontinuation of hypnotics may initially be associated with withdrawal symptoms, it may result in a reversal of drug-induced impairment after some time. Two further metaanalyses of studies before 2000 by Barker et al. (2004b) focused on the changes after discontinuation and impairment at long-term follow-up. The authors concluded that while some recovery of function was observed after discontinuation, previous users displayed impairment in many areas some years after discontinuation. Some studies examined recovery in older adults. One of them was conducted by McAndrews et al. (2003), who examined a sample of outpatients (aged 50 years and older) presenting to a sleep clinic with

100

complaints of sleep problems. Twenty-five patients completed a cognitive test battery before discontinuation and at 1 month postdiscontinuation. The battery comprised measures in three cognitive domains: attention/speed of processing, psychomotor speed, and learning/ memory. Results showed that the performance of benzodiazepine users at baseline showed modest reductions in attention and psychomotor speed compared to controls. Importantly, however, benzodiazepine users improved more from baseline to postrecovery reassessment than controls. So, recovery from benzodiazepine use was associated with subtle and reversible effects on cognition. Another study was conducted by Curran et al. (2003). They examined 192 long-term users of benzodiazepine hypnotics aged 65 years or older. Of these, 104 who wished to withdraw from benzodiazepines were randomly allocated to one of two groups under double-blind, placebo-controlled conditions. In group 1, the benzodiazepine dose was tapered from week 1 of the trial. Group 2 were administered their usual dose for 12 weeks, which was then tapered thereafter. An additional group of 35 patients who did not wish to withdraw from benzodiazepines participated as controls. All patients were assessed at 0, 12, and 24 weeks, and one-half of these patients were reassessed at 52 weeks. Sixty percent of the patients had been taking the drug continuously for > 10 years, while 27% of the patients had been taking the drug continuously for > 20 years. Of all the patients beginning the trial, 80% had withdrawn successfully 6 months later. There was little difference between groups 1 and 2, but both groups differed from the controls in that the performance of the withdrawers on several cognitive and psychomotor tasks showed improvements at 24 and 52 weeks relative to baseline. Withdrawers and controls did not differ in sleep (as measured by subjective ratings of problems sleeping or intensity of dreaming) or benzodiazepine withdrawal symptoms. The results of this study imply that withdrawal from benzodiazepines produces some

subtle cognitive advantages for elderly people, yet little in the way of withdrawal symptoms or emergent sleep difficulties. Novel hypnotics Most currently available prescription hypnotics enhance the effects of the sleep-promoting neurotransmitter GABA, whereas many OTC treatments induce sedation by blocking the wake promoting neurotransmitter histamine. These drugs contain antihistamines, such as diphenhydramine and hydroxyzine, which are antagonists for histamine H1 receptors and can penetrate the central nervous system. Several antidepressant and antipsychotic drugs have similar effects on H1 receptors and can induce pronounced sedation. In fact, some are antidepressants that are currently in development for the treatment of insomnia, for example, low doses of doxepin (Krystal et al., 2010). The effects of H1 antagonism on cognition have been extensively studied and were recently reviewed by van Ruitenbeek et al. (2010). They conclude that H1 antagonism primarily affects psychomotor functions and attention and have little effect on memory. Melatonin is involved in the circadian regulation of sleep and is therefore also a target for the treatment of sleep problems. Administration of exogenous melatonin or analogs such as ramelteon, which is licensed in the United States, can promote sleep onset (Wilson et al., 2010). Effects on cognitive functions have not been studied as extensively as those of benzodiazepines and antihistamines. Most studies assessed residual effects on performance using a DSST and no significant residual effects of ramelteon on this test have been reported. Mayer et al. (2009) evaluated the residual effects of ramelteon in adults with chronic insomnia using the DSST and a word recall test. Although these authors report a consistently reduced sleep onset, with no next-morning residual effects, it is, however, still too early to draw final conclusions about its efficacy and adverse effects.

101

Nevertheless, drugs binding to the melatonin receptors seem to be interesting as potential hypnotics, given the main role of melatonin in the sleep–wake cycle. Other targets for the treatment of insomnia are antagonists for serotonin 5HT-2 receptors and orexin/hypocretin receptors (c.f., Wafford and Ebert, 2008). As these compounds are very new and mostly still under development, little information is published yet on the effects on cognition. Summary and conclusion To summarize, hypnotic drugs can have deteriorating effects on cognitive performance, which is expected from all drugs when taken in doses that induce sedation and promote sleep. BzRAs can have effects on memory that are at least partially independent of their sedative effects. Other sedating drugs, with different mechanisms of action such as histamine H1 antagonists or melatonin agonists, may have less effect on memory and learning. For patients using hypnotic drugs, the effects on cognition are relevant to the extent that a drug dose affects daytime performance. Therefore, most hypnotics are studied to determine whether they produce residual sedation and impairing effects on performance the morning after bedtime use. Some drugs and doses produce severe residual effects, whereas others seem to have no or only minor impairing effects on next-day performance. No hypnotic has been found yet to improve daytime performance, either directly or by improving sleep. Studies on long-term use of benzodiazepine hypnotics suggest that effects on daytime performance may diminish over time due to tolerance. However, there are also studies showing that performance may improve after discontinuation of chronic benzodiazepine use, which suggests that tolerance may not be complete. Selection of a drug that has minimal residual sedating effects

the next day is therefore important, whatever its mechanism of action. More studies are needed to determine whether use of a hypnotic drug can actually improve cognitive functions in patients complaining of insomnia. References Abduljawad, K. A., Langley, R. W., Bradshaw, C. M., & Szabadi, E. (2001). Effects of clonidine and diazepam on prepulse inhibition of the acoustic startle response and the N1/P2 auditory evoked potential in man. Journal of Psychopharmacology, 15, 237–242. Barbone, F., McMahon, A. D., Davey, P. G., Morris, A. D., Reid, I. C., McDevitt, D. G., et al. (1998). Association of road-traffic accidents with benzodiazepine use. Lancet, 352, 1331–1336. Barker, M. J., Greenwood, K. M., Jackson, M., & Crowe, S. F. (2004a). The cognitive effects of long-term benzodiazepine use: A meta-analysis. CNS Drugs, 18, 37–48. Barker, M. J., Greenwood, K. M., Jackson, M., & Crowe, S. F. (2004b). Persistence of cognitive effects after withdrawal from long-term benzodiazepine use: A meta-analysis. Archives of Clinical Neuropsychology, 19, 437–454. Beracochea, D. (2006). Anterograde and retrograde effects of benzodiazepines on memory. The Scientific World, 6, 1460–1465. Coenen, A. M. L., & van Luijtelaar, E. L. J. M. (1997). Effects of benzodiazepines, sleep and sleep deprivation on vigilance and memory. Acta Neurologica Belgica, 97, 123–129. Coenen, A. M. L., van Poppel, H. C. A. J. M., Gribnau, F. W. J., Vossen, J. M. H., & van Luijtelaar, E. L. J. M. (1989). Benzodiazepines and cognition: Effects on intentional and incidental learning. In J. Horne (Ed.), Sleep ‘88 (pp. 201–204). Stuttgart, New York: Gustav Fischer Verlag. Coenen, A. M. L., & Van Luijtelaar E. L. J. M. (1991). Pharmacological dissociation of EEG and behavior: a basic problem in sleep-wake classification. Sleep 14, 464–465. Curran, H. V. (1999). Effects of anxiolytics on memory. Human Psychopharmacology: Clinical and Experimental, 14(S1), S72–S79. Curran, H. V., & Birch, B. (1991). Differentiating the sedative, psychomotor and amnestic effects of the benzodiazepines: A study with midazolam and the antagonist, flumazenil. Psychopharmacology, 103, 519–523. Curran, H. V., Collins, R., Fletcher, S., Kee, S. C., Woods, B., & Iliffe, S. (2003). Older adults and withdrawal from benzodiazepine hypnotics in general practice: Effects on cognitive function, sleep, mood and quality of life. Psychological Medicine, 33, 1223–1237. Fafrowicz, M., Unrug, A., Marek, T., van Luijtelaar, G., Noworol, C., & Coenen, A. (1995). Effects of diazepam

102 and buspirone on reaction times of saccadic eye movements. Neuropsychobiology, 32, 156–160. Gorissen, M. E. E., Curran, H. V., & Eling, P. A. T. M. (1998). Proactive interference and temporal context encoding after diazepam intake. Psychopharmacology (Berlin), 138, 334–343. Gorissen, M. E. E., & Eling, P. A. T. M. (1998). Dual task performance after diazepam intake: Can resource depletion explain the benzodiazepine-induced amnesia? Psychopharmacology (Berlin), 138, 354–361. Gorissen, M., Eling, P., van Luijtelaar, G., & Coenen, A. (1995). Effects of diazepam on encoding processes. Journal of Psychophysiology, 9, 113–121. Gorissen, M., Tielemans, M., & Coenen, A. (1997). Alertness and memory after sleep deprivation and diazepam intake. Journal of Psychopharmacology, 11, 233–239. Hinrichs, J. V., Ghoneim, M. M., & Mewaldt, S. P. (1984). Diazepam and memory: Retrograde facilitation produced by interference reduction. Psychopharmacology, 84, 158–162. Johnson, L. C., & Chernik, D. A. (1982). Sedative-hypnotics and human performance. Psychopharmacology, 76, 101–113. Krystal, A. D., Durrence, H. H., Scharf, M., Jochelson, P., Rogowski, R., Ludington, E., et al. (2010). Efficacy and safety of doxepin 1 mg and 3 mg in a 12-week sleep laboratory and outpatient trial of elderly subjects with chronic primary insomnia. Sleep, 33, 1553–1561. Leufkens, T. R. M., Lund, J. S., & Vermeeren, A. (2009a). Highway driving performance and cognitive functioning the morning after bedtime and middle-of-the-night use of gaboxadol, zopiclone and zolpidem. Journal of Sleep Research, 18, 387–396. Leufkens, T. R. M., Ramaekers, J. G., de Weerd, A. W., Riedel, W. J., & Vermeeren, A. (2009b). Residual effects of zopiclone 7.5 mg on highway driving in elderly insomnia patients: A placebo controlled study. In T. R. M. Leufkens (Ed.), Hypnotics and anxiolytics: Field and laboratory measures of drug safety in driving performance and cognitive functions. Thesis, Maastricht University. Leufkens, T. R. M., Ramaekers, J. G., de Weerd, A. W., Riedel, W. J., & Vermeeren, A. (2009c). On the road driving performance and driving related skills in untreated insomnia patients and chronic users of hypnotics. In T. R. M. Leufkens (Ed.), Hypnotics and anxiolytics: Field and laboratory measures of drug safety in driving performance and cognitive functions. Thesis, Maastricht University. Leufkens, T. R. M., & Vermeeren, A. (2009). Highway driving in the elderly the morning after bedtime use of hypnotics: A comparison between temazepam 20 mg, zopiclone 7.5 mg, and placebo. Journal of Clinical Psychopharmacology, 29, 432–438. Lister, R. G. (1985). The amnestic action of benzodiazepines in man. Neuroscience Biobehavioural Reviews, 9, 87–94.

