Вы находитесь на странице: 1из 1757

Principles and Practice of

Fifth Edition

Principles and Practice of

Meir H. Kryger, MD, FRCPC
Clinical Professor of Medicine
University of Connecticut
Director of Research and Education
Gaylord Sleep Medicine
Wallingford, Connecticut

Thomas Roth, PhD

Professor, Department of Psychiatry
Wayne State University School of Medicine
Director and Division Head
Sleep Disorders and Research Center
Henry Ford Hospital
Detroit, Michigan
Clinical Professor, Department of Psychiatry
University of Michigan Medical School
Ann Arbor, Michigan

William C. Dement, MD, PhD

Lowell W. and Josephine Q. Berry Professor of Psychiatry
and Behavioral Sciences, and Director
Sleep Disorders and Research Center
Stanford University School of Medicine
Palo Alto, California
3251 Riverport Lane
St. Louis, Missouri 63043
Premium Edition: 978-1-4377-0731-1
Copyright 2011, 2005, 2000, 1994, 1989 by Saunders, an imprint of Elsevier Inc.

No part of this publication may be reproduced or transmitted in any form or by any means, electronic
or mechanical, including photocopying, recording, or any information storage and retrieval system,
without permission in writing from the publisher. Details on how to seek permission, further
information about the Publishers permissions policies and our arrangements with organizations such as
the Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website:
This book and the individual contributions contained in it are protected under copyright by the
Publisher (other than as may be noted herein).


Knowledge and best practice in this field are constantly changing. As new research and experience
broaden our understanding, changes in research methods, professional practices, or medical
treatment may become necessary.
Practitioners and researchers must always rely on their own experience and knowledge in
evaluating and using any information, methods, compounds, or experiments described herein. In
using such information or methods they should be mindful of their own safety and the safety of
others, including parties for whom they have a professional responsibility.
With respect to any drug or pharmaceutical products identified, readers are advised to check the
most current information provided (i) on procedures featured or (ii) by the manufacturer of each
product to be administered, to verify the recommended dose or formula, the method and duration
of administration, and contraindications. It is the responsibility of practitioners, relying on their own
experience and knowledge of their patients, to make diagnoses, to determine dosages and the best
treatment for each individual patient, and to take all appropriate safety precautions.
To the fullest extent of the law, neither the Publisher nor the authors, contributors, or editors,
assume any liability 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.

Library of Congress Cataloging-in-Publication Data

Principles and practice of sleep medicine / [edited by] Meir H. Kryger, Thomas Roth, William C.
Dement.5th ed.
p. ; cm.
Includes bibliographical references and index.
ISBN 978-1-4160-6645-3
1. Sleep disorders. 2. Sleep. I. Kryger, Meir H. II. Roth, T. (Tom) III. Dement, William C.,
[DNLM: 1. Sleep Disorders. 2. Sleepphysiology. WM 188 P957 2011]
RC547.P75 2011

Senior Acquisitions Editor: Dolores Meloni

Senior Developmental Editor: Anne Snyder
Associate Developmental Editor: Julie Goolsby
Publishing Services Manager: Patricia Tannian
Senior Project Manager: Sarah Wunderly
Design Direction: Steve Stave

Working together to grow

libraries in developing countries
Printed in Canada www.elsevier.com | www.bookaid.org | www.sabre.org

Last digit is the print number: 9 8 7 6 5 4 3 2 1

We dedicate this volume to our children:

Shelly, Michael, and Steven

Daniel, Adam, Jonathan, and Andrea

Elizabeth, Catherine, and Nicholas

But the tigers come at night,
With their voices soft as thunder,
As they tear your hope apart,
As they turn your dream to shame.
From I Dreamed a Dream, LES MISRABLES,
with permission, Cameron Mackintosh, producer
1985 Alain Boublil Music Ltd.
Used by permission 1991, CMI.

Blessings on him who first invented sleep.It covers a man

all over, thoughts and all, like a cloak.It is meat for the
hungry, drink for the thirsty, heat for the cold, and cold for
the hot.It makes the shepherd equal to the monarch, and
the fool to the wise.There is but one evil in it, and that is
that it resembles death, since between a dead man and a
sleeping man there is but little difference.
By Saavedra M. de Cervantes

To sleep! To forget! he said to himself with the serene

confidence of a healthy man that if he is tired and sleepy,
he will go to sleep at once. And the same instant his head
did begin to feel drowsy and he began to drop off into
forgetfulness. The waves of the sea of unconsciousness had
begun to meet over his head, when all at onceit was as
though a violent shock of electricity had passed over him.
He started so that he leapt up on the springs of the sofa, and
leaning on his arms got in a panic on to his knees. His eyes
were wide open as though he had never been asleep. The
heaviness in his head and the weariness in his limbs that he
had felt a minute before had suddenly gone.
By Leo Tolstoy
Contributors ix


Peter Achermann, PhD Charles W. Atwood, Jr., MD

Professor, Institute of Pharmacology and Toxicology, Associate Professor and Director, Sleep Medicine
University of Zurich, Zurich, Switzerland Fellowship, Department of Medicine, Division of
Sleep Homeostasis and Models of Sleep Regulation Pulmonary, Allergy, and Critical Care Medicine,
University of Pittsburgh School of Medicine; Assistant
Alon Y. Avidan, MD, MPH Chief of Medicine and Sleep Laboratory Director,
Associate Professor of Neurology; Associate Director, Department of Medicine, Pulmonary and Sleep
Sleep Disorders Program; Director, UCLA Neurology Medicine Section, VA Pittsburgh Healthcare System,
Residency Program; Director, UCLA Neurology Pittsburgh, Pennsylvania
Clinic, Department of Neurology, University of Medical Therapy for Obstructive Sleep Apnea
California, Los Angeles, Los Angeles, California
Physical Examination in Sleep Medicine Fiona C. Baker, PhD
Honorary Senior Research Fellow, Brain Function
Torbjrn kerstedt, PhD Research Group, School of Physiology, University of
Head Professor, Stress Research Institute, Stockholm the Witwatersrand, Johannesburg, South Africa; Sleep
University; Affiliated Professor, Department of Physiologist, Center for Health Sciences, SRI
Clinical Neuroscience, Karolinska Institute, International, Menlo Park, California
Stockholm, Sweden Sex Differences and Menstrual-Related Changes in
Introduction Sleep and Circadian Rhythms
Sleep, Stress, and Burnout
Thomas J. Balkin, PhD
Ravi Allada, MD Director, Behavioral Biology Branch Center for Military
Associate Professor, Neurobiology and Physiology, Psychiatry and Neuroscience, Walter Reed Army
Associate Director, Center for Sleep and Circadian Institute of Research, Silver Spring, Maryland
Biology, Northwestern University, Evanston, Illinois Performance Deficits during Sleep Loss: Effects of
Genetic Basis of Sleep in a Simple Model Organism: Time Awake, Time of Day, and Time on Task
Bilgay Izci Balserak, BA, MA, PhD
Richard P. Allen, PhD, FAASM Division of Sleep Medicine and Center for Sleep and
Associate Professor, Department of Neurology, School Respiratory Neurobiology, Department of Medicine,
of Medicine, Johns Hopkins University, Baltimore, School of Medicine, University of Pennsylvania,
Maryland Philadelphia, Pennsylvania
Restless Legs Syndrome and Periodic Limb
Sleep Disturbances and Sleep-Related Disorders in
Movements during Sleep
Sonia Ancoli-Israel, PhD Siobhan Banks, PhD
Professor of Psychiatry, University of California, San Research Fellow, Centre for Sleep Research, University
Diego, La Jolla, California of South Australia, Adelaide, Australia
Sleep and Fatigue in Cancer Patients
Chronic Sleep Deprivation
Insomnia in Older Adults
Actigraphy Steven R. Barczi, MD, FAASM
Roseanne Armitage, PhD Associate Professor of Medicine, Department of
Professor of Psychiatry, Adjunct Professor of Medicine, Division of Geriatrics and Gerontology,
Psychology, Director, Sleep and Chronophysiology University of Wisconsin School of Medicine and
Laboratory, University of Michigan Medical School, Public Health; Associate Director, Education and
Ann Arbor, Michigan Evaluation, Madison VA Geriatric Research,
Sex Differences and Menstrual-Related Changes in Education & Clinical Center, Wm. S. Middleton
Sleep and Circadian Rhythms Veterans Memorial Hospital, Director of Education,
University of Wisconsin Center for Sleep Medicine
Isabelle Arnulf, MD, PhD and Sleep Research, Madison, Wisconsin
Sleep Disorders Unit, Piti-Salptrire Hospital, Paris, Medical and Psychiatric Disorders and the
France Medications Used to Treat Them

x Contributors

Joseph T. Bass, MD, PhD Michel A. Cramer Bornemann, MD, D-ABSM,

Associate Professor, Department of Medicine and FAASM
Neurology, Northwest University, Evanston, Illinois Director, Minnesota Regional Sleep Disorders Center;
Animal Models for Disorders of Chronobiology: Cell Lead Investigator, Sleep Forensics Associates,
and Tissue Hennepin County Medical Center; Assistant Professor,
Department of Neurology, Department of Medicine,
Claudio L. Bassetti, MD
University of Minnesota School of Medicine; Faculty
Director, Neurocenter of Southern Switzerland, Lugano;
Instructor, Department of Biomedical Engineering,
Chairman, Neurology Department, Professor of
Bakken/MIND Laboratory, University of Minnesota
Neurology, University Hospital, Zrich, Switzerland
Idiopathic Hypersomnia Graduate School, Minneapolis, Minnesota
Sleep and Stroke Sleep Forensics
Non-REM Arousal Parasomnias
Gregory Belenky, MD
Peter Buchanan, MB, ChB, MD, FRACP
Research Professor and Director, Sleep and Performance
Senior Clinical Research Fellow, Sleep & Circadian
Research Center, Washington State University,
Group, Woolcock Institute of Medical Research;
Spokane, Washington
Introduction Senior Staff Specialist, Department of Respiratory
Fatigue Risk Management Medicine, Liverpool Hospital; Consultant Sleep
Sleep and Performance Monitoring in the Workplace: Medicine Physician, St. Vincents Clinic, Sydney,
The Basis for Fatigue Risk Management Australia
Positive Airway Pressure Treatment for Obstructive
Ruth M. Benca, MD, PhD Sleep Apnea-Hypopnea Syndrome
Professor, Department of Psychiatry, University of
Wisconsin-Madison School of Medicine, Madison, Orfeu M. Buxton, PhD
Wisconsin Instructor in Medicine, Division of Sleep Medicine,
Mood Disorders Harvard Medical School; Associate Neuroscientist,
Department of Medicine, Brigham and Womens
Kathleen L. Benson, PhD Hospital, Boston Massachusetts
Brain Imaging Center, McLean Hospital, Belmont, The Human Circadian Timing System and SleepWake
Massachusetts; Associate Clinical Professor (retired), Regulation
Department of Psychiatry and Behavioral Sciences,
Stanford University School of Medicine, Stanford, Daniel J. Buysse, MD
California; Director, Sleep Disorders Program Professor, Psychiatry, Clinical, and Translational
(retired), Psychiatry Service, VA Palo Alto Health Science, University of Pittsburgh School of Medicine;
Care System, Palo Alto, California Director, Neuroscience Clinical and Translational
Schizophrenia Research Center, University of Pittsburgh School of
Medicine, Pittsburgh, Pennsylvania
Donald L. Bliwise, PhD Clinical Pharmacology of Other Drugs Used as
Professor of Neurology, Psychiatry, Behavioral Sciences Hypnotics
and Nursing; Director, Program in Sleep, Aging and Insomnia: Recent Developments and Future
Chronobiology, Emory University School of Medicine, Directions
Atlanta, Georgia Treatment Guidelines for Insomnia
Normal Aging Georgina Cano, PhD
Sleep in Independently Living and Institutionalized
Research Assistant Professor, Neuroscience, University
of Pittsburgh, Pittsburgh, Pennsylvania
Bradley F. Boeve, MD Models of Insomnia
Professor of Neurology, Mayo College of Medicine, Michelle T. Cao, DO
Department of Neurology, Mayo Clinic, Rochester, Clinical Instructor, Division of Sleep Medicine, Stanford
Minnesota University School of Medicine, Palo Alto, California.
Alzheimers Disease and Other Dementias
Narcolepsy: Diagnosis and Management
Michael H. Bonnet, PhD Sleep Neuromuscular Diseases
Professor, Department of Neurology, Wright State Clinical Features and Evaluation of Obstructive Sleep
Apnea and Upper Airway Resistance Syndrome
University, Director of Sleep Laboratory; Neurology,
Dayton Department of Veterans Affairs Medical Colleen E. Carney, PhD
Center, Dayton, Ohio Assistant Professor, Department of Psychology, Ryerson
Acute Sleep Deprivation University, Toronto, Ontario, Canada
Shift Work, Shift Work Disorder, and Jet Lag Psychological and Behavioral Treatments for
Insomnia II: Implementation and Specific
Alexander A. Borbly, MD Populations
Professor Emeritus of Pharmacology, Institute of
Pharmacology and Toxicology, University of Zurich,
Zurich, Switzerland
Sleep Homeostasis and Models of Sleep Regulation
Contributors xi

Mary A. Carskadon, PhD Yves Dauvilliers, MD, PhD

Professor, Psychiatry and Human Behavior, Alpert Professor, Neurology, Guide Chauliac Hospital,
Medical School of Brown University; Director of Montpellier, France
Chronobiology and Sleep Research, Child and Idiopathic Hypersomnia
Adolescent Psychiatry, E.P. Bradley Hospital,
Alec J. Davidson, PhD
Providence, Rhode Island
Normal Human Sleep: An Overview Assistant Professor, Department of Neurobiology,
Daytime Sleepiness and Alertness Morehouse School of Medicine, Atlanta, Georgia
Animal Models for Disorders of Circadian Functions:
Rosalind Cartwright, PhD, FAASM Whole Organism
Professor Emeritus, Neuroscience Program, Graduate
ONeill F. DCruz, MD, MBA
College, Rush University Medical Center, Chicago,
Medical Director, Neurology, UCB Pharma, Raleigh,
Dreaming as a Mood-Regulation System North Carolina
Cardinal Manifestations of Sleep Disorders
Chien Lin Chen, MD
Tom de Boer, PhD
Lecturer, Department of Medicine, Tzu Chi University,
Assistant Professor, Molecular Cell Biology, Leiden
School of Medicine; Director, Gastrointestinal
University Medical Center, Leiden, The Netherlands
Motility, Laboratory; Attending Physician, Division of Body Temperature, Sleep, and Hibernation
Gastroenterology and Hepatology, Tzu Chi General
Hospital, Hualien, Taiwan William C. Dement, MD, PhD
Gastrointestinal Monitoring Techniques Lowell W. and Josephine Q. Berry Professor of
Psychiatry and Behavioral Sciences; Director, Sleep
Ronald D. Chervin, MD, MS
Disorders and Research Center, Stanford University
Professor of Neurology, Michael S. Aldrich Collegiate
School of Medicine, Palo Alto, California
Professor of Sleep Medicine, Department of History of Sleep Physiology and Medicine
Neurology and Sleep Disorders Center, University of Normal Human Sleep: An Overview
Michigan Medical School; Director, Sleep Disorders Daytime Sleepiness and Alertness
Center; Co-Director, Center for Sleep Science, Sleep Medicine, Public Policy, and Public Health
University of Michigan Health System, Ann Arbor,
Michigan Ronald Denis, MD
Use of Clinical Tools and Tests in Sleep Medicine Faculty of Dental Medicine, Universit de Montral,
Montral, Qubec, Canada
Peter A. Cistulli, MBBS, PhD, MBA, FRACP, FCCP Pain and Sleep
Professor, Respiratory Medicine, Head of Discipline of
Sleep Medicine, Sydney Medical School, University of Derk-Jan Dijk, PhD
Sydney; Director, Centre for Sleep Health & Professor of Sleep and Physiology, Surrey Sleep
Research, Department of Respiratory Medicine, Royal Research Centre, University of Surrey, Guildford,
North Shore Hospital; Research Leader, Sleep & United Kingdom
Circadian Group, Woolcock Institute of Medical Genetic Basis of Sleep in Healthy Humans
Research, Sydney, Australia David F. Dinges, PhD
Oral Appliances for Sleep-Disordered Breathing
Professor and Chief, Division of Sleep and
Anita P. Courcoulas, MD, MPH, FACS Chronobiology, Department of Psychiatry, University
Chief, Minimally Invasive Bariatric and General Surgery; of Pennsylvania School of Medicine, Philadelphia,
Associate Professor of Surgery, University of Pennsylvania
Pittsburgh, University of Pittsburgh Medical Center, Chronic Sleep Deprivation
Pittsburgh, Pennsylvania Circadian Rhythms in Sleepiness, Alertness, and
Obstructive Sleep Apnea, Obesity, and Bariatric Performance
Surgery Sleep Medicine, Public Policy, and Public Health