Louwerens, J. W., Gloerich, A. B. M., De Vries, G., Brookhuis, K. A., & O'Hanlon, J. F. (1987). The relationship between drivers’ blood alcohol concentration (BAC) and actual driving performance during high speed travel. In P. C. Noordzij & R. Roszbach (Eds.), International congress on alcohol, drugs and traffic safety, T86 (pp. 183–186). Amsterdam: Excerpta Medica. Mayer, G., Wang-Weigand, S., Roth-Schechter, B., Lehmann, R., Staner, C., & Partinen, M. (2009). Efficacy and safety of 6-month nightly ramelteon administration in adults with chronic primary insomnia. Sleep, 32, 351–360. McAndrews, M. P., Weiss, R. T., Sandor, P., Taylor, A., Carlen, P. L., & Shapiro, C. M. (2003). Cognitive effects of long-term benzodiazepine use in older adults. Human Psychopharmacology, 18, 51–57. Neutel, C. I. (1995). Risk of traffic accident injury after a prescription for a benzodiazepine. Annals of Epidemiology, 5, 239–244. Neutel, C. I. (1998). Benzodiazepine-related traffic accidents in young and elderly drivers. Human Psychopharmacology: Clinical and Experimental, 13, 115s–123s. Nutt, D. J., & Stahl, S. M. (2010). Searching for perfect sleep: The continuing evolution of GABAA receptor modulators as hypnotics. Journal of Psychopharmacology, 24, 1601–1612. O'Hanlon, J. F. (1984). Driving performance under the influence of drugs: Rationale for, and application of, a new test. British Journal of Clinical Pharmacology, 18, 121s–129s. Paterniti, S., Dufouil, C., & Alpérovitch, A. (2002). Long-term benzodiazepine use and cognitive decline in the elderly: The epidemiology of vascular aging study. Journal of Clinical Psychopharmacology, 22(3), 285–293. Puustinen, J., Nurminen, J., Kukola, M., Vahlberg, T., Laine, K., & Kivelä, S. L. (2007). Associations between use of benzodiazepines or related drugs and health, physical abilities and cognitive function: A non-randomised clinical study in the elderly. Drugs & Aging, 24(12), 1045–1059. Quaglio, G., Lugoboni, F., Fornasiero, A., Lechi, A., Gerra, G., & Mezzelani, P. (2005). Dependence on zolpidem: Two case reports on detoxification with flumazenil infusion. International Clinical Psychopharmacology, 20, 285–287. Ramaekers, J. G. (2003). Antidepressants and driver impairment: Empirical evidence from a standard on-theroad test. The Journal of Clinical Psychiatry, 64, 20–29. Squire, L. R. (1992). Memory and the hippocampus: A synthesis from findings with rats, monkeys, and humans. Psychological Review, 99, 195–231. Theunissen, E. L., Vermeeren, A., Vuurman, E. F. P. M., & Ramaekers, J. G. (2009). Drugs, driving and traffic safety in allergic rhinitis. In J. C. Verster, S. R. PandiPemural, J. G. Ramaekers & J. J. De Gier (Eds.), Drugs, driving and traffic safety. (pp. 371–382). Basel: Birkhäuser.

103 Tiplady, B., Hiroz, J., Holmes, L., & Drummond, G. (2003). Errors in performance testing: A comparison of ethanol and temazepam. Journal of Psychopharmacology, 17, 41–49. Unrug, A., Coenen, A. M. L., & van Luijtelaar, E. L. J. M. (1997a). Effects of the tranquillizer diazepam and the stimulant methylphenidate on vigilance and memory. Neuropsychobiology, 36, 42–48. Unrug, A., van Luitelaar, E. L. J. M., Coles, M. G. H., & Coenen, A. M. L. (1997b). Event related potentials in a passive and active auditory condition: Effects of diazepam and buspirone in slow wave positivity. Biological Psychology, 46, 101–111. Van Leeuwen, T. H., Verbaten, M. H., Koelega, H. S., Camfferman, G., van der Gugten, J., & Slangen, J. L. (1994). Effects of oxazepam on eye movements and performance in vigilance tasks with static and dynamic stimuli. Psychopharmacology, 116, 499–507. Van Leeuwen, T. H., Verbaten, M. N., Koelega, H. S., Kenemans, J. L., & Slangen, J. L. (1992). Effects of bromazepam on single-trial event-related potentials in a visual vigilance test. Psychopharmacology, 106, 555–564. Van Ruitenbeek, P., Vermeeren, A., & Riedel, W. J. (2010). Cognitive domains affected by histamine H1-antgonism in humans: A literature review. Brain Research Reviews, 64, 263–282. Van Steveninck, A. L., Wallnöfer, A. E., Schoemaker, R. C., Pieters, M. S., Danhof, M., van Gerven, J. M., et al.

(1997). A study of the effects of long-term use on individual sensitivity to temazepam and lorazepam in a clinical population. British Journal of Clinical Pharmacology, 44(3), 267–275. Vermeeren, A. (2004). Residual effects of hypnotics: Epidemiology and clinical implications. CNS Drugs, 18, 297–328. Vermeeren, A., Danjou, P. E., & O'Hanlon, J. F. (1998). Effects of evening and middle of the night doses of zaleplon 10 and 20 mg on memory and actual driving performance. Human Psychopharmacology, 13, S98–S107. Vermeeren, A., Leufkens, T. R. M., & Verster, J. C. (2009). Effects of anxiolytics on driving. In J. C. Verster, S. R. PandiPemural, J. G. Ramaekers & J. J. De Gier (Eds.), Drugs, driving and traffic safety. (pp. 289–306). Basel: Birkhäuser. Wafford, K. A., & Ebert, B. (2008). Emerging anti-insomnia drugs: Tackling sleeplessness and the quality of wake time. Nature Reviews. Drug Discovery, 7, 530–540. Wilson, S. J., Nutt, D. J., Alford, C., Argyropoulos, S. V., Baldwin, D. S., Bateson, A. N., et al. (2010). British Association for Psychopharmacology consensus statement on evidence-based treatment of insomnia, parasomnias and circadian rhythm disorders. Journal of Psychopharmacology, 24, 1577–1601. Winsky-Sommerer, R. (2009). Role of GABAA receptors in the physiology and pharmacology of sleep. The European Journal of Neuroscience, 29, 1779–1794.

H. P. A. Van Dongen and G. A. Kerkhof (Eds.) Progress in Brain Research, Vol. 190 ISSN: 0079-6123 Copyright Ó 2011 Elsevier B.V. All rights reserved.

CHAPTER 6

Effects of caffeine on sleep and cognition Jan Snel{,* and Monicque M. Lorist{,} {

Department of Psychonomics, University of Amsterdam, Amsterdam, The Netherlands { Department of Experimental Psychology, University of Groningen, Groningen, The Netherlands } BCN-NeuroImaging Center, University Medical Center Groningen, Groningen, The Netherlands

Abstract: Caffeine can be used effectively to manipulate our mental state. It is beneficial in restoring low levels of wakefulness and in counteracting degraded cognitive task performance due to sleep deprivation. However, caffeine may produce detrimental effects on subsequent sleep, resulting in daytime sleepiness. This justifies a careful consideration of risks related to sleep deprivation in combination with caffeine consumption, especially in adolescents. The efficacy of caffeine to restore detrimental effects of sleep deprivation seems to be partly due to caffeine expectancy and to placebo effects. The claim that stimulant effects of caffeine are related to withdrawal or withdrawal reversal seems to be untenable. Keywords: caffeine; modafinil; staying awake; falling asleep; expectancy; withdrawal; recovery sleep; mental state. Introduction

people might take a nap, go for a walk, or put on bright lights. However, for centuries, one of the most popular means to manipulate our physiological and mental state is the use of caffeine, mostly prepared as coffee. On one hand, it is deliberately used to counteract fatigue, to stay alert, perform at satisfying levels, and postpone sleep, while on the other hand, it is intentionally avoided by many to get a good night's sleep. Caffeine is generally accepted to be a mild stimulant. It is affordable and easily available throughout the world and found in many products (see Table 1). After oral ingestion of caffeine, mostly in the form

People are continuously engaged to find the optimum of their mental and physiological state. To reach that optimum, diverse strategies are used. In the case of preparing to go to bed, for example, people relax, lights are dimmed, and some people take a nightcap. When sleepy at times that one is expected to be alert and wakeful,

*Corresponding author. Tel.: þ31 20 5256855; Fax: þ31 20 6391656 E-mail: [email protected] DOI: 10.1016/B978-0-444-53817-8.00006-2

105

106

of coffee or tea, 99% of it is absorbed from the gastrointestinal tract into the bloodstream, peaking 30–60 min after ingestion. Faster absorption of caffeine is found for caffeine-containing chewing gum, with maximum levels reached between 45 and 80 min postadministration, while absorption rate for caffeine-containing capsules lies between 85 and 120 min (Kamimori et al., 2002). Caffeine diffuses throughout the entire body; it passes all biological membranes, including the blood–brain barrier and the placental barrier. Most of the biological effects of caffeine, at levels reached during normal human consumption, are caused by way of antagonizing adenosine receptors, in particular, A1 and A2A receptors, and to a lesser degree, the A2B and A3 receptors. A1 and A2A adenosine receptors affect several mechanisms located in widespread areas of the

Table 1. Caffeine contents of common products Coffee, cup ¼ 125 ml Filtered, percolated Drip Instant Pads, dark regular Pads, mild Cappuccino Espresso cup ¼ 50 ml Decaffeinated coffee cup ¼ 125 ml Tea, cup ¼ 125 ml Soft drinks per 100 ml Cola's general Cola’ light Ice tea Energy drinks per 100 ml Chocolate containing drinks per 100 ml Chocolate/50 g Milk Dark Extra dark Chocolate candy, bars, ice cream Alcoholic drinks or shooters per 100 ml Prescription and non-predescription medication Source: Wendte et al. (2003).

Caffeine (mg) 60–100 44 35–50 90–95 75–80 60 50–60 2–4 20–45 3–11 0–15 3–12 30 2–4 2–25 8–60 30–210 2–10 50–120 25–100

brain, involved in the regulation of sleep, arousal, and cognition (Ribeiro and Sebastiao, 2010). Therefore, it is no surprise that caffeine, as an adenosine receptor antagonist, can modulate physiological and mental states (Table 2). This is supported by findings indicating that caffeine indeed attenuates the buildup of sleep propensity associated with wakefulness (Landolt, 2008b), although in rats, it was found (Wurts and Edgar, 2000) that caffeine did not block compensatory non-rapid eye movement (non-REM) sleep and sleep continuity. Moreover, it potently attenuates EEG markers of non-REM sleep homeostasis during sleep, as well as during wakefulness (Landolt, 2008b). Targeting the adenosine system by caffeine consumption therefore seems an effective tool to modulate individual vulnerability to the detrimental effects of sleep deprivation on cognitive performance, and sleep. To be able to determine the effect of caffeine in people, individual differences have to be taken into account. Metabolic rate and the tolerance for effects of caffeine vary considerably from one person to the other. The half-life of caffeine is on average 3.7 h, ranging from 2 to 10 h, dependent on endogenous and exogenous factors. For example, metabolic speed of caffeine is increased by 30–50% in nicotine users, whereas a decrease is observed in pregnant women and woman taking oral contraceptives. Also there are large interindividual differences in sensitivity to caffeine due to genetic variations in the adenosine A2A receptor gene (Retey et al., 2007), the role adenosine and adenosine receptors play in non-REM sleep homeostasis, (Landolt, 2008b) and genotype-dependent differences in sleep (Landolt, 2008a). However, no systematic difference in the metabolism of caffeine is observed between men and women. Besides the purposive use or avoidance of caffeine, it is frequently consumed unintentionally. People are not always aware that caffeine is added to many products to increase flavor and taste. The awareness (or unawareness) of caffeine consumption has important implications, not only for our well-being, but also for caffeine research.

107 Table 2. Central adenosine receptors affected by typical caffeine exposure

Receptor

Localization

Types of neurons

A1

Almost all brain areas, especially hippocampus, cerebral and cerebellar cortex, certain thalamic nuclei Dopamine-rich regions: striatum, nucleus accumbens, tuberculum olfactorium, hippocampus? cortex?

All types of neurons (aspecific), especially linked to dopamine D1 receptors Colocalized with dopamine D2 receptors

A2A

Effects of caffeine

Caffeine action

Antagonistic

Disinhibition of transmitter release

Antagonistic

Increase transmission via dopamine D2 receptors

Source: Lorist and Tops (2003).