Antonio Culebras, MD, FAAN, FAHA G. William Domhoff, PhD

Professor, Neurology, SUNY Upstate Medical Research Professor of Psychology, Department of
University; Consultant, The Sleep Center, Community Psychology, University of California, Santa Cruz,
General Hospital, Syracuse, New York. Santa Cruz, California
Other Neurological Disorders Dream Content: Quantitative Findings

Charles A. Czeisler, PhD, MD, FRCP Neil J. Douglas, Kt, MD, DSc, FRCP, FRCPE
Baldino Professor of Sleep Medicine and Director, Professor of Respiratory and Sleep Medicine, Respiratory
Division of Sleep Medicine, Harvard Medical School; Medicine, University of Edinburgh; Consultant
Chief, Division of Sleep Medicine, Senior Physician, Physician, Sleep Department, Royal Infirmary of
Department of Medicine, Brigham and Womens Edinburgh, United Kingdom
Respiratory Physiology: Understanding the Control of
Hospital, Boston, Massachusetts Ventilation
The Human Circadian Timing System and
Sleep in Patients with Asthma and Chronic
SleepWake Regulation
Obstructive Pulmonary Disease
xii Contributors

Christopher L. Drake, PhD Karl A. Franklin, MD

Bioscientific Staff Investigator, Sleep Disorders and Associate professor, Surgery and Respiratory Medicine,
Research Center, Henry Ford Hospital; Associate Ume University, Ume, Sweden
Professor, Department of Psychiatry and Behavioral Coronary Artery Disease and Obstructive Sleep
Neuroscience, Wayne State University, Detroit, Apnea
Michigan Philippa Gander, PhD
Shift Work, Shift-Work Disorder, and Jet Lag
Professor and Centre Director, Sleep/Wake Research
Jack D. Edinger, PhD Centre, Massey University, Wellington, New Zealand
Clinical Professor, Department of Psychiatry and Fatigue Risk Management
Behavioral Sciences, Duke University School of Charles F.P. George, MD, FRCPC
Medicine; Senior Psychologist, Durham VA Medical Professor of Medicine, University of Western Ontario;
Center, Durham, North Carolina Director, Sleep Medicine Clinic and Laboratory,
Psychological and Behavioral Treatments for
Insomnia II: Implementation and Specific London Health Sciences Centre, London, Ontario,
Populations Canada
Sleep and Stroke
Mark W. Elliott, MD, FRCP Cognition and Performance in Patients with
Senior Lecturer, Clinical Medicine, University of Leeds, Obstructive Sleep Apnea
Department of Respiratory Medicine, St. Jamess Rachel Givelber, MD
University Hospital, Leeds, West Yorkshire, England Assistant Professor of Medicine, Division of Pulmonary,
Noninvasive Ventilation to Treat Chronic Ventilatory
Failure Allergy, and Critical Care Medicine, University of
Pittsburgh, Pittsburgh, Pennsylvania
Colin A. Espie, MappSci, PhD FBPsS, CPsychol Medical Therapy for Obstructive Sleep Apnea
Director, University of Glasgow Sleep Centre; Head,
Shelagh K. Gleeson, MD
Section of Psychological Medicine, Section of
Staff Physician, Pulmonary, Critical Care & Sleep
Psychological Medicine, Faculty of Medicine,
Disorders, Sentara Pulmonary and Critical Care
University of Glasgow; Professor, Clinical Psychology,
Specialists, Norfolk, Virginia
University of Glasgow Sleep Centre, Sackler Institute Wake-Promoting Medications: Efficacy and Adverse
of Psychobiological Research, Southern General Effects
Hospital, Glasgow, Scotland, United Kingdom
Models of Insomnia Paul B. Glovinsky, PhD
Clinical Director, St. Peters Sleep Center, St. Peters
Juliette H. Faraco, PhD Hospital, Albany, New York
Senior Research Scientist, Stanford Center for Assessment Techniques for Insomnia
Narcolepsy Research, Stanford University, Palo Alto,
California Namni Goel, PhD
Genetics of Sleep and Sleep Disorders in Humans Research Assistant Professor, Division of Sleep and
Chronobiology, Department of Psychiatry, University
Irwin Feinberg, MD of Pennsylvania School of Medicine, Philadelphia,
Psychiatry and Behavioral Sciences, University of Pennsylvania
California, Davis, Davis, California Circadian Rhythms in Sleepiness, Alertness, and
Schizophrenia Performance
Kathleen A. Ferguson, BSc, MD, FRCPC Joshua J. Gooley, PhD
Associate Professor of Medicine, Schulich School of Assistant Professor, Neuroscience and Behavioral
Medicine and Dentistry, University of Western Disorders, Duke-NUS Graduate Medical School,
Ontario, London, Ontario, Canada Singapore; Instructor of Medicine, Harvard Medical
Oral Appliances for Sleep-Disordered Breathing
School, Boston, Massachusetts
Luigi Ferini-Strambi, MD Anatomy of the Mammalian Circadian System
Associate Professor, General Psychology, Universita R. Curtis Graeber, PhD
Vita-Salve San Raffaele; Director, Sleep Disorders President, The Graeber Group, Ltd., Kirkland,
Center, Department of Neuroscience, Milan, Italy Washington
Restless Legs Syndrome and Periodic Limb Fatigue Risk Management
Movements during Sleep

Paul Franken, PhD

Matre dEnseignement et de Recherche, Center for
Integrative Genomics, University of Lausanne,
Lausanne, Switzerland
Genetic Basis of Sleep in Rodents
Contributors xiii

Ronald Grunstein, MB, BS, MD, PhD, FRACP John H. Herman, PhD
Professor of Sleep Medicine, Woolcock Institute of Professor, Departments of Psychiatry and Pediatrics,
Medical Research, The University of Sydney Medical University of Texas Southwestern Medical Center at
School; Head and Senior Staff Specialist, Centre for Dallas; Director, Sleep Disorders Center for Children,
Respiratory Failure and Sleep Disorders, Department Childrens Medical Center, Dallas, Texas; Associate
of Respiratory and Sleep Medicine, Royal Prince Professor, Departments of Psychiatry and Medicine,
Alfred Hospital, Camperdown, Australia; Chief Baylor College of Medicine; Director, Sleep Disorders
Investigator, NHMRC, Centre for Integrated Research and Research Center, Michael E. DeBakey VA
and Understanding of Sleep (CIRUS), The University Medical Center; Clinical Director, Methodist Hospital
of Sydney, Glebe, Sydney, Australia Diagnostic Sleep Laboratory, VA Medical Center
Positive Airway Pressure Treatment for Obstructive Sleep Center, Houston, Texas
Sleep Apnea-Hypopnea Syndrome Chronobiologic Monitoring Techniques
Endocrine Disorders
Victor Hoffstein, PhD, MD, FRCP(C), FCCP
Batrice Guardiola-Lematre, PhD Professor of Medicine, Department of Medicine,
Scientific Director, R & D, Scientific Direction, ADIR, University of Toronto; Staff Respirologist,
Servier International Research Institute, Courbevoie, Department of Medicine, St. Michaels Hospital,
France Toronto, Canada
Melatonin and the Regulation of Sleep and Circadian Snoring
Max Hirshkowitz, PhD
Christian Guilleminault, MD, DM, DBiol Tenured Associate Professor, Medicine, Baylor College
Professor, Division of Sleep Medicine, Stanford of Medicine, Houston, Texas; Sleep Center Director,
University School of Medicine, Palo Alto, California Sleep Disorders and Research Center, Michael E.
Narcolepsy: Diagnosis and Management DeBakey VA Medical Center, Houston, Texas
Sleep and Neuromuscular Diseases Monitoring and Staging Human Sleep
Clinical Features and Evaluation of Obstructive Sleep Monitoring Techniques for Evaluating Suspected
Apnea and Upper Airway Resistance Syndrome Sleep-Disordered Breathing
Ronald M. Harper, PhD Evaluating Sleepiness
Department of Neurobiology and the Brain Research Richard L. Horner, PhD
Institute, David Geffen School of Medicine University Canada Research Chair in Sleep and Respiratory
of California, Los Angeles, Los Angeles, California Neurobiology, Associate Professor, Departments of
Cardiovascular Physiology: Central and Autonomic Medicine and Physiology, University of Toronto,
Toronto, Canada
Allison G. Harvey, PhD Respiratory Physiology: Central Neural Control of
Professor of Clinical Psychology, Department of Respiratory Neurons and Motoneurons during
Psychology; Director, Golden Bear Sleep and Mood
Clinic, University of California, Berkeley, Berkeley, Christer Hublin, MD, PhD
California Adjunct Professor, Neurology, Helsinki University;
Insomnia: Diagnosis, Assessment, and Outcomes Assistant Chief Medical Officer, Brain@Work
Jan Hedner, MD, PhD Research Center, Finnish Institute of Occupational
Professor, Sleep Medicine, Department of Internal Health, Helsinki, Finland
Epidemiology of Sleep Disorders
Medicine, Gothenburg University, Gothenburg,
Sweden Steven R. Hursh, PhD
Coronary Artery Disease and Obstructive Sleep Adjunct Professor, Department of Psychiatry and
Apnea Behavioral Sciences, Johns Hopkins University School
Raphael C. Heinzer, MD, MPH of Medicine; President, Institutes for Behavior
Senior Lecturer and Researcher, Pulmonary Department; Resources, Inc., Baltimore, Maryland
Co-Director, Center for Investigation and Research in Fatigue and Performance Modeling
Fatigue, Performance, Errors, and Accidents
Sleep, Lausanne University Hospital, Lausanne,
Switzerland Nelly T. Huynh, PhD
Normal Physiology of the Upper and Lower Airways Faculty of Dentistry, Universit de Montral, Centre
dEtude du Sommeil, Hpital du Sacr-Coeur,
Montral, Qubec, Canada
Sleep Bruxism
xiv Contributors

Shahrokh Javaheri, MD Samuel T. Kuna, MD

Professor Emeritus of Medicine, University of Cincinnati Associate Professor of Medicine, Pulmonary, Allergy and
College of Medicine; Medical Director of Sleepcare Critical Care, Department of Medicine, University of
Diagnostics, Cincinnati, Ohio Pennsylvania; Chief, Pulmonary, Critical Care and
Sleep and Cardiovascular Disease: Present and Sleep Section, Philadelphia VA Medical Center Sleep
Future Center, Philadelphia, Pennsylvania
Cardiovascular Effects of Sleep-Related Breathing Anatomy and Physiology of Upper Airway
Disorders Obstruction
Systemic and Pulmonary Hypertension in Obstructive
Sleep Apnea Clete A. Kushida, MD, PhD, RPSGT
Heart Failure Professor, Division of Sleep Medicine; Acting Medical
Director, Stanford Sleep Medicine Center; Director,
Mark E. Josephson, MD Stanford Center for Human Sleep Research, Palo Alto,
Herman Dana Professor of Medicine, Department of California
Medicine, Harvard Medical School; Chief, Division of Clinical Features and Evaluation of Obstructive Sleep
Cardiovascular Medicine; Director, Harvard- Apnea and Upper Airway Resistance Syndrome
Thorndike Electrophysiology Institute and Arrhythmia
Service, Beth Israel Deaconess Medical Center, Hans-Peter Landolt, PhD
Boston, Massachusetts Professor, Institute of Pharmacology and Toxicology,
University of Zurich, Zurich, Switzerland
Jonathan Jun, MD Genetic Basis of Sleep in Healthy Humans
Post-Doctoral Fellow, Medicine, Pulmonary and Critical
Care, Johns Hopkins University School of Medicine, Paola A. Lanfranchi, MD
Johns Hopkins Hospital, Baltimore, Maryland Assistant Professor, Medicine, Universit de Montral;
Obstructive Sleep Apnea and Metabolic Dysfunction Cardiologist, Department of Medicine, Division of
Cardiology, Hpital du Sacr-Coeur de Montral,
Gran Kecklund, PhD Montral, Qubec, Canada
National Institute for Psychosocial Medicine, Cardiovascular Physiology: Autonomic Control in
Department of Public Health Sciences, Karolinska Health and in Sleep Disorders
Institute, Stockholm, Sweden
Sleep, Stress, and Burnout Gilles Lavigne, DMD, FRCD, PhD, HC
Dean, Faculty of Dental Medicine, Universit de
Sharon Keenan, PhD, REEGT, RPSGT, D-ABSM Montreal; Scientist Clinician, Surgery, Trauma and
Director, The School of Sleep Medicine, Inc., Palo Alto, Sleep, Hpital du Sacr-Coeur de Montral, Montral,
California Qubec, Canada
Monitoring and Staging Human Sleep Relevance of Sleep Physiology for Sleep Medicine
Kurt Kruchi Sleep Bruxism
Head, Thermophysiological Chronobiology, Centre for Pain and Sleep
Chronobiology, Psychiatric Hospital of the University
of Basel, Basel, Switzerland Kathryn Lee, RN, PhD, FAAN, CBSM
Body Temperature, Sleep, and Hibernation Professor and Associate Dean for Research; James &
Marjorie Livingston Endowed Chair; Director, T32
James M. Krueger, PhD, MDHC Nurse Research Training in Symptom Management,
Regents Professor, Sleep and Performance Research School of Nursing, University of California, San
Center, Washington State University, Pullman, Francisco Department of Family Health Care Nursing,
Washington San Francisco, California
Sleep and Host Defense Sleep Disturbances and Sleep-Related Disorders in
Meir H. Kryger, MD, FRCPC Pregnancy
Clinical Professor of Medicine, University of
Connecticut; Director of Research and Education, Patrick Leger, MD, FCCP
Gaylord Hospital, Sleep Medicine, Wallingford, Laboratoire du Sommeil, Service de Pneumologie, Pierre
Connecticut Bnite, Lyon, France
Management of Obstructive Sleep Apnea-Hypopnea Noninvasive Ventilation to Treat Chronic Ventilatory
Syndrome Failure
Restrictive Lung Disorders
Monitoring Techniques for Evaluating Suspected Christopher Li, MD, FRCP (C), DABSM
Sleep-Disordered Breathing Assistant Professor, Department of Medicine, University
of Toronto; Staff Respirologist, Medicine, St.
Andrew D. Krystal, MD, MS Michaels Hospital, Toronto, Ontario, Canada
Professor, Psychiatry and Behavioral Sciences, Duke Snoring
University School of Medicine, Durham, North
Pharmacologic Treatment: Other Medications
Contributors xv

Kenneth L. Lichstein, PhD Juan F. Masa, MD

Professor, Department of Psychology, The University of Head of Respiratory Research in Centro de Investigacin
Alabama, Tuscaloosa, Alabama Biomdica en Red de Enfermedades Respiratorias;
Insomnia: Epidemiology and Risk Factors (CIBERES); Head of Pulmonary Division, Pulmonary
Unit, San Pedro de Alcntara Hospital, Cceres, Spain
Alan A. Lowe, DMD, PhD, FRCD(C) Restrictive Lung Disorders
Professor and Chair, Division of Orthodontics, Faculty
of Dentistry, The University of British Columbia, Christina S. McCrae, PhD
Vancouver, British Columbia, Canada Associate Professor, Department of Clinical and Health
Oral Appliances for Sleep-Disordered Breathing Psychology, University of Florida, Gainesville, Florida
Insomnia: Epidemiology and Risk Factors
James G. MacFarlane, PhD
Assistant Professor of Pediatrics and Psychiatry, Jennifer McDonald, PhD
University of Toronto; Research Associate Postdoctoral Fellow, Center for Behavioral Health
(Neurology), The Hospital for Sick Children; Director Research and Services, University of Alaska
of Education, MedSleep (Network of Clinics), Anchorage, Anchorage, Alaska
Toronto, Ontario, Canada Sleep and Performance Monitoring in the Workplace:
Fibromyalgia and Chronic Fatigue Syndromes The Basis for Fatigue Risk Management