It is important to realize that estimating habitual caffeine intake is difficult and most likely results in lower estimates than actually consumed. Hence, the selection of participants based on their self-reported daily consumption of caffeine is often biased and unreliable. Careful screening of participants, who were selected on the basis of their habitual intake of 100–500 mg caffeine a day, by Wendte et al. (2003) revealed an actual caffeine consumption of 154–1285 mg ( 1.5–15 cups of coffee); quantities which were up to 250% higher than the self-reported values. In addition to the underreporting of consumed caffeine quantity, factors such as differences in brewing method, used coffee blend (Arabica coffee contains 2% of caffeine, Robusta coffee 4%), or serving size of caffeine-containing food hamper adequate estimation of caffeine intake. In this chapter, effects of caffeine are described; in particular, effects on sleep–wake rhythmicity in sleep-deprived versus well-rested individuals. Caffeine and sleep deprivation A 24-h economy demands individuals to operate at times which are not “in sync” with their circadian clock. As a result, daily rhythms can become disrupted and, consequently, negatively affect our well-being. If the timing of sleep is not adapted to the circadian clock, this misalignment may result in a so-called circadian rhythm sleep disorder.

Disturbances of our circadian rhythmicity may cause disruptions in sleeping patterns and influence cognitive task performance, which can lead to suboptimal performance and even errors (Crochet et al., 2009; Ker et al., 2010). Jet lag and shift work are important factors that contribute to these deteriorations of performance and have indeed been found to be related to an increase in risk of injury (as discussed elsewhere in this volume). There is a strong need for interventions to guarantee that persons who run such risk can do their job safely and are able to restore disturbed circadian rhythms as soon and efficiently as possible. Different pharmacological aids are used to counteract work-related sleep problems and jet lag, and the question is whether caffeine might qualify as one of them (Coste and Lagarde, 2009). Evidence suggests that sleep-loss-induced deficits in alertness and vigilance can indeed be reversed or mitigated by stimulants such as caffeine (Lorist and Snel, 2008; Snel et al., 2004). Sleep deprivation has been used in different studies to elucidate the role of caffeine in offsetting these effects. Benitez et al. (2009), for example, used a severe form of sleep deprivation in 14 males and 4 females (mean age 25.8 years, SD ¼ 4.3), all mild or non-caffeine users (< 300 mg/day). Participants had a normal 8-h night of sleep, followed by 77 h of continuous wakefulness. Placebo or caffeine (200 mg), in the form of two sticks of Alert gum, containing 100 mg of caffeine each, was assigned randomly at 1.00, 3.00, 5.00, and 7.00 a.m. during the

108

three subsequent nights. Performance was tested periodically over the 77-h period of continuous wakefulness, using the psychomotor vigilance test (PVT). The PVT is a simple visual reaction task, used to measure sustained attention. It measures the time it takes to respond to a visual stimulus. A biomathematical model was build to describe performance during the period of extended wakefulness. This model identified patterns in the data that suggested the presence of a performance inhibitor, called fatigue, that increased and saturated over the 77 h of continuous wakefulness. Caffeine was found to be able to inhibit the effects of fatigue. This result confirms the findings of former research (Balkin et al., 2004; Wesensten et al., 2002, 2004, 2005) with simple psychomotor tasks and tasks of executive functions, indicating that caffeine taken both incidentally or with the purpose of counteracting drowsiness in the morning, is effective in maintaining alertness and performance after prolonged periods of total sleep deprivation (i.e., 54.6–85 h). The efficacy of caffeine in restoring cognitive processes like emotional perception, judgment, risk-taking, and planning after sleep deprivation was targeted in a series of double-blind studies by Killgore et al. (2009). They studied the effects of caffeine in 54 participants (age 18–36 years), who were sleep deprived for 45–50 h and subsequently tested. Before performing different cognitive tasks, they received one dose of 600 mg caffeine (n ¼ 12), 400 mg modafinil (n ¼ 12), 20 mg dextroamphetamine (n ¼ 16), or placebo (n ¼ 14). Like caffeine, modafinil promotes alertness and wakefulness. Modafinil is especially FDA approved to manage fatigue in narcolepsy and residual fatigue in sleep apnea and shift work sleep disorder (Rosenthal et al., 2008). Dextroamphetamine also has stimulant properties, known to promote wakefulness. It has been provided to pilots on long missions to help them remain focused and alert. However, dextroamphetamine can produce some side effects like palpitations, tachycardia, and elevated blood pressure (Caldwell and Caldwell, 2005). Modafinil has a

disadvantage that it can be used only after informed consent. Participants in the Killgore et al. (2009) study were administered a test to measure the formation of abstract concepts 1 h after drug administration (i.e., after 45 h awake). Tests measuring planning time and cognitive processing started 3.5 h (i.e., 47.5 h awake) and 4.5 h after drug administration (i.e., after 48.5 h awake). The results of these studies showed that the three stimulants differentially affected the outcome of the three cognition tasks, suggesting that the stimulating effects of caffeine especially affected cognitive planning processes. It seems clear that performance efficiency is affected by sleep deprivation and part of the performance deteriorations can indeed be counteracted by caffeine or other stimulants. However, it is important to realize that task performance not only relies on specific task demands and related cognitive abilities. The interaction between task, participant, and environment actually determines the quality of performance, especially during suboptimal situations. It is clear that some tasks are more challenging than others, and related changes in motivation might mask fatigue effects created by sleep deprivation. Fatigue is a well-known and common phenomenon in sustained operations, such as long-distance driving and long-term continuous work (see Chapters 9 and 11 of this volume), in which a low information load does not promote motivation to perform. In many of such real-life situations, a sufficient level of motivation to work is required to perform adequately. Kilpelaiinen and colleagues studied the effect of caffeine and placebo on sustained attention and learning in such a vigilance situation. They also assessed subjective ratings of sleepiness, mood, motivation, and perceived task performance in their study (Kilpelaiinen et al., 2010). Fifteen military pilot students (age 23–24 years) took part in a series of tests in a flight simulator, in the 37 h sleep deprivation study. They received either placebo or 200 mg of caffeine twice a day (Kilpelaiinen et al., 2010). During the experiment, vigilance

109

was assessed six times (Mackworth clock test) and learning four times. The learning task consisted of learning the association between Japanese hiragana symbols and spoken syllables. Learning was tested 2 h later with five hiraganas from the previous learning session. As expected, sleep deprivation decreased the amount of correct detections and increased reaction times in both the caffeine and placebo groups. The increase in number of false alarms was limited to the placebo group. Working memory, as tested in the learning task, remained unaffected during sleep deprivation. Kilpelaiinen et al. (2010) argued that the absence of performance decline in the learning task might partly be due to the stimulating nature of this task following a very exhausting and longlasting vigilance task. With increasing sleep pressure, they observed that participants even wanted to perform the learning task during task breaks to help them stay awake, illustrating the importance of taking into account the environmental factors in sustained performance and sleep deprivation studies. Although a clear deterioration in vigilance performance was observed after sleep deprivation, subjective feelings of success remained stable across sustained wakefulness in the caffeine group. The feeling of success was measured with the phrase “How successful do you think you were in this task?” which was rated on a visual analog scale running from “not at all” to “very much.” Similarly, Baranski (2007) in his study on confidence in judgment found that one night of sleep deprivation did not result in an impaired assessment of cognitive performance. It should be noted that the overconfidence in caffeine participants might have serious consequences in real-life work environments, like in aviation, because realistic self-perception is highly important in avoiding risks. Caffeine and recovery sleep There is little doubt among laymen and health professionals about the fundamental importance of sufficient, restorative sleep in maintaining

one's physical and mental health. Maintaining a good sleep quality involves avoidance of substances that stimulate mind and body and disturb sleep habits. Caffeine, with its proven efficacy to counteract sleepiness, is one of the stimulants that may produce detrimental effects on subsequent sleep, especially when sleep is initiated at a time when the biological clock sends a strong waking signal as happens during daytime. This means a prudent use of caffeine. In other words, using caffeine at times that high mental alertness and physical activation hampers sleep quality should be avoided, in particular, in the hours shortly before going to sleep. In people working shift hours, sleep habits have to be adapted to their irregular working schedule, which might cause specific problems in combination with caffeine consumption. Pecotic et al. (2008) evaluated sleep habits and explored whether these were influenced by caffeine consumption in 130 medical students, 68 physicians at the postgraduate study program, 162 specialists, and 93 nurses. Results indicated that the hours of sleep needed for feeling well rested depended on age, gender, work demands, and work schedule. However, respondents who consumed caffeine reported more trouble staying awake while listening to lectures or learning and while driving a car. Based on these results, the authors argued that caffeine consumption may impair sleep habits and quality of sleep, thereby hampering cognitive performance and wakefulness during nonsleep hours. Whether caffeine use is consistently the cause of impaired sleep in everyday situation and not vice versa is questionable (Orbeta et al., 2006; Roehrs and Roth, 2008; Whalen et al., 2007). It was found that feeling tired in the morning induced high caffeine use which was associated with subsequent impaired sleep, indicating that cause and effect concerning the relation between caffeine and impaired sleep is not always clear. Another factor that may induce sleep disturbance in interaction with caffeine is vulnerability to stress (Drake et al., 2006). Drake and colleagues

110

showed that individuals with an objectively identified vulnerability to stress-induced sleep disturbance showed a stronger sleep-reactivity in response to a 3-mg/kg BW caffeine challenge than nonvulnerable individuals. The effects of caffeine on daytime recovery sleep after 25 h of sleep deprivation were studied by Carrier et al. (2009). Young (age 20–30 years) and middle-aged volunteers (age 45–60 years) participated in a caffeine (200 mg) and placebo condition, spaced 1 month apart. Three hours before daytime recovery sleep, the participants took either the first 100 mg caffeine or placebo capsule, while the second 100 mg dose was taken 1 hour before daytime recovery sleep. An effect of caffeine on daytime recovery sleep was observed in both age groups, reflected in a decrease in sleep efficiency, sleep duration, slowwave sleep, and REM sleep. Moreover, caffeine reduced non-REM sleep EEG synchronization during daytime recovery sleep. These results provide additional evidence that using caffeine to cope with night work and jet lag might result in detrimental effects on subsequent sleep. However, no detrimental effects on recovery sleep were observed in the Philip et al. (2006) study. These authors examined nighttime driving performance between 2.00 and 3.30 a.m. after placebo, 30 min of napping, or 200 mg caffeine. An important difference between the Philip et al. (2006) study and studies in which a clear effect of caffeine on recovery sleep was observed might be the length of the sleep deprivation period used. Carrier et al. (2009), for example, exposed their participants to 25 h of sleep deprivation, while participants in the Philip et al. (2006) study were allowed to go to sleep immediately after the nighttime driving session (i.e., after 3.30 a.m.). Besides the duration of the sleep deprivation period, circadian rhythm might be another important factor that has to be taken into account in explaining differential effects of caffeine on recovery sleep. The participants in the Carrier et al. (2009) study were instructed to maintain a regular sleep–wake schedule before

the experimental session and the experimenters monitored these individuals during the 3 days before each experimental session to verify the absence of sleep deprivation. Effects of caffeine on sleep variables in individuals working night shifts or suffering from jet lag may be more pronounced because of stronger influences of circadian rhythms and related physiological processes in these individuals. Effects of caffeine use on subsequent sleeping patterns have been related to its pharmacological actions. Although it is widely accepted that the predominant effect of caffeine is to block specific adenosinergic receptors, other mechanisms may play a role, as Ataka et al. (2008) presented in their study on candidate antifatigue substances on mental fatigue. Caffeine, used to increase alertness and wakefulness, might affect sleep by an effect on branched-chain amino acids. Branched-chain amino acids are used for the synthesis of proteins and are regarded as a biomarker of mental fatigue. Ataka et al. (2008) examined levels of these amino acids in 17 healthy participants who randomly received 100 mg/day caffeine or placebo twice a day for 8 consecutive days. Fatigue was induced by mental task performance (Uchida-Kraepelin psychodiagnostic test and advanced trail-making test). Task performance of the caffeine group was better than performance observed in the placebo group. However, subjective perception of fatigue, motivation, and sleepiness did not differ between both groups. These results are in line with the findings of Kilpelaiinen et al. (2010), suggesting that administration of caffeine improves task performance without decreasing the sensation of fatigue. An important observation of Ataka et al. (2008) was that plasma branched-chain amino acid levels in the caffeine group were lower than those observed in the placebo group, after the fatigue-inducing mental tasks, as well as after the recovery period (as a trend). These results indicate that caffeine can accelerate mental fatigue through increased activation of the brain, without an accompanying sensation of