Mark W. Mahowald, MD Dennis McGinty, PhD

Professor, Neurology, University of Minnesota Medical Adjunct Professor, University of California Los Angeles;
School; Co-Director, Minnesota Regional Sleep Chief, Neurophysiology Research, VA Greater Los
Disorders Center, Hennepin County Medical Center, Angeles Healthcare System, Los Angeles, California
Minneapolis, Minnesota Neural Control of Sleep in Mammals
Sleep Forensics
Epilepsy, Sleep, and Sleep Disorders Melanie K. Means, PhD
Non-REM Arousal Parasomnias Staff Psychologist, Veterans Affairs Medical Center;
REM Sleep Parasomnias Assistant Professor, Department of Psychiatry and
Other Parasomnias Behavioral Sciences, Duke University Medical Center,
Durham, North Carolina
Jeannine A. Majde, PhD Psychological and Behavioral Treatments for
Adjunct Professor, College of Veterinary Medicine, Insomnia II: Implementation and Specific
Department of Veterinary and Comparative Anatomy, Populations
Pharmacology, and Physiology, Washington State
University, Pullman, Washington Thomas A. Mellman, MD
Sleep and Host Defense Professor, Psychiatry, Howard University, Washington,
Beth A. Malow, MD, MS Dreams and Nightmares in Posttraumatic Stress
Professor and Director, Sleep Disorders Division, Disorder
Neurology; Director, Vanderbilt Sleep Disorders Anxiety Disorders
Center, Neurology, Vanderbilt University School of Wallace Mendelson, MD
Medicine, Nashville, Tennessee Professor of Psychiatry and Clinical Pharmacology,
Approach to the Patient with Disordered Sleep
Neurologic Monitoring Techniques University of Chicago, Chicago, Illinois
Physical Examination in Sleep Medicine Hypnotic Medications: Mechanisms of Action and
Pharmacologic Effects
Rachel Manber, PhD
Emmanuel Mignot, MD, PhD
Professor, Psychiatry and Behavioral Science, Stanford
Director, Center for Sleep Sciences and Medicine,
Sleep Medicine Center, Stanford School of Medicine,
Stanford University, Palo Alto, California
Stanford University, Palo Alto, California Genetics of Sleep and Sleep Disorders in Humans
Psychological and Behavioral Treatments for
Wake-Promoting Medications: Basic Mechanisms and
Insomnia II: Implementation and Specific
Narcolepsy: Pathophysiology and Genetic
Christiane Manzini Predisposition
Research Assistant, University of Montreal Faculty of Ralph E. Mistlberger, PhD
Medicine and Dentistry; Center for the Study of Sleep Professor, Psychology, Simon Fraser University,
and Biological Rhythms, Hpital du Sacr-Coeur de Burnaby, British Columbia, Canada
Montral, Montral, Qubec, Canada Circadian Rhythms in Mammals: Formal Properties
Sleep Bruxism and Environmental Influences
Pierre Maquet, MD, PhD
Research Director FNRS, Cyclotron Research Centre,
University of Lige, Lige, Belgium
What Brain Imaging Reveals about Sleep Generation
and Maintenance
xvi Contributors

Murray A. Mittleman, MD, DrPH Seiji Nishino, MD

Associate Professor of Medicine and Epidemiology, Professor, Psychiatry and Behavioral Sciences, Stanford
Harvard Schools of Medicine and Public Health; University School of Medicine; Director, Sleep and
Director, Cardiovascular Epidemiology Research Unit, Circadian Neurobiology Laboratory, Stanford
Medicine, Cardiology, Beth Israel Deaconess Medical University School of Medicine, Palo Alto, California
Center, Boston, Massachusetts Wake-Promoting Medications: Basic Mechanisms and
Sleep-Related Cardiac Risk Pharmacology

Karen E. Moe, PhD Eric A. Nofzinger, MD

Research Associate Professor and Assistant Vice Provost Director, Sleep Neuroimaging Research Program,
for Research, Psychiatry and Behavioral Sciences, Psychiatry, University of Pittsburgh School of
University of Washington, Seattle, Washington Medicine, Pittsburgh, Pennsylvania
Menopause What Brain Imaging Reveals about Sleep Generation
and Maintenance
Harvey Moldofsky, MD, Dip Psych, FRCPC
Professor Emeritus, Department of Psychiatry; Member Louise M. OBrien, PhD
Emeritus, Institute of Medical Science, School of Assistant Professor, Sleep Disorders Center, Department
Graduate Studies, University of Toronto; Honorary of Neurology; Assistant Research Scientist,
Staff, Department of Psychiatry, Toronto Western Department of Oral and Maxillofacial Surgery,
Hospital, University Health Network; President and University of Michigan, Ann Arbor, Michigan
Medical Director, Sleep Disorders Clinics, Centre for Sex Differences and Menstrual-Related Changes in
Sleep and Chronobiology, Toronto, Ontario, Canada Sleep and Circadian Rhythms
Fibromyalgia Bruce F. OHara, PhD
Jacques Montplaisir, MD, PhD, CRCP Associate Professor, Biology, University of Kentucky,
Professor, Psychiatry and Neuroscience, Universit de Lexington, Kentucky
Montral; Director, Center for Advanced Studies in Genetic Basis of Sleep in Rodents
Sleep Medicine, Hpital du Sacr-Coeur de Montral, Eric J. Olson, MD
Montreal, Quebec, Canada Associate Professor, Medicine, College of Medicine;
Restless Legs Syndrome and Periodic Limb Co-Director, Center for Sleep Medicine, Division of
Movements during Sleep
Alzheimers Disease and Other Dementias
Pulmonary and Critical Care Medicine, Mayo Clinic,
Rochester, Minnesota
Charles M. Morin, PhD Obstructive Sleep Apnea, Obesity, and Bariatric
Professor, Psychology, Universit Laval; Director, Sleep Surgery
Research Center, Centre de Recherche Universit Mary B. OMalley, MD, PhD
Laval Robert-Giffard, Qubec, Qubec, Canada Private Practice
Psychological and Behavioral Treatments for
Insomnia I: Approaches and Efficacy
Great Barrington, Massachusetts
Wake-Promoting Medications: Efficacy and Adverse
Tore Nielsen, PhD Effects
Professor, Psychiatry, Universit de Montral; Director, William C. Orr, PhD
Dream & Nightmare Laboratory, Sleep Research President and CEO, Lynn Health Science Institute;
Center, Hpital du Sacr-Coeur de Montral, Clinical Professor of Medicine, University of
Montral, Qubec, Canada Oklahoma Health Sciences Center, Oklahoma City,
Introduction: The Changing Historical Context of
Dream Research
Gastrointestinal Physiology in Relation to Sleep
Ultradian, Circadian, and Sleep-Dependent Features
Gastrointestinal Disorders
of Dreaming
Gastrointestinal Monitoring Techniques
Dream Analysis and Classification: The Reality
Simulation Perspective Edward F. Pace-Schott, PhD
Idiopathic Nightmares and Dream Disturbances Postdoctoral Fellow, Psychology, University of
Associated with SleepWake Transitions
Disturbed Dreaming as a Factor in Medical
Massachusetts, Amherst, Massachusetts; Associate
Conditions Researcher, Psychiatry, Massachusetts General
Hospital; Faculty, Division of Sleep Medicine, Harvard
F. Javier Nieto, MPH, MD, PhD Medical School, Boston, Massachusetts
Helfaer Professor of Public Health; Professor and Chair The Neurobiology of Dreaming
of Population Health Sciences, University of
Markku Partinen, MD, PhD
Wisconsin School of Medicine and Public Health,
Adjunct Professor, Department of Clinical
Madison, Wisconsin
Systemic and Pulmonary Hypertension in Obstructive Neurosciences, University of Helsinki; Director,
Sleep Apnea Helsinki Sleep Clinic, Vitalmed Research Center,
Helsinki, Finland
Epidemiology of Sleep Disorders
Contributors xvii

Dipali Patel, BSc (Hons) Barbara A. Phillips, MD, MSPH, FCCP

Sleep and Performance Monitoring in the Workplace: Professor, Division of Pulmonary, Critical Care and
The Basis for Fatigue Risk Management Sleep Medicine; Internal Medicine, University of
Kentucky College of Medicine; Medical Director,
John H. Peever, PhD
University of Kentucky Good Samaritan Sleep Center,
Associate Professor, Cell and Systems Biology,
Lexington, Kentucky
University of Toronto, Toronto, Ontario, Canada Management of Obstructive Sleep Apnea-Hypopnea
Sensory and Motor Processing during Sleep and
Obstructive Sleep Apnea in the Elderly
Philippe Peigneux, PhD Wilfred R. Pigeon, PhD, CBSM
Professor, Chair Clinical Neuropsychology, Assistant Professor; Director, Sleep and Neurophysiology
Neuropsychology and Functional Neuroimaging Research Laboratory, Psychiatry, University of
Research Unit, Universit Libre de Bruxelles, Rochester Medical Center, Rochester, New York
Bruxelles, Belgium; Scientific Collaborator, Cyclotron Dreams and Nightmares in Posttraumatic Stress
Research Centre, Universit de Lige, Lige, Belgium Disorder
Memory Processing in Relation to Sleep
Vsevolod Y. Polotsky, MD, PhD
Yksel Peker, MD, PhD Associate Professor, Medicine, Johns Hopkins University
Associate Professor, Pulmonary Medicine, Sahlgrenska School of Medicine, Baltimore, Maryland
Academy, University of Gothenburg, Gothenburg, Obstructive Sleep Apnea and Metabolic Dysfunction
Sweden; Medical Doctor, Neurology, Rehabilitation
and Sleep Medicine, Skaraborg County Hospital, Nelson B. Powell, MD, DDS, FACS
Skvde, Sweden Department of Otolaryngology Head and Neck Surgery
Coronary Artery Disease and Obstructive Sleep and Department of Psychiatry and Behavioral Science,
Apnea Stanford University School of Medicine, Palo Alto,
Michael Perlis, PhD Surgical Management for Obstructive Sleep-
Associate Professor, Psychiatry; Director, Behavioral Disordered Breathing
Sleep Medicine Program, University of Pennsylvania,
Philadelphia, Pennsylvania Naresh M. Punjabi, MD, PhD
Models of Insomnia Associate Professor of Medicine and Epidemiology,
Department of Medicine, Division of Pulmonary and
Aleksander Perski, PhD Critical Care Medicine, Johns Hopkins University,
Assistant Professor, Stress Research Institute, Stockholm Baltimore, Maryland
University, Stockholm, Sweden Obstructive Sleep Apnea and Metabolic Dysfunction
Sleep, Stress, and Burnout
Maria Antonia Quera-Salva, MD
Michael J. Peterson, MD, PhD Director of Sleep Unit, Hpital Raymond Poincar
Assistant Professor, Psychiatry, University of Wisconsin, APHP, Paris Ouest University, Paris, France
School of Medicine and Public Health; Director of Melatonin and the Regulation of Sleep and Circadian
Consultation and Liaison Psychiatry, University of Rhythms
Wisconsin Hospital and Clinics, Madison, Wisconsin
Mood Disorders Holly Ramsawh, PhD
Assistant Project Scientist, Psychiatry, University of
Dominique Petit, PhD California, San Diego, La Jolla, California
Research Assistant, Psychiatry, Universit de Montral; Anxiety Disorders
Senior Research Assistant, Center for Advanced
Studies in Sleep Medicine, Hpital du Sacr-Coeur de Kathryn Moynihan Ramsey, PhD
Montral, Montreal, Quebec, Canada Post Doctoral Fellow, Departments of Neurobiology &
Alzheimers Disease and Other Dementias Physiology and Medicine, Northwestern University,
Evanston, Illinois
Pierre Philip, MD, PhD Animal Models for Disorders of Chronobiology: Cell
Universit Bordeaux 2, GENPPHASS, Clinique du and Tissue
Sommeil, Centre Hospitalier Universitaire Pellegrin,
Bordeaux, France Susan Redline, MD, MPH
Drowsy Driving Professor, Medicine and Center of Clinical Investigation,
Case Western Reserve University, Cleveland, Ohio
Genetics of Obstructive Sleep Apnea

Kathryn J. Reid, PhD

Research Assistant Professor, Department of Neurology,
Northwestern University Feinberg School of
Medicine, Chicago, Illinois
Circadian Disorders of the SleepWake Cycle
xviii Contributors

John E. Remmers, MD Patricia Sagaspe, PhD

Professor, Internal Medicine and of Physiology and Genpphass, Centre Hospitalier Universitaire Pellegrin,
Biophysics, University of Calgary Faculty of Medicine; Bordeaux, France; Laboratoire Exploitation,
Physician, Foothills Hospital, Calgary, Alberta, Canada Perception, Simulateurs et Simulations; Institut
Anatomy and Physiology of Upper Airway National de Recherche sur les Transports et leur
Obstruction Scurit; Laboratoire Central des Ponts et Chausses,
Paris, France
Robert W. Riley, DDS, MD, FACS Drowsy Driving
Adjunct Clinical Professor, Surgery, Department of
Otolaryngology Head and Neck Surgery; Adjunct Charles Samuels, MD
Clinical Associate Professor, Sleep Disorders Medicine, Medical Director, Centre for Sleep and Human
Department of Psychiatry and Behavioral Science, Performance; Clinical Assistant Professor, Faculty of
Stanford University School of Medicine, Stanford Medicine, University of Calgary, Calgary, Alberta,
University Sleep and Research Center, Palo Alto, Canada
California Sleep Problems in First Responders and the Military
Surgical Management for Obstructive Sleep-
Disordered Breathing Mark H. Sanders, MD, FCCP, DABSM
Professor of Medicine (retired), Division of Pulmonary,
Dominique Robert, PUPH Allergy and Critical Care Medicine, Department of
Professor, Claude Bernard University; Medical Doctor, Medicine, University of Pittsburgh School of
Intensive Care, Hospices Civils de Lyon; President, Medicine, Pittsburgh, Pennsylvania
Home Care Program, Association Lyonnaise de Sleep in Chronic Kidney Disease
Logistique Post-hospitalire, Lyon, France
Noninvasive Ventilation to Treat Chronic Ventilatory Clifford B. Saper, MD, PhD
Failure James Jackson Putnam Professor, Department of
Neurology, Program in Neuroscience, and Division of
Timothy Roehrs, PhD Sleep Medicine, Harvard Medical School; Chairman,
Professor, Department of Psychiatry and Behavioral Department of Neurology, Beth Israel Deaconess
Neuroscience, Wayne State University, School of Medical Center, Boston, Massachusetts
Medicine; Director of Research, Sleep Disorders of Anatomy of the Mammalian Circadian System
Research Center, Henry Ford Hospital, Detroit,
Michigan Aliya Sarwar, MD
Daytime Sleepiness and Alertness Assistant Professor of Neurology, Baylor College of
Medication and Substance Abuse Medicine; Associate Director, Clinical Care,
Parkinsons Disease Research, Education and Clinical
Alan M. Rosenwasser, PhD Center (PADRECC), Michael E. DeBakey VA
Professor, Department of Psychology, University of Medical Center, Houston, Texas
Maine; Cooperating Professor, School of Biology and Evaluating Sleepiness
Ecology, University of Maine, Orono, Maine
Physiology of the Mammalian Circadian System Michael J. Sateia, MD
Professor of Psychiatry, Sleep Medicine, Dartmouth
Thomas Roth, PhD Medical School; Chief, Section of Sleep Medicine,
Professor, Department of Psychiatry, Wayne State Dartmouth-Hitchcock Medical Center, Lebanon, New
University School of Medicine; Director and Division Hampshire
Head, Sleep Disorders and Research Center, Henry Treatment Guidelines for Insomnia
Ford Hospital, Detroit, Michigan; Clinical Professor,
Department of Psychiatry, University of Michigan Jose Savard, PhD
Medical School, Ann Arbor, Michigan Professor, School of Psychology, Universit Laval;
Daytime Sleepiness and Alertness Researcher, Laval University Cancer Research Center,
Pharmacologic Treatment of Insomnia: Qubec, Qubec, Canada
Benzodiazepine Receptor Agonists Sleep and Fatigue in Cancer Patients
Medication and Substance Abuse
Carlos H. Schenck, MD
Megan E. Ruiter, MA Professor, Department of Psychiatry, University of
Graduate Student, Department of Psychology, The Minnesota Medical School; Staff Psychiatrist,
University of Alabama, Tuscaloosa, Alabama Minnesota Regional Sleep Disorders Center,
Insomnia: Epidemiology and Risk Factors Hennepin County Medical Center, Minneapolis,
Benjamin Rusak, PhD, FRSC Minnesota
REM Sleep Parasomnias
Professor and Director of Research, Psychiatry,
Dalhousie University; Director, Chronobiology and Michael Schredl, PhD
Sleep Program, Capital District Health Authority; Head of Research, Sleep Laboratory, Central Institute of
Professor, Psychology, Pharmacology, Dalhousie Mental Health; Associate Professor, Department of
University, Halifax, Nova Scotia, Canada Psychology, Fakultt fr Sozialwissenschaften,
Circadian Rhythms in Mammals: Formal Properties University of Mannheim, Mannheim, Germany
and Environmental Influences Dreams in Patients with Sleep Disorders
Contributors xix