111

increased fatigue. Bartley and Chute (1947) argued that mental fatigue might be regarded as a protection mechanism warning an individual that he or she needs a rest in order to prevent an overload of our cognitive system. Results of Ataka et al. (2008) imply that this warning mechanism, essential in reducing acute fatigue, is less effective after caffeine consumption. This might explain why caffeine used during daytime, particularly at later times of day, may have detrimental effects on subsequent sleep periods. Caffeine and self-imposed sleep deprivation Sleep disturbances have been associated with an increased risk of work absenteeism, decrements in vitality, social functioning, physical and mental health, and general quality of life (Lund et al., 2010). In young people, who tend to have irregular life styles and consequently do not get the sleep they need, sleep problems might give rise to academic problems. Caffeine can be used strategically to optimize the level of wakefulness, thereby improving daytime functioning in this group. It is important, though, to realize that nonjudicious use of caffeine may intensify their sleeping problems, since there is evidence for a greater physiological need for sleep in adolescents compared to other age groups. Lund et al. (2010) examined sleep quality in college students (age 17–24 years). Disturbed sleep was reported by 60% of the 1125 students, who completed a cross-sectional online survey about sleep habits. Students overwhelmingly stated that emotional stress and academic stress were important factors that negatively impacted sleep and explained 24% of the variance of the scores on the Pittsburgh Sleep Quality Index. It was noteworthy that caffeine consumption, consistency of sleep schedule, and daily hours of television and video game exposure were no significant predictors of sleep quality (Lund et al., 2010). A different pattern of results was observed by Calamaro et al. (2009), who performed a study

in middle and high school students (age 12–18 years) to examine the relation between technology use (e.g., watching television, text messaging, playing video games, surfing internet), caffeine intake, and quality of sleep. Calamaro et al. hypothesized that with increased technology use, especially late at night, more caffeine is consumed to stay awake. This behavior pattern, coupled with the early start times for middle schools and high schools that demand earlier weekday rise times, was expected to result in sleep deficits. The results showed that the hours spent with technology activities at night were indeed related to a decrease in sleep duration. Not surprisingly, the ability to stay alert and function adequately during the subsequent day was impaired by excessive daytime sleepiness in those students who got less sleep. In addition, caffeine consumption tended to be 76% higher in this group. Important implications of such a strategy concern the additional negative effects on nighttime sleep when trying to compensate daytime sleepiness by taking caffeine (Orbeta et al., 2006, Roehrs and Roth, 2008; Whalen et al., 2007). Altogether, these results warrant a careful consideration of risks related to sleep deprivation in combination with caffeine consumption, especially in middle and high school age groups. Caffeine, sleepiness, and work quality Caffeine promotes alertness during times of desired wakefulness in persons with jet lag or shift work disorder (for a review see Lorist and Snel, 2008). An important benefit of the effects of caffeine could be a reduced risk on injury and error during these periods. Tieges et al. (2004) showed that doses of 3 and 5 mg/kg body weight of caffeine in well-rested habitual caffeine consumers indeed reduced the number of errors compared to a placebo condition. Based on related changes in brain activity shown as enlarged error-related negativity, an event-related brain component that reflects anterior cingulate cortex activity, they

112

concluded that coffee consumption increased monitoring of ongoing cognitive processes for signs of erroneous outcomes. The observed reduction in self-consciousness resulting in overconfidence after caffeine consumption (Kilpelaiinen et al., 2010) seems to contradict these findings, and might have consequences in real-life environments. Realistic self-perception is essential in avoiding risks. The important question whether under suboptimal conditions, for example, due to extended wakefulness, caffeine is still able to enhance the detection of erroneous responses and consequently minimize the risk of errors, was extensively studied in recent years (e.g., Lorist and Snel, 2008). Ker et al. (2010) examined more specifically the role of caffeine in preventing errors or injuries caused by impaired alertness in individuals with jet lag or shift work. Their systematic review did not elucidate a relation between caffeine and injuries because none of the studies included measured injury outcomes. Concerning the occurrence of errors, Ker and colleagues observed that caffeine significantly reduced the number of errors compared to placebo. One of the studies examined by Ker et al. (2010) was a study by Dagan and Doljansky (2006). These researchers evaluated the efficacy of caffeine (200 mg) and modafinil (200 mg) in maintaining cognitive performance after sleep deprivation in a flight simulation task. According to their results, both caffeine and modafinil significantly decreased the deviation from assigned altitude and velocity values compared to baseline levels during the nocturnal drop in cognitive performance, supporting that caffeine has a performance-increasing effect, especially under suboptimal conditions. A second study reviewed by Ker et al. (2010) was performed by Philip et al. (2006). These authors measured nighttime highway driving performance of 12 young men immediately after 200 mg of caffeine, decaffeinated coffee (containing 15 mg of caffeine), and after napping in the car for 30 min. An increase was found in line crossings during nighttime driving compared with

the daytime driving session. Lateral deviations have been found to be a frequent cause of sleeprelated accidents. If daytime highway driving (between 6.00 p.m. and 7.30 p.m.) was used as a point of reference for nighttime driving (between 2.00 and 3.30 a.m.), no difference in driving performance was observed in 75% of the participants who consumed caffeine and 66% of them drove as well after a nap, indicating that drinking coffee or napping significantly reduced line crossing errors. These results illustrate the common practice that caffeine is used as an efficient countermeasure for sleep-related accidents which are known to occur most frequently in the middle of the night. It is important to note that similar effects have also been found in well-rested individuals (Attwood et al., 2006; Childs and de Wit, 2006; Haskell et al., 2005; Hewlett and Smith, 2007). A common comment on caffeine's ability to improve performance is that little account is taken of the fact that caffeine withdrawal and withdrawal reversal might possibly obscure the net effects of caffeine. In a great deal of the experimental studies on the effects of caffeine, researchers have used the naturally occurring overnight caffeine abstinence period. In addition, participants are asked to abstain from their usual morning caffeine consumption prior to laboratory testing, and caffeine consumption is delayed until the experimental session. Improvements in performance following caffeine ingestion under these circumstances could reflect reversal of the adverse effects of the overnight caffeine withdrawal. It is known that due to regular caffeine intake, the number of adenosine receptors in the central nervous system increases. As a result of this adaptive caffeine response, individuals become more sensitive to adenosine. A subsequent reduction of intake of caffeine as a blocker of adenosinergic receptors will increase the normal physiological effects of adenosine, resulting in withdrawal symptoms in tolerant caffeine users. Withdrawal symptoms, including headache, irritability, and an inability to concentrate, usually appear within

113

12–24 h after discontinuation of caffeine intake, peaking around 48 h (Juliano and Griffiths, 2004). These effects last from 1 to 5 days, representing the time required for the number of adenosine receptors in the brain to revert to “normal” levels, uninfluenced by caffeine consumption. Keane and James (2008) examined the chronic effects of caffeine consumption in 15 healthy individuals (age 17–19 years). Participants alternated weekly between ingesting placebo and caffeine (1.75 mg/kg) three times daily for 4 consecutive weeks following either usual sleep or sleep restriction (40% of the usual amount). The effects of caffeine on brain activity, performance, and mood were examined after 6 days in which caffeine was consumed or after 6 days in which participants did not use caffeine-containing substances. The authors argued that the 6-day period warrants that the effects observed in the abstinence condition cannot be ascribed to withdrawal-related effects, since these effects last up to 5 days. Their results showed that the effects of caffeine on EEG activity were trivial and inconsistent, and no clear evidence was found of restorative effects of caffeine for performance and mood variables. They argued that caffeine is of no use to enhance human function or to reverse the negative effects of sleep loss. Other studies, however, did show increases in participants who were well-rested and not deprived of caffeine (e.g., Attwood et al., 2006; Childs and de Wit, 2006; Haskell et al., 2005; Hewlett and Smith, 2007). The lack of clear results in the Keane and James (2008) study might be related to the sleep restriction protocol they used. Keane and James did not observe changes in performance scores due to the sleep restrictions. In addition, no main effects were observed on EEG power, suggesting that the sleep restriction manipulation might have been too weak or the dependent measure too insensitive to caffeine effects to induce differential effects of caffeine between the usual sleep condition and the sleep restriction condition.

In order to avoid confounding of caffeine effects with tolerance or withdrawal, Michael et al. (2008) examined the effects of caffeine in 12 participants (age 18–29 years), who consumed either no caffeine or very little caffeine on a regular basis. The wellrested participants were tested on 2 separate days, using vigilance tests scheduled at baseline (around 9.00 a.m.) and at 30, 60, 120, 180, and 240 min after placebo or caffeine (200 mg) administration. During task performance, eye blink variables were measured to assess alertness. In contrast with the outcome of the Keane and James (2008) study, the result of Michael et al. (2008) showed that, even though the participants were well rested, caffeine was able to reduce drowsiness, as deduced from ocular movements and reaction times, and these changes persisted for 3–4 h. The use of different ocular variables seemed to provide a sensitive measure to detect subtle changes in alertness induced by caffeine. Self-reports of sleepiness were not as sensitive; differences between the caffeine and placebo condition were only observed 30 min after substance administration. These findings support the conclusion that caffeine can have beneficial effects on performance and alertness (Michael et al., 2008), and that these effects do not seem to be related to withdrawal or withdrawal reversal since it is unlikely that tolerance has developed in the individuals who did not consume caffeine on a regular base. The general conclusion, so far, seems clear; caffeine can be used effectively to manipulate mental state. It was found to be beneficial in restoring low levels of wakefulness and counteracting degraded task performance. However, caffeine may produce detrimental effects on subsequent sleep, resulting in daytime sleepiness. Remaining issues Expectancy Caffeine is generally regarded as a stimulant, frequently used to make people feel more alert and ready to face daily challenges and to counteract

114

sleepiness. It is assumed that these properties are the main reason for consuming caffeine in a broad range of daily life situations and in research, as well. The stimulant effects of caffeine on physiological and mental states seem to profoundly be mediated by its pharmacological actions as an adenosine receptor antagonist. Van Dongen et al. (2001), for example, found in their participants, who were sleep deprived for 88 h and who received sustained low caffeine doses (0.3 mg/kg BW/h) during the last 66 h, that inertia was largely overcome by caffeine. They concluded that the caffeine-induced antagonism of adenosine receptors on the central nervous system results into an increase of unused adenosine in the brain upon awakening which might be the cause of sleep inertia. However, expectations of caffeine (i.e., placebo) effects have been found to be an important additional factor to its psychostimulant effects. These expectations can trigger a series of physiological and psychological reactions, usually related to the pharmacological effects of caffeine consumption. Sun et al. (2007), for example, studied the effect of caffeine expectations on vigilance and cognitive task performance during 28 h of sleep deprivation. They informed 10 healthy male participants (age 18–20 years) that the capsules they had to ingest contained caffeine and gave them information about the stimulating effects of caffeine to increase the expectation of caffeine effects. Participants arrived at the laboratory at 6.00 a.m. and went through 28 successive hours of total sleep deprivation. Cognitive tests (letter cancelation task, continuous addition test) were administered every 2 h from 12.00 to 10.00 a.m. of the second day. Sun et al. (2007) found that an initial dose of 200 mg caffeine, administered at 12.00 a.m., followed 4 h later by a placebo helped to maintain cognitive performance during the period without sleep to a similar extent as the condition in which a double dose of caffeine was given. The placebo effect extended the cognitive boost without incurring the rise in blood pressure and heart rate that sometimes go together with caffeine consumption. This result

suggests that this caffeine-plus-placebo regimen could be used when work schedules demand extended periods of alertness without sleep to maximize attention but minimize negative side effects. Caffeine expectations were also examined in 16 young healthy volunteers (age 18–25 years) by Anderson and Horne (2008). In their study, participants performed a three-times 30-min PVT (separated by a 2-min break after every 30 min) after they had a light lunch; during this period, an early afternoon “dip” is usually experienced. Sleepiness was further enhanced by requiring participants to limit their prior night's sleep to 5 h. Participants were tested twice, either after they consumed a cup of decaffeinated coffee which was accompanied by verbal information that the coffee was decaffeinated (control) or after they consumed decaffeinated coffee after the experimenter informed them about the highly alerting effects of the “super” type coffee they were going to receive (expectancy). Significantly, fewer lapses and shorter reaction times were observed in the expectancy condition than after the control condition during the first hour of task performance, indicating that expectancy about consuming caffeine was effective in improving performance or preventing performance decrement in moderately sleepy people. It is surprising that while caffeine can take 30–40 min to become pharmacologically effective, the effects of expectancy seemed more rapid. Anderson and Horne (2008) argued that the effects of caffeine expectations might be related to classical conditioning, that is, the expectancy effect is a conditioned response. Moreover, this effect might have been enhanced by knowledge of the effects of caffeinated beverages. Support for a role of classical conditioning in caffeine expectancy was provided by Attwood et al. (2008). They examined whether the effects of caffeine could be conditioned to the context of administration in 16 volunteers (age 18–26 years). Four conditioning trials were followed by a test session, in which participants received placebo before performing a simple reaction time task. Attwood et al. observed that

115

in the test session, the group who received 250 mg caffeine during the conditioning trials performed significantly faster during the test session than the group who had received placebo, indicating the development of a conditioned response. Possible involvement of expectancy effects in this study was reduced by presenting caffeine in a novel drink, and by not informing the participants about the contents of the beverage, making the finding all the more salient. Caffeine expectancy effects are not limited to cognitive task performance but may be also present in physical task performance. Pollo et al. (2008) observed across participants who thought they consumed caffeine, that mean muscle work which was not accompanied by a decrease of perceived muscle fatigue, was regarded as evidence that caffeine expectancy could indeed counteract the symptoms of physical fatigue. In sum, the effects elicited after administration of a placebo instead of caffeine illustrate that the efficacy of caffeine to restore detrimental effects of sleep deprivation on performance is partly due to caffeine expectancy. Conclusions Caffeine, a well-known antagonist of adenosinergic receptors, can be used effectively to modulate our mental state. Caffeine is found to be beneficial in restoring low levels of wakefulness and to counteract deteriorations in task performance related to sleep deprivation. However, the results also indicate that caffeine may produce detrimental effects on subsequent sleep, resulting in daytime sleepiness. The efficacy of caffeine to restore mental performance decline in suboptimal conditions seems to be partly due to caffeine expectancy. No support was found for the claim that stimulant effects of caffeine are related to withdrawal or withdrawal reversal. In conclusion, caffeine provides an adequate and common way to strategically adjust mental state, provided the effects on recovery sleep are taken into consideration.

References Anderson, C., & Horne, J. A. (2008). Placebo response to caffeine improves reaction time performance in sleepy people. Human Psychopharmacology, 23(4), 333–336. Ataka, S., Tanaka, M., Nozaki, S., Mizuma, H., Mizuno, K., Tahara, T., et al. (2008). Effects of oral administration of caffeine and d-ribose on mental fatigue. Nutrition, 24(3), 233–238. Attwood, A. S., Terry, P., & Higgs, S. (2006). Exploring factors that mediate responsiveness to caffeine. Appetite, 47, 258. Attwood, A., Terry, P., & Higgs, S. (2008). Evidence of conditioned cognitive and mood effects of caffeine in humans using a differential conditioning paradigm. Appetite, 50(2–3), 556. Balkin, T. J., Kamimori, G. H., Redmond, D. P., Vigneulle, R. M., Thorne, D. R., Belenky, G., et al. (2004). On the importance of countermeasures in sleep and performance models. Aviation Space and Environmental Medicine, 75(3 Suppl.), A155–A157. Baranski, J. V. (2007). Fatigue, sleep loss, and confidence in judgment. Journal of Experimental Psychology: Applied, 13 (4), 182–196. Bartley, S. H., & Chute, E. (1947). Fatigue and impairment in man. London: McGraw-Hill. Benitez, P. L., Kamimori, G. H., Balkin, T. J., Greene, A., & Johnson, M. L. (2009). Modeling fatigue over sleep deprivation, circadian rhythm, and caffeine with a minimal performance inhibitor model. Methods in Enzymology, 454, 405–421. Calamaro, C. J., Mason, T. B. A., & Ratcliffe, S. J. (2009). Adolescents living the 24/7 lifestyle: Effects of caffeine and technology on sleep duration and daytime functioning. Pediatrics, 123(6), e1005–e1010. Caldwell, J. A., & Caldwell, J. L. (2005). Fatigue in military aviation: An overview of us military-approved pharmacological countermeasures. Aviation Space and Environmental Medicine, 76(Suppl. 7), C39–C51. Carrier, J., Paquet, J., Fernandez-Bolanos, M., Girouard, L., Roy, J., Selmaoui, B., et al. (2009). Effects of caffeine on daytime recovery sleep: A double challenge to the sleepwake cycle in aging. Sleep Medicine, 10(9), 1016–1024. Childs, E., & de Wit, H. (2006). Subjective, behavioral, and physiological effects of acute caffeine in light, nondependent caffeine users. Psychopharmacology (Berlin), 185(4), 514–523. Coste, O., & Lagarde, D. (2009). Clinical management of jet lag: What can be proposed when performance is critical? Travel Medicine and Infectious Disease, 7(2), 82–87. Crochet, P., Aggarwal, R., Misha, A., Sirimanna, P., & Darzi, A. (2009). Caffeine and taurine reverse the deterioration in laparoscopic and cognitive skill following sleep

116 deprivation. Journal of the American College of Surgeons, 209(3), S111. Dagan, Y., & Doljansky, J. T. (2006). Cognitive performance during sustained wakefulness: A low dose of caffeine is equally effective as modafinil in alleviating the nocturnal decline. Chronobiology International, 23(5), 973–983. Drake, C. L., Jefferson, C., Roehrs, T., & Roth, T. (2006). Stress-related sleep disturbance and polysomnographic response to caffeine. Sleep Medicine, 7(7), 567–572. Haskell, C. F., Kennedy, D. O., Wesnes, K. A., & Scholey, A. B. (2005). Cognitive and mood improvements of caffeine in habitual consumers and habitual non-consumers of caffeine. Psychopharmacology (Berlin), 179(4), 813–825. Hewlett, P., & Smith, A. (2007). Effects of repeated doses of caffeine on performance and alertness: New data and secondary analyses. Human Psychopharmacology: Clinical and Experimental, 22(6), 339–350. Juliano, L. M., & Griffiths, R. R. (2004). A critical review of caffeine withdrawal: Empirical validation of symptoms and signs, incidence, severity, and associated features. Psychopharmacology (Berlin), 176(1), 1–29. Kamimori, G. H., Karyekar, C. S., Otterstetter, R., Cox, D. S., Balkin, T. J., Belenky, G. L., et al. (2002). The rate of absorption and relative bioavailability of caffeine administered in chewing gum versus capsules to normal healthy volunteers. International Journal of Pharmaceutics, 234(1–2), 159–167. Keane, M. A., & James, J. E. (2008). Effects of dietary caffeine on EEG, performance and mood when rested and sleep restricted. Human Psychopharmacology: Clinical and Experimental, 23(8), 669–680. Ker, K., Edwards, P. J., Felix, L. M., Blackhall, K., & Roberts, I. (2010). Caffeine for the prevention of injuries and errors in shift workers. Cochrane Database of Systematic Reviews, 5, CD008508. Killgore, W. D., Kahn-Greene, E. T., Grugle, N. L., Killgore, D. B., & Balkin, T. J. (2009). Sustaining executive functions during sleep deprivation: A comparison of caffeine, dextroamphetamine, modafinil, lab. Sleep, 32(2), 205–216. Kilpelaiinen, A. A., Huttunen, K. H., Lohi, J. J., & Lyytinen, H. (2010). Effect of caffeine on vigilance and cognitive performance during extended wakefulness. The International Journal of Aviation Psychology, 20(2), 144–159. Landolt, H. P. (2008a). Genotype-dependent differences in sleep, vigilance, and response to stimulants. Current Pharmaceutical Design, 14(32), 3396–3407. Landolt, H. P. (2008b). Sleep homeostasis: A role for adenosine in humans? Biochemical Pharmacology, 75(11), 2070–2079. Lorist, M. M., & Snel, J. (2008). Caffeine, sleep and quality of life. In D. Streiner, S. R. Pandi-Perumal & J. C. Verster (Eds.), Sleep and quality of life in clinical medicine. USA: Humana Press, Springer, 978-1-60327-340-4.

Lorist, M. M., & Tops, M. (2003). Caffeine, fatigue, and cognition. Brain and Cognition, 53(1), 82–94. Lund, H. G., Reider, B. D., Whiting, A. B., & Prichard, J. R. (2010). Sleep patterns and predictors of disturbed sleep in a large population of college students. The Journal of Adolescent Health, 46(2), 124–132. Michael, N., Johns, M., Owen, C., & Patterson, J. (2008). Effects of caffeine on alertness as measured by infrared reflectance oculography. Psychopharmacology (Berlin), 200 (2), 255–260. Orbeta, R. L., Overpeck, M. D., Ramcharran, D., Kogan, M. D., & Ledsky, R. (2006). High caffeine intake in adolescents: Associations with difficulty sleeping and feeling tired in the morning. The Journal of Adolescent Health, 38(4), 451–453. Pecotic, R., Valic, M., Kardum, G., Sevo, V., & Dogas, Z. (2008). Sleep habits of medical students, physicians and nurses regarding age, sex, shift work and caffein consumption. Lijecnicki Vjesnik, 130(3–4), 87–91. Philip, P., Taillard, J., Moore, N., Delord, S., Valtat, C., Sagaspe, P., et al. (2006). The effects of coffee and napping on nighttime highway driving: A randomized trial. Annals of Internal Medicine, 144(11), 785–791. Pollo, A., Carlino, E., & Benedetti, F. (2008). The top-down influence of ergogenic placebos on muscle work and fatigue. The European Journal of Neuroscience, 28(2), 379–388. Retey, J. V., Adam, M., Khatami, R., Luhmann, U. F., Jung, H. H., Berger, W., et al. (2007). A genetic variation in the adenosine a2a receptor gene (adora2a) contributes to individual sensitivity to caffeine effects on sleep. Clinical Pharmacology and Therapeutics, 81(5), 692–698. Ribeiro, J. A., & Sebastiao, A. M. (2010). Caffeine and adenosine. Journal of Alzheimer's Disease, 20(Suppl. 1), 3–15. Roehrs, T., & Roth, T. (2008). Caffeine: Sleep and daytime sleepiness. Sleep Medicine Reviews, 12(2), 153–162. Rosenthal, T. C., Majeroni, B. A., Pretorius, R., & Malik, K. (2008). Fatigue: An overview. American Family Physician, 78(10), 1173–1179. Snel, J., Lorist, M. M., & Tieges, Z. (2004). Coffee, caffeine and cognitive performance. In A. Nehlig (Ed.), Coffee, tea, chocolate and the brain (pp. 53–73). (1st ed.). Boca Raton: CRC Press LLC. Sun, Y., Zhang, Y., He, N., Liu, X., & Miao, D. (2007). Caffeine and placebo expectation: Effects on vigilance, cognitive performance, heart rate, and blood pressure during 28 hours of sleep deprivation. Journal of Psychophysiology, 21(2), 91–99. Tieges, Z., Richard Ridderinkhof, K., Snel, J., & Kok, A. (2004). Caffeine strengthens action monitoring: Evidence from the error-related negativity. Brain Research. Cognitive Brain Research, 21(1), 87–93. Van Dongen, H. P., Price, N. J., Mullington, J. M., Szuba, M. P., Kapoor, S. C., & Dinges, D. F. (2001). Caffeine eliminates psychomotor vigilance deficits from sleep inertia. Sleep, 24(7), 813–819.