Richard J. Schwab, MD Carlyle Smith, PhD

Professor, Department of Medicine, Division of Sleep Lifetime Professor Emeritus, Psychology, Trent
Medicine, Pulmonary, Allergy and Critical Care University, Peterborough, Ontario, Canada; Adjunct
Division, University of Pennsylvania Medical Center, Professor, Centre for Neuroscience Studies, Queens
Philadelphia, Pennsylvania University, Kingston, Ontario, Canada
Anatomy and Physiology of Upper Airway Memory Processing in Relation to Sleep
Michael T. Smith, PhD
Paula K. Schweitzer, PhD Associate Professor, Psychiatry and Behavioral Sciences;
Director of Research, Sleep Medicine and Research Director, Behavioral Sleep Medicine Program, The
Center, St. Lukes Hospital, St. Louis, Missouri Johns Hopkins Bayview Medical Center, Baltimore,
Drugs That Disturb Sleep and Wakefulness Maryland
Pain and Sleep
Frdric Sris, MD
Professor, Department of Medicine, Laval University; Virend K. Somers, MD, PhD
Director, Sleep Laboratory, Pneumology, Institut de Professor of Medicine, Cardiovascular Division,
Cardiologie et de Pneumologie de Qubec, Qubec, Department of Internal Medicine; Consultant,
Qubec, Canada Division of Cardiology Department of Medicine,
Normal Physiology of the Upper and Lower Airways Mayo Clinic and Mayo Foundation, Rochester,
Barry J. Sessle, MDS, PhD, DSc(hc), FRSC Minnesota
Cardiovascular Physiology: Autonomic Control in
Professor, Canada Research Chair, Faculty of Dentistry, Health and in Sleep Disorders
University of Toronto; Consultant, Wasser Pain Cardiovascular Effects of Sleep-Related Breathing
Management Centre, Mount Sinai Hospital, Toronto, Disorders
Ontario, Canada; Adjunct Professor of Dentistry,
School of Medicine and Dentistry, University of Arthur J. Spielman, PhD
Rochester, Rochester, New York Professor, Psychology, The City College of the City
Sensory and Motor Processing during Sleep and University of New York; Associate Director, Center
Wakefulness for Sleep Medicine, Weill Cornell Medical College,
Cornell University, New York, New York; Director of
Amir Sharafkhaneh, MD, PhD, DABSM Research, Center for Sleep Disorders Medicine and
Associate Professor of Medicine, Baylor College of Research, New York Methodist Hospital, Brooklyn,
Medicine, Baylor University; Medical Director, Sleep New York
Disorders and Research Center, Michael E. DeBakey Assessment Techniques for Insomnia
VA Medical Center, Houston, Texas Insomnia: Diagnosis, Assessment, and Outcomes
Evaluating Sleepiness
Murray B. Stein, MD, MPH
Paul J. Shaw, PhD Professor, Psychiatry and Family and Preventive
Assistant Professor, Anatomy and Neurobiology, Medicine, University of California, San Diego, La
Washington University in St. Louis School of Jolla, California; Staff Psychiatrist, Psychiatry, VA San
Medicine, St. Louis, Missouri Diego Healthcare System, San Diego, California
Models of Insomnia Anxiety Disorders
Tamar Shochat, DSc Robert Stickgold, PhD
Associate Professor, Nursing, University of Haifa, Haifa, Associate Professor, Psychiatry, Harvard Medical School;
Israel Associate Professor, Psychiatry, Beth Israel-Deaconess
Insomnia in Older Adults Medical Center, Boston, Massachusetts
Margaret Shouse, PhD Why We Dream
Professor, Neurobiology, School of Medicine, University Katie L. Stone, MA, PhD
of California, Los Angeles, Los Angeles, California Senior Scientist, Research Institute, California Pacific
Epilepsy, Sleep, and Sleep Disorders Medical Center, San Francisco, California
Jerome M. Siegel, PhD Actigraphy
Professor, Psychiatry and Biobehavioral Sciences, David Robyn Stremler, RN, PhD
Geffen School of Medicine at UCLA, Los Angeles, Assistant Professor, Lawrence S. Bloomberg Faculty of
California; Chief, Neurobiology Research, VA Greater Nursing, University of Toronto; Adjunct Scientist,
Los Angeles Healthcare System-Sepulveda, Center for The Hospital for Sick Children, Toronto, Ontario,
Sleep Research, North Hills, California Canada
REM Sleep The Postpartum Period
Sleep in Animals: A State of Adaptive Inactivity
xx Contributors

Patrick J. Strollo, Jr., MD Irene Tobler, PhD

Associate Professor, Medicine and Clinical and Professor, Institute of Pharmacology and Toxicology,
Translational Research, Division of Pulmonary Allergy University of Zurich, Zurich, Switzerland
and Critical Care Medicine, University of Pittsburgh Phylogeny of Sleep Regulation
School of Medicine, Pittsburgh, Pennsylvania
Medical Therapy for Obstructive Sleep Apnea Claudia Trenkwalder, MD
Paracelsus Elena Hospital Kassel, Professor of
Ronald Szymusiak, PhD Neurology, Department of Clinical Neurophysiology,
Adjunct Professor, Departments of Medicine and University of Gottingen, Gottingen, Germany; Chair,
Neurobiology, David Geffen School of Medicine at Paracelsus-Elena Hospital, Center of Parkinsonism
UCLA; Research Scientist, VA Greater Los Angeles and Movement Disorders, Kassel, Germany
Healthcare System, Los Angeles, California Parkinsonism
Neural Control of Sleep in Mammals
Fred W. Turek, PhD
J. Taillard, MD, PhD Charles E. and Emma H. Mison Professor, Biology,
Universit Bordeaux 2, GENPPHASS, Clinique du Center for Sleep and Circadian Biology, Northwestern
Sommeil, Centre Hospitalier Universitaire Pellegrin, University, Evanston, Illinois
Bordeaux, France Introduction, Section 3
Drowsy Driving Circadian Clock Genes
Genetic Basis of Sleep in Rodents
Esra Tasali, MD Introduction: Master Circadian Clock and Master
Assistant Professor, Medicine, Pulmonary & Critical Circadian Rhythm
Care Medicine, University of Chicago Medical Center, Gastrointestinal Physiology in Relation to Sleep
Chicago, Illinois Physiology of the Mammalian Circadian System
Endocrine Physiology in Relation to Sleep and Sleep Animal Models for Disorders of Circadian Functions:
Disturbances Whole Organism

Daniel J. Taylor, PhD Mark L. Unruh, MD, MSc

Associate Professor, Department of Psychology, Assistant Professor, Medicine, Renal-Electrolyte
University of North Texas, Denton, Texas Division, University of Pittsburgh School of Medicine,
Insomnia: Epidemiology and Risk Factors Pittsburgh, Pennsylvania
Sleep in Chronic Kidney Disease
Mihai Teodorescu, MD
Clinical Assistant Professor of Medicine, University of Eve Van Cauter, PhD
Wisconsin School of Medicine and Public Health, Professor, Department of Medicine, Chicago, Illinois
Madison, Wisconsin Endocrine Physiology in Relation to Sleep and Sleep
Medical and Psychiatric Disorders and the Disturbances
Medications Used to Treat Them
Hans P.A. Van Dongen, MS, PhD
Jiuan Su Terman, PhD Research Professor and Assistant Director, Sleep and
Clinical Coordinator, Center for Light Treatment and Performance Research Center, Washington State
Biological Rhythms, Columbia University Medical University, Spokane, Washington
Center, New York, New York Circadian Rhythms in Sleepiness, Alertness, and
Light Therapy Performance
Fatigue and Performance Modeling
Michael Terman, PhD Fatigue, Performance, Errors, and Accidents
Professor, Psychiatry, College of Physicians & Surgeons,
Bradley V. Vaughn, MD
Columbia University; Research Scientist VI, Clinical
Professor, Neurology and Biomedical Engineering,
Chronobiology, New York State Psychiatric Institute;
University of North Carolina School of Medicine,
Director, Center for Light Treatment and Biological
Chapel Hill, North Carolina
Rhythms, New York Presbyterian Hospital, Columbia Cardinal Manifestations of Sleep Disorders
University Medical Center; President, Center for
Environmental Therapeutics, New York, New York Richard L. Verrier, PhD, FACC
Light Therapy Associate Professor of Medicine, Harvard Medical
School, Department of Medicine, Division of
Michael J. Thorpy, MD
Cardiovascular Medicine, Beth Israel Deaconess
Director, Sleep/Wake Disorders Center, Montefiore
Medical Center, Boston, Massachusetts
Medical Center, and Albert Einstein College of Cardiovascular Physiology: Central and Autonomic
Medicine New York, New York Regulation
Classification of Sleep Disorders Sleep-Related Cardiac Risk
Cardiac Arrhythmogenesis during Sleep:
Mechanisms, Diagnosis, and Therapy
Contributors xxi