117 Wendte, R., Snel, J., & Tieges, Z. (2003). Misclassification of caffeine use: Causes and examples. In Fac. Psychologie, Programmagroep Psychonomie (pp. 1–20). Amsterdam: Universiteit van Amsterdam. Wesensten, N. J., Belenky, G., Kautz, M. A., Thorne, D. R., Reichardt, R. M., & Balkin, T. J. (2002). Maintaining alertness and performance during sleep deprivation: Modafinil versus caffeine. Psychopharmacology (Berlin), 159(3), 238–247. Wesensten, N. J., Belenky, G., Thorne, D. R., Kautz, M. A., Balkin, T. J., & Balkin, T. J. (2004). Modafinil vs caffeine: Effects on fatigue during sleep deprivation. Aviation, Space, and Environmental Medicine, 75(6), 520–525.

Wesensten, N. J., Killgore, W. D., & Balkin, T. J. (2005). Performance and alertness effects of caffeine, dextroamphetamine, and modafinil during sleep deprivation. Journal of Sleep Research, 14(3), 255–266. Whalen, D. J., Silk, J. S., Semel, M., Forbes, E. E., Ryan, N. D., Axelson, D. A., et al. (2008). Caffeine consumption, sleep, and affect in the natural environments of depressed youth and healthy controls. Journal of Pediatric Psychology, 33(4), 358–367. Wurts, S. W., & Edgar, D. M. (2000). Caffeine during sleep deprivation: Sleep tendency and dynamics of recovery sleep in rats. Pharmacology, Biochemistry and Behavior, 65(1), 155–162.

H. P. A. Van Dongen and G. A. Kerkhof (Eds.) Progress in Brain Research, Vol. 190 ISSN: 0079-6123 Copyright Ó 2011 Elsevier B.V. All rights reserved.

CHAPTER 7

Can light make us bright? Effects of light on cognition and sleep Sarah Laxhmi Chellappa{,{, Marijke C. M. Gordijn},* and Christian Cajochen{ {

Centre for Chronobiology, Psychiatric Hospital of the University of Basel, Basel, Switzerland { CAPES Foundation/Ministry of Education of Brazil, Brasilia-DF, Brazil } Department of Chronobiology, University of Groningen, Groningen, The Netherlands

Abstract: Light elicits robust nonvisual effects on numerous physiological and behavioral variables, such as the human sleep–wake cycle and cognitive performance. Light effects crucially rely on properties such as dose, duration, timing, and wavelength. Recently, the use of methods such as fMRI to assess light effects on nonvisual brain responses has revealed how light can optimize brain function during specific cognitive tasks, especially in tasks of sustained attention. In this chapter, we address two main issues: how light impinges on cognition via consolidation of human sleep–wake cycles; and how light directly impacts on sleep and cognition, in particular in tasks of sustained attention. A thorough understanding of how light affects sleep and cognitive performance may help to improve light settings at home and at the workplace in order to improve well-being. Keywords: light; cognition; sleep; circadian clock; human.

Introduction

whereas in nocturnal animals, the light phase comprises the rest phase, thus representing a different temporal niche for sleep. The human visual system is designed to accommodate the needs of a diurnal species from a visual perspective. However, also the nonvisual responses to light point to the role of light as the mediator of inducing daytime physiology in humans. Hormonal secretion, heart rate, body temperature, sleep propensity, alertness, pupillary constriction, and gene expression are all immediately influenced by light—or even hours after light exposure ended—in order to pursue optimal adaptation to the imposed light–dark cycle

The 24-h reoccurrence of light and darkness represents the most systematic time cue on earth. Thus, it is not surprising that all living organisms integrated the light–dark cycle in their physiology and optimally adapted their anatomy and behavior to anticipate dawn and dusk. In humans, light is intuitively linked with an alert or wakeful state, *Corresponding author. Tel.: þ31 50 3637658; Fax: þ31 50 3632148 E-mail: [email protected] DOI: 10.1016/B978-0-444-53817-8.00007-4

119

120

(Berson, 2003; Bromundt et al., 2011; Cajochen et al., 2005; Hatori and Panda, 2010; Lavoie et al., 2003; Muñoz et al., 2005). Recently, these long-term and acute effects of light on physiology have been described as belonging to the non-image forming (NIF) system, given that these responses are not associated with the classical involvement of rod and cone photopigments (Guler et al., 2008). This has opened a new area of research, which, apart from the role of light in regulating circadian rhythms, focuses on the physiological and anatomical underpinnings of light in modulating sleep and cognition. Thus, in this chapter, we address two issues: (1) how light impinges on cognition via consolidation of human sleep–wake cycles and (2) how light directly impacts on sleep and cognition, in particular, in tasks of sustained attention. Effects of light on the circadian timing system and sleep–wake cycles The response of the central circadian pacemaker located in the suprachiasmatic nuclei (SCN) to light pulses plays a crucial role in the synchronization to the environmental light–dark cycles. Light pulses presented during the subjective day rapidly induce expression of the immediate early gene c-fos (Rusak et al., 1990) and the clock gene Per1 within the SCN (Albrecht et al., 2001; Edelstein et al., 2003), resulting in phase shifts of behavioral circadian rhythms. Thus, the mammalian Per genes are not only light-responsive components of the circadian oscillator but also are involved in resetting of the circadian clock (for a review, see Oster et al., 2002). In humans, exposure to light late in the biological day (dusk) leads to a delay in human sleep onset, while exposure to light early in the biological day (dawn) advances activity onset (for a review, see Czeisler and Gooley, 2007). This phase-shifting property of light denotes a clear NIF effect of light, which relies on circadian phase and substantially impacts on the temporal organization of sleep and wakefulness.

Thus, light as the major “Zeitgeber” (i.e., time giver) is a prerequisite for ideal synchronization between the temporal organization of sleep and wakefulness and the external light–dark cycle. As a consequence, attenuated Zeitgeber strength (i.e., not enough light, or light of an inappropriate wavelength, i.e., light at longer wavelengths, > 600 nm) and light at inappropriate times (i.e., during the biological night) can lead to improper entrainment between internal (i.e., circadian) and external (i.e., 24-h earth rotation) time, which often occurs in people working on rotating shifts and/or older individuals as well as visually impaired people. There is ample evidence that timed light exposure is a successful countermeasure of circadian misalignment in shift work (Burgess et al., 2002). Exposure to bright light did not only stabilize sleep–wake rhythms in demented older people, but significantly attenuated the decline in mental capabilities over the investigated time span of 4.5 years (Riemersma-van der Lek et al., 2008). Similarly, bright light promoted circadian alignment and prevented the detrimental effects of night work on sustained attention, as measured in an increased response speed on the psychomotor vigilance test (Santhi et al., 2008). The beneficial effects of light on circadian entrainment clearly carry on to cognitive performance. Thus, in this way, it is rather the stabilization of sleep–wake rhythms that leads to better brain function, than the exposure to bright light per se. Strong evidence for this stems from the fact that, if sleep and wakefulness occur out of phase with internal biological time, this impairs several cognitive functions such as learning in humans (Wright et al., 2006). Further, fragmentation of the rest-activity rhythm correlates with age-related cognitive deficits (Oosterman et al., 2009). Thus, stable and consolidated circadian sleep–wake rhythms are an essential requirement for proper cognitive functioning in health and disease, and light is only a means to an end. However, there is recent evidence that light per se may directly impinge on sleep, alertness, cognitive performance, and even mood

121

levels probably even without its action via the central circadian pacemaker. Light directly impacts on sleep and cognition In nocturnal animals, there is recent evidence that the acute light-induction of sleep is mediated by melanopsin-based photoreception (Lupi et al., 2008; Tsai et al., 2009), potentially reflecting an additional clock-independent photic input to sleep. To our knowledge, a similar wakefulnessinducing effect of light during sleep has not been substantiated as such in the “diurnal” humans, as it is difficult to apply light while sleeping, and testing the effects of short-wavelength light during sleep is impossible because of the filtering properties of the eyelids (Moseley et al., 1988). Therefore, most of the evidence of acute light effects on sleep in humans comes from studies applying light shortly before or after sleep. Bright light in the morning has been shown to shorten sleep duration (Dijk et al., 1989) and advance circadian rhythms (i.e., melatonin profile), without effects on nonrapid eye movement (NREM) sleep homeostasis (Carrier and Dumont 1995; Dijk et al., 1989). An artificial dawn in the morning during the last 30 min of sleep caused more superficial sleep along with a faster decline in skin temperatures and less sleepiness after waking up (Van de Werken et al., 2010). Despite positive effects on alertness in the morning, the use of the same artificial dawn during 2 weeks did not induce a significant change in circadian phase (Giménez et al., 2010). Bright light in the evening may lead to an increase in sleep latency to NREM sleep stage 2 (Cajochen et al., 1992; Carrier and Dumont 1995) and changes in the temporal dynamics of electroencephalographic (EEG) slow-wave activity (SWA), such that SWA is lower during the first and higher during the fourth NREM–REM sleep cycle as compared to dim light condition (Cajochen et al., 1992). Exposure to bright polychromatic light (2500 lux) in the morning (06:00 h–09:00 h) or in the evening

(18:00 h–21:00 h) for 3 consecutive days can result in earlier sleep termination following morning light than after evening light (Gordijn et al., 1999). Interestingly, the duration of the first REM sleep episode was longer after morning light than after evening light, most likely due to a phase advance of the circadian influence on REM sleep production. This is supported by the observed advance of the circadian rhythm of melatonin. Thus, both sleep termination and REM sleep duration can be manipulated by light exposure. All these effects have been interpreted as reflecting a carryover effect of light's alerting action into sleep (i.e., longer sleep latencies, reduced SWA in the first cycle with an intrasleep rebound of SWA in the last cycle), reflecting the repercussion of the immediate induction of a circadian phase advance or delay on the following sleep episode, and showing the positive effects of artificial dawn on the dissipation of sleep inertia after awakening, which could not be explained by circadian mechanisms.

Circadian and homeostatic influences on the alert state Wakefulness is a construct associated with high levels of environmental awareness, which can be tracked down by a wide array of responses, such as subjective perception, and behavior, subcortical and cortical activity (Buysse et al., 2003). Basically, it comprises self-reported low levels of fatigue or sleepiness, fast and more accurate responses in tasks of sustained attention, low power densities in the EEG theta frequency range (4–8 Hz), and high power densities in the EEG beta frequency range (12–30 Hz; Badia et al., 1991; Daurat et al., 2000). Subjective perception of wakefulness crucially relies on a time-of-day dependency, to the extent that diurnal fluctuations of alertness follow similar dynamics of core body temperature (CBT), with its maximum in the evening and nadir in the early morning (Kleitman, 1987).

122

Two important protocols have been developed in order to dissect out the relative contributions of circadian and sleep–wake homeostatic processes in humans: the constant routine and the forced desynchrony protocols (Duffy and Dijk, 2002). In the first, the amplitude and phase of many circadian rhythms can be elucidated without the masking effects of food, posture, light, and so forth. In the second protocol, subjects live on artificially very long or very short days so that the circadian system is no longer entrained to the imposed sleep–wake cycle. The desynchronized subjects sleep at different circadian phases of the entire 24-h cycle, which enables to differentiate the contribution of the sleep homeostatic process or the circadian system to a given variable (Dijk et al., 1997). Usually—but not always—both factors contribute substantially to measures as alertness, mood, and neurobehavioral performance (Cajochen et al., 1999a; Koorengevel et al., 2003). Accordingly, some forced desynchrony studies have revealed that the deleterious effects of prior wakefulness on alertness were strongest during the minimum of the endogenous CBT rhythm. This strongly indicates that optimal levels of alertness can be achieved when the phase relationship between the endogenous circadian timing system and the sleep–wake cycle is such that the former opposes the wake-dependent deterioration of alertness and performance, as conceptualized in the “two-process” model (Borbély 1982; Daan et al., 1984). The most effective means of obtaining this occurs when the waking day starts  2 h after the endogenous circadian minimum of CBT rhythm, which corresponds to  2 h after the circadian maximum of the plasma melatonin rhythm. Taken together, this implies that the circadian process plays a wake-promoting role that counteracts the accumulating homeostatic drive for sleep during wakefulness. The result of the interaction between these two fundamental properties emanating from the central nervous system is that humans are able to maintain alert wakefulness for  15–17 h (Czeisler et al., 1994; Dijk and Czeisler

1994). Considering the temporal dynamics of these processes on alertness, one can hypothesize that light exerts its alerting effects most strongly when the circadian drive for sleep is at its maximum (i.e., in the early morning at the CBT minimum), under high homeostatic sleep pressure (i.e., after more than 16 h of wakefulness). The impact of light on sleep and wakefulness, however, does not happen in a homogenous manner for all types of light exposure, but intimately depends on intensity, timing, duration, and wavelength. Thus, in the next sections, we describe how these aspects of light exposure can play a crucial role on wakefulness.

Do light effects impact on sleep and wakefulness irrespective of timing? While applying light during sleep is difficult in humans, measuring light effects during wakefulness is easy, but its interpretation rather multifaceted, since factors such as the duration of prior wakefulness, endogenous circadian phase, and prior environmental light exposure all interact with each other. Most studies on the effects of light in humans are conducted at night (Badia et al., 1991; Campbell and Dawson 1990; Foret et al., 1996; Lockley et al., 2006). Considering how light acts on the circadian and homeostatic systems (as described in the previous section), one does in fact expect that, even during the biological night in extended wakefulness (when sleep pressure is rising), light can dramatically enhance subjective alertness and reduce objective markers of sleepiness (e.g., slow eye movements). From a circadian perspective, there is a considerable body of evidence that suggests that, at night, the strong melatonin suppression caused by exposure to high-intensity light may be one of the underlying mechanisms (Cajochen et al., 1998; Sack et al., 1992). Although not consistently shown, it has been hypothesized that melatonin attenuates SCN-dependent mechanisms

123

responsible for promoting and maintaining cortical and behavioral arousal at particular times in the circadian cycle (Cajochen et al., 1999b; Dijk and Czeisler 1995; Sack et al., 1997; Wright, 1997). However, it is very likely that there are additional mechanisms that mediate the alerting effects of light. Light exposure at night on the nasal part of the retina does induce suppression of melatonin, without effects on alertness (Ruger et al., 2005a). Also during the biological day, when melatonin is at minimal levels, light does impact on alertness. A study in which participants were exposed to either bright light (5000 lux) or dim light < 10 lux (control condition) either between 12:00 and 16:00 h or between 00:00 and 04:00 h showed that bright light had a timedependent effect on heart rate and CBT, such that bright-light exposure at night, but not in daytime, increased heart rate and increased CBT. However, the effects of bright light on the psychological variables was time independent, since both nighttime and daytime bright light reduced sleepiness and fatigue significantly and similarly (Fig. 1a and b; Ruger et al., 2006). Further evidence in support of daytime effects of light on alertness was found in an “in-lab” study, with daytime exposure to short (21 min) white light at > 7000 lux; cortical activity was enhanced during an oddball task and subjective alertness improved in a dynamic manner (Vandewalle et al., 2006). These alerting effects declined within minutes after the end of the light stimulus. During the light exposure, brain activity showed various region-specific time courses, such as enhanced responses in the posterior thalamus, including the pulvinar nucleus, which has been implicated in visual attention and alertness regulation. This suggests that light may modulate activity of subcortical structures involved in alertness, thereby promoting cortical activity in networks involved in ongoing nonvisual cognitive processes. It remains inconclusive and controversial whether the timing of light exposure does have an impact on

human sleep. For instance, in an ultrashort sleep–wake schedule (Kubota et al., 2002), exposure to evening bright light (5000 lux) delayed the diurnal fluctuation of sleep propensity, which suggests that light may have the potential to phase-advance or -delay sleep phase in a similar manner as to the phase-response curve (PRC) derived from CBT— or melatonin—rhythm. Melatonin has been hypothesized to act as a mediator to convey the output of the circadian pacemaker to the sleep–wake system (Lavie, 1997). Nocturnal bouts of sleep propensity observed in an ultrashort sleep–wake schedule seem to occur in parallel with an increase in melatonin secretion near to habitual bedtime (Kubota et al., 2002). However, it is also likely that the magnitude of the phase change in melatonin secretion after bright-light exposure may not correlate with that in sleep propensity rhythm. An alternative hypothesis suggests that an indirect effect of melatonin on sleep–wake cycle mediated via temperature may be essential in considering coupling mechanisms between the circadian pacemaker and sleep, since melatonin acts strongly on the temperature rhythm (Van Someren, 2000). However, timing per se is not the only factor one should bear in mind when considering how light impacts on alertness. Since the aforementioned studies used polychromatic light above 1000 lux, it may be that the intensity of the light exposure is, in fact, responsible for such alerting effects, which appear to be irrespective of timerelated dependency. This leads to the next question: What is the threshold of light intensity that can keep us awake?

Dose–response effects of light: Is there a saturation point? Light is a powerful synchronizer which resets the endogenous circadian pacemaker to the 24-h day in an intensity-dependent manner. Although it is clearly recognized that bright light (1000 lux or more) is an effective synchronizer in humans, one might believe that the human circadian pacemaker

(a) A

Daytime light exposure

B Nighttime light exposure

(b) A Daytime light exposure

75

Heart rate (beats/min) (mean ± sem)

**

8

**

Sleepiness scores (KSS) (mean ± sem)

ns 70 65 60 55

B

Nighttime light exposure

**

6

4

2

50

ns

10 8 6 4 2

Fatigue scores (VAS-Fatigue) (mean ± sem)

ns

12

0 37.0

Core body temperature (°C) (mean ± sem)

0 10 8

** **

6 4 2 0

ns

*

36.8 36.6 36.4 36.2

0.0 08:00 10:00 12:00 14:00 16:00 18:0020:00 22:00 00:00 02:00 04:00 06:00 External time (h) External time (h) Dim light condition

Bright light condition

Energy scores (VAS-Energy) (mean ± sem)

Cortisol concentration (pg/ml) (mean ± sem)

0 14

10

#

#

8 6 4 2 0 08:00 10:00 12:00 14:00 16:00 18:00 20:00 22:00 00:00 02:00 04:00 06:00 External time (h) Dim light condition

External time (h) Bright light condition

Fig. 1. (a) Time course of heart rate, cortisol concentration, and core body temperature for the two experiments, before and during bright-light exposure versus dim light. Hatched bars, period of light exposure (daytime experiment: noon until 4.00 p.m.; nighttime experiment: midnight until 4:00 a.m.; Ruger et al., 2006). (b) Time course of subjective sleepiness, fatigue, and energy for the two experiments, before and during bright-light exposure versus dim light. Hatched bars, period of light exposure (daytime experiment: noon until 4:00 p.m.; nighttime experiment: midnight until 4:00 a.m.; (Ruger et al., 2006). Significances indicated refer to post hoc ANOVAs for the daytime and nighttime experiments separately (#p500 ms) on the PVT. The first two test bouts of each waking period, which may be affected by sleep inertia, are omitted in order to focus on the homeostatic and circadian patterns in the data. Notice that the acute total sleep deprivation condition (left panel) exposed the entire 24 h of the circadian cycle in performance, whereas the nadir of the circadian cycle was masked by scheduled sleep in the chronic sleep restriction conditions (right panel). Vertical bars indicate scheduled sleep periods. Figure adapted from McCauley et al. (2009) with permission.

160

impairment due to chronic sleep restriction can reach considerable magnitude. The dose-dependent cumulative buildup of cognitive impairment across days was also seen in another study of sleep restriction, to 3, 5, or 7 h per day and including sleep extension to 9 h per day for 7 days (Belenky et al., 2003; see Fig. 5). The 3-h daily sleep condition of this study revealed that when sleep was restricted to below 4 h, the accumulation of performance impairment was disproportionately fast. This same conclusion was drawn from a study involving 3 days with only 2 h of sleep per day (Van Dongen and

20 Worse

Performance (PVT lapses)

15

10

5 Better 0

0

1

2

3

4

5 6 Days

7

8

9 10

Fig. 5. Performance on the PVT across days of sleep restriction or extension as a function of sleep dose. A total of 66 healthy young adults were subjected to one of four laboratory sleep conditions (Belenky et al., 2003). Details are the same as for Fig. 4, except that subjects received one of four doses of sleep for 7 consecutive days, followed by three recovery days with 8 h TIB. The sleep doses were 3 h TIB per day for 13 subjects (circles), 5 h TIB per day for 13 subjects (boxes), 7 h TIB per day for 14 subjects (diamonds), and 9 h TIB per day (sleep extension) for 16 subjects (triangles, control group). Awakening was scheduled at 07:00 each day. Performance on a number of cognitive tasks including the PVT was tested daily at 09:00, 12:00, 15:00, and 21:00. In the 5 h TIB condition an additional test bout occurred at midnight, and in the 3 h TIB condition yet another one took place 2 h after midnight. The graphs show group averages for the number of lapses (reaction times greater than 500 ms) on the PVT, omitting the first test bout to avoid sleep inertia. Figure adapted from McCauley et al. (2009) with permission.

Dinges, 2003), and was recognized to apply to total sleep deprivation as well (Van Dongen et al., 2003). Neither the accumulation of performance impairment across days of sleep restriction per se nor the accelerated accumulation when sleep is curtailed below 4 h is accurately predicted by the two-process model (Van Dongen et al., 2003). The data displayed in Fig. 5 also suggested that the rate of recuperation in the recovery days at the end of the study depended on prior sleep dose, and that recuperation was incomplete following the 3- and 5-h sleep conditions after 3 days with 8 h scheduled for recovery sleep. This was interpreted as indicating that prior sleep restriction essentially causes a gradual change in the brain, whereby the set point of maximal performance drifts to a suboptimal level (Belenky et al., 2003). By extension, it was hypothesized— and recently shown—that sleep extension can actually improve the set point of maximal performance (Rupp et al., 2009). Figure 4 and particularly Fig. 5 also show that when sleep dose is decreased, the amplitude of within-day changes is increased, with deficits growing particularly in the morning hours (Mollicone et al., 2010). This feature has previously been linked to a presumed sigmoidal effect on performance of the homeostatic process (Jewett and Kronauer, 1999), but is nowadays believed to be attributable to the circadian process in interaction with the homeostatic process (Dijk and Franken, 2005). The long-lasting effects of chronic sleep restriction suggest that some aspect of the homeostatic process undergoes a gradual change that is slow to recover. The sleep, activity, fatigue, and task effectiveness (SAFTE) model of fatigue and performance (Hursh et al., 2004b) was the first to address this issue. The model contains a “sleep reservoir” to represent the homeostatic process, with depletion of the sleep reservoir corresponding to the buildup of pressure for sleep. During chronic sleep restriction, the capacity of the sleep reservoir is gradually downregulated to account for the cumulative buildup of performance impairment.

161

During recovery sleep, the reservoir capacity is gradually restored, but this takes a while. As such, the recuperation rate is initially reduced after a period of sustained sleep restriction, thereby extending the time to full recovery. The SAFTE model also has a limit on the maximum rate of replenishment during sleep, such that sleep periods greater than about 4 h per day ultimately result in stable levels of reduced performance, but sleep restriction to less than about 4 h per day leads to continuous performance degradation. This model behavior captures essential features of the data shown in Figs. 4 and 5. The SAFTE model, which is routinely used to predict performance in a range of operational settings (Dean et al., 2007), has a special provision to account for sleep quality, which in the model is modulated by two factors. First, the circadian process directly affects the rate of sleep accumulation to the sleep reservoir. As a result, a period of sleep during the day, when circadian activation is high (for normal nighttime sleepers), is less beneficial to restoring performance than an equivalent period of sleep during the night, when circadian activation is low. Second, in the model, sleep accumulation does not start immediately after the onset of a sleep period. A nominal delay of 5 min is imposed between the start of sleep and beneficial recovery of the reservoir. As a result, an environment that results in sleep fragmentation (i.e., brief interruptions of sleep) causes repetitive 5 min penalties on sleep accumulation, which accounts for the reduced restorative value of fragmented sleep (Bonnet and Arand, 2003). Recently, another published model of performance built on the foundation of the two-process model—the three-process model of alertness (Folkard and Åkerstedt, 1987)—was also expanded to capture cumulative performance changes in chronic sleep restriction scenarios (Åkerstedt et al., 2008). This model, in which the third of the three processes describes the transient effect of sleep inertia, focuses on predicting (subjective) alertness, although it has also been shown to predict sleep (Åkerstedt and Folkard, 1996).

The model developers noticed that the exponential recovery rate during sleep in the original version of their model was too high to be able to account for the cumulative effects of chronic sleep restriction. They therefore posited that there is a level of alertness below which the exponential recovery function switches to a linear recovery function (this point being 12.2 on a scale from 1 to 21), such that recovery is less efficient when sleepiness is high. Although this modification improved the amount of variance captured in data from an experimental study in which sleep was restricted to 4 h per day for five consecutive days, it is an ad hoc solution (Åkerstedt et al., 2008) which does not provide much insight into the potential underlying mechanisms. McCauley et al. (2009) took a different approach to modifying the homeostatic process to capture the data from chronic sleep restriction experiments. Based in part on an idea originally proposed by Johnson et al. (2004), McCauley and colleagues introduced an additional homeostatic process (process u) with a longer time constant than the original homeostatic process. The additional process also builds up during wakefulness and dissipates during sleep, and its state codetermines the rate by which the original homeostatic process builds up during wakefulness and dissipates during sleep. This dual homeostatic process model was implemented as a set of firstorder ordinary differential equations, with the circadian process included as a nonhomogeneity in the differential equation system. A consequence of this model structure is a dynamic interaction between the homeostatic and circadian processes, which naturally captures the observed modulation of circadian amplitude as a function of sleep dose (see Fig. 5). Analysis of the dynamics of the model of McCauley and colleagues revealed the existence of a bifurcation (i.e., a qualitative change in model behavior): for sleep durations of about 4 h per day or more, the model predicts that daily average cognitive performance converges to a steady state, whereas for sleep durations shorter

162

than about 4 h per day, the model predicts that daily average performance escalates and does not reach a steady state. This feature, which is an emerging behavior of the model and was not purposely inserted into the equations, provided a new explanation for the equilibrium states of suboptimal performance under conditions of moderate chronic sleep restriction down to approximately 4 h per day and also for the disproportionate buildup of performance impairment observed in the 3 h per day condition (Belenky et al., 2003) and the total sleep deprivation condition (Van Dongen et al., 2003) relative to the chronic sleep restriction conditions with daily sleep durations of 4 h or more (see Figs. 4 and 5). This explanation was tentatively linked to underlying neurobiological processes: it was hypothesized that the original homeostatic process would reflect a metabolic cascade involving adenosine production, and that the newly added process would reflect up- and downregulation of the adenosine receptor (McCauley et al., 2009). Mathematical models predicting fatigue and performance changes over time generally do not make distinctions between various cognitive skills or functions, such as memory, attention, reasoning, language, or decision making. Rather, the models predict variation in general capacity that is variously reflected in the speed and accuracy of these functions. The justification for this is twofold: (1) many aspects of cognitive function appear to covary as a function of sleep and circadian rhythm, perhaps with differences in sensitivity to sleep deprivation and circadian phase; and (2) most modeling applications are for individuals performing a range of activities that depend on many cognitive faculties, so having a model that represents the general effects on all cognitive functions may typically be more useful than a model that only predicts effects on a single function. Even so, efforts are underway to develop computational models that make predictions of moment-to-moment performance changes in selected cognitive tasks (Gunzelmann et al., 2009;

Ratcliff and Van Dongen, 2009), which may eventually lead to detailed predictions of cognitive impairment in specific operational contexts. Predicting fatigue-related accident risk Mathematical models can be useful in managing fatigue, performance, and accident risk in many operational settings. To aid in transitioning the theoretical models to operational use, several steps are involved. One involves converting model outputs to usable parameters, such as a fatigue rating or performance effectiveness scale. A second involves calibrating the outputs of the model to actual data, which could include the risk of accidents. For mathematical models of cognitive performance based on sleep, a conversion of work schedules into predicted sleep times of the workers is also useful. The Federal Railroad Administration (FRA) completed a research program to calibrate and validate mathematical models of sleep and performance for use in predicting and managing fatigue risk in railroad workers (Hursh et al., 2006, 2008). Because such models offer the possibility of objectively assessing and forecasting fatigue and performance, employees and employers can use them to help schedule work and rest to avoid fatigue and improve performance. A useful model needs to be calibrated to the demands of a particular job so that the model predictions can be related to the risk of meaningful failures of human performance. One important part of preparing a model for use as a fatigue risk management tool is an assessment of whether the tool can predict the risk of human factor error or the risk of having a human factor accident (HFA). Accidents are rare, random events that are often modeled as probabilities using Poisson distributions, which describe the relationship between periods of time and the probability that a certain number of accidents will occur in that period of time (Parzen, 1960). Train accidents

163 Table 1. Human factor accidents (HFAs), proportion of work time, employee-hours, human factors accidents per employee-hour, and probability of HFAs, as a function of performance effectiveness predicted by the SAFTE model

SAFTE-predicted effectiveness (inverse of fatigue) Human factor accidents (HFA) Proportion of work time Employee-hours (e-h) HFA per e-h (m) Probability of  1 HFA per 200,000 e-h

< 50 (severely fatigued)

50–59 (extremely fatigued)

60–69 (very fatigued)

70–79 (moderately fatigued)

80–89 (mildly fatigued)

90–100 (not fatigued)

33 0.027 20,457,144 1.61  10 6 0.276

38 0.043 31,961,042 1.19  10 6 0.212

95 0.118 88,125,804 1.08  10 6 0.194

123 0.158 118,489,773 1.04  10 6 0.188

183 0.234 175,186,642 1.04  10 6 0.189

259 0.420 314,829,738 0.82  10 6 0.152

have been modeled in this fashion.1 In the case of estimating accident rates for fatigue—that is, the conditional probability of a HFA given a particular level of fatigue—the problem is twofold: (1) the frequency of accidents due to various levels of fatigue must be known; and (2) the number for employee-hours (e-h) exposure by level of fatigue must be known. Databases such as the FRA safety database can provide the number of various types of accidents and e-h for a specified time period, but they have no information about the involvement of fatigue in those accidents. This limitation was addressed with mathematical modeling of fatigue and performance in the FRA research program (Hursh et al., 2006, 2008). The research thus provided information on the frequency of accidents as a function of mathematically predicted fatigue, and allowed for the estimation of e-h exposure and risk as a function of predicted fatigue. Table 1 shows the number of HFA caused by freight railroad operators (locomotive engineers and conductors) in the period from January 2003 to June 2005 for BNSF Railway, CSX

If m is the mean rate of accidents that occur in time t, then the probability that one or more accidents will occur equals 1  e mt. For example, if in a particular year, there were 1000 HFAs and 800  106 employee-hours (e-h) worked, the mean accident rate per e-h would be m ¼ 1000/800  106 ¼ 1.25  10 6. Consequently, the probability of one or more HFA per 200,000 e-h would be 1  e 1.25  10–6  200,000 ¼ 0.22.

1

Transportation, Kansas City Southern Railway, Norfolk Southern Railroad, and Union Pacific Railroad combined, as a function of operator effectiveness (which is inversely related to fatigue). Here, effectiveness (scaled 0–100) was predicted with the SAFTE model based on operators’ 30-day work histories (Hursh et al., 2008). The SAFTE model based these predictions on estimated amounts of sleep that would occur under the work schedules at hand (Hursh et al., 2004a). Based on this study, Table 1 shows the estimated proportion of work time that operators spent in each fatigue level. Only 790 of approximately 148,500 employees were involved in the observed HFA; therefore, the actual work time for the operators at hand did not provide sufficient information to reliably determine the accident rate. However, operational data for the railroads, including e-h, could be obtained from the FRA safety database. In the period from January 2003 to June 2005, more than 749 million e-h were worked at the five railroads. The e-h data were partitioned into fatigue levels using the proportion of work time information in the table. Given the frequency of accidents and the e-h exposure at each fatigue level, a HFA rate per e-h was then calculated, as shown in the table. The probability of one or more HFA per 200,000 e-h (a standard exposure metric for occupational accidents) is also shown. It is well known from epidemiology (Lilienfeld and Lilienfeld, 1980) that outcome risk is

164

Cumulative proportion of work time

determined by both magnitude and duration of exposure. In the present case, HFA risk should therefore be determined by fatigue severity as well as duration of exposure to fatigue. In agreement with this, the results in Table 1 show that the probability of one or more HFA per 200,000 e-h increased as predicted fatigue increased (i.e., as effectiveness decreased, from right to left). Indeed, there was a significant correlation (r ¼ 0.93) between crew fatigue (effectiveness) and the risk of a HFA (Hursh et al., 2008); no such relationship was found for non-HFAs (of which 1,000 were evaluated). It was also found that the 95% confidence interval for the cumulative relative risk of a HFA at predicted effectiveness scores below 70 was significantly greater than chance and the cumulative relative risk of a non-HFA. This suggests that a SAFTE-predicted effectiveness of 70 may be a suitable fatigue threshold in freight railroad operations; below 70 fatigue becomes unacceptable because the increase in accident risk at that level compromises safety. The FRA research program also found that cause codes associated with accidents that occurred at predicted effectiveness scores below 70 showed an overrepresentation of the type of HFA that might be expected of a fatigued crew, such as passing a stop signal or exceeding the maximum authorized speed. This confirmed that

the detected relationship between accident risk and predicted effectiveness is meaningful (Hursh et al., 2008). Table 1 shows that up to about 20% of work time occurred at predicted effectiveness scores below 70. This suggests that greater than 20% of work time at an effectiveness level below 70 is a good first-order approximation of severity and duration exposure for establishing a fatigue benchmark for rail operations. A reasonable goal for fatigue risk management would be to keep employee work time within that threshold. Estimating the distributions of work time by fatigue level for different groups of employees can be a useful tool for determining where fatigue problems exist and where to allocate risk management resources. Over the past 5 years, the FRA conducted a series of studies in which a random sample of railroad employees in a particular occupation was asked to keep a 2-week diary of their work and sleep times (Gertler and DiFiore, 2009; Gertler and Viale, 2006a,b, 2007). Work and sleep time information was used to calculate the cumulative proportion of work time at various levels of SAFTE-predicted effectiveness (fatigue). For example, Fig. 6 shows the cumulative proportion of work time as a function of predicted effectiveness from a diary study of locomotive engineers and conductors (Gertler and DiFiore, 2009).

1.00 0.80 0.60 0.40 0.20 0.00