Bryan Vila, PhD Nancy J. Wesensten, PhD

Professor and Director, Simulated Hazardous Behavioral Biology Branch, Center for Military
Operational Tasks Laboratory, Sleep and Performance Psychiatry and Neuroscience, Walter Reed Army
Research Center, Washington State University, Institute of Research, Silver Spring, Maryland
Spokane, Washington Pharmacologic Management of Performance Deficits
Sleep Problems in First Responders and the Military Resulting from Sleep Loss and Circadian
Martha Hotz Vitaterna, PhD
Research Associate Professor, Center for Functional David P. White, MD
Genomics, Northwestern University Center for Clinical Professor, Medicine, Harvard Medical School;
Functional Genomics, Evanston, Illinois Clinical Professor of Medicine, Brigham and Womens
Circadian Clock Genes Hospital, Boston, Massachusetts
Central Sleep Apnea and Periodic Breathing
James K. Walsh, PhD
Visiting Professor, Department of Psychiatry, Stanford Amy R. Wolfson, PhD
University School of Medicine, Palo Alto, California; Professor of Psychology; Associate Dean for Faculty
Executive Director and Senior Scientist, Sleep Development, Office of the Dean, College of the Holy
Medicine and Research Center, St. Lukes Hospital; Cross, Worcester, Massachusetts
Adjunct Professor, Department of Psychology, Saint The Postpartum Period
Louis University, St. Louis, Missouri Kenneth P. Wright, Jr., PhD
Sleep Medicine, Public Policy, and Public Health
Pharmacologic Treatment of Insomnia:
Assistant Professor, University of Colorado, Boulder,
Benzodiazepine Receptor Agonists Colorado
Shift Work, Shift-Work Disorder, and Jet Lag
Arthur S. Walters, MD
Chien-Ming Yang, PhD
Professor and Associate Director, Sleep Medicine,
Assistant Professor, Department of Psychology, National
Neurology, Vanderbilt University School of Medicine,
Chengchi University, Taipei, Taiwan
Nashville, Tennessee Assessment Techniques for Insomnia
Restless Legs Syndrome and Periodic Limb
Movements during Sleep Terry Young, MS, PhD
Erin J. Wamsley, PhD Professor, Department of Population Health Sciences,
Instructor, Psychiatry, Harvard Medical School; Epidemiology Section, University of Wisconsin School
Instructor, Psychiatry, Beth Israel Deaconess Medical of Medicine and Public Health, Madison, Wisconsin
Systemic and Pulmonary Hypertension in Obstructive
Center, Boston, Massachusetts Sleep Apnea
Why We Dream
Antonio Zadra, PhD
Terri E. Weaver, PhD, RN, FAAN
Professor, Psychology, Universit de Montral; Research
Professor and Dean, University of Illinois at Chicago
Professor, Centre dEtude du Sommeil et des Rythmes
College of Nursing, Chicago, Illinois
Cognition and Performance in Patients with Biologiques, Hpital du Sacr-Coeur de Montral,
Obstructive Sleep Apnea Montral, Qubec, Canada
Dream Content: Quantitative Findings
John V. Weil, MD Idiopathic Nightmares and Dream Disturbances
Emeritus Professor, Medicine, University of Colorado Associated with SleepWake Transitions
School of Medicine, Denver, Colorado Phyllis C. Zee, MD, PhD
Respiratory Physiology: Sleep at High Altitudes
Professor, Neurology; Director, Sleep Disorders Center,
Ian D. Weir, DO Northwestern University Feinberg School of
Associate Director, the Sleep Disorders Center; Director Medicine, Chicago, Illinois
of the Insomnia Center, Norwalk Hospital, Norwalk, Circadian Disorders of the SleepWake Cycle
Connecticut Marco Zucconi, MD
Wake-Promoting Medications: Efficacy and Adverse
Effects Professor, Department of Neurology, Sleep Disorders
Center, H San Raffaele Scientific Institute and
Andrew Wellman, MD Hospital, Universit Vita-Salute San Raffaele, Milan,
Instructor, Medicine, Division of Sleep Medicine, Italy
Harvard Medical School, Boston, Massachusetts Pain and Sleep
Central Sleep Apnea and Periodic Breathing
Foreword xxiii


Medicine has only recently discovered the importance of It is an honor for us, representing the Sleep Research
sleep, and how sleep symptoms can be the canary in the Society, to help introduce the 5th edition of Principles and
mine of serious medical and psychiatric problems that can Practice of Sleep Medicine. This volume appears nearly fifty
affect all people. I can attest firsthand about the impor- years after the first professional sleep research meeting in
tance of sleep, and how symptoms affecting sleep can the United States on March 25 and 26, 1961, a meeting
impact a persons life. I was the Canadian Force Com- which directly led to the formation of the Sleep Research
mander of the United Nations Assistance Mission for Society. According to records of Al Rechtschaffen, main-
Rwanda between October 1993 and August 1994. During tained in the University of Chicago Library, the first
that time, a genocide resulted in the deaths of 800,000 meeting was titled the Conference on Research in EEG,
people and I was an eyewitness having heard, smelled, Sleep and Dreams. Two days of scientific sessions included
seen, and touched thousands and thousands of mutilated topics such as: Methods and Merits of Various Systems
bloated bodies of innocent civilians while trying to arrange of Scoring, Equipment and Technical Problems and
a peace during a civil war while much of the world stood The Relation of EEG to Verbal Report. As the session
idly by. My sleep suffered, my health suffered, and I devel- titles suggest, in 1961 sleep scientists were necessarily
oped the symptoms of posttraumatic stress disorder. As the focused upon some of the most basic tenants of research:
mission was winding down observation, measurement, standardization, and tech
After prayers, I climbed into my vehicle and took off nology. It seems very unlikely that any of the thirty-six
without telling anyone. It wasnt the first time. I had begun scientists in attendance, including Bill Dement, would
to suffocate in the headquarters, with its endless stream of have envisioned the exponential growth of knowledge
problems and demands. I had been inventing trips to get about sleep and its disorders represented in the many pages
me away from it, deciding that I had to see the troops in of the current volume.
the field or just tour the country. In every village, along Sleep research has evolved to include the breadth of
every road, in every church, in every school were unburied modern scientific approaches, and sleep medicine is
corpses. My dreams at night became my reality of the day, germane to most medical specialties and public health
and increasingly I could not distinguish between the two. concerns. In the 149 chapters of this text (including more
By this point, I wasnt bothering to make excuses than 50 new chapters and new sections on Genetics, Occu-
anymore to disguise my quest for solitude. I would just pational Sleep Medicine, Sleep Medicine in Older People)
sneak away and then drive around thinking all manner of experts describe the intricate mechanisms of biological
black thoughts that I couldnt permit myself to say to timing, the genetic polymorphisms conferring risk for
anyone for fear of the effect on the morale of my troops. sleep disorders, sleep-immune interactions, the morbidity
Without my marking the moment, death became a desired and mortality risks of sleep apnea, and the impact of sleep
option. I hoped I would hit a mine or run into an ambush disturbance on workplace and transportation safety, just
and just end it all. I think some part of me wanted to join to highlight as few areas. We commend the authors on
the legions of the dead, whom I felt I had failed. I could their excellent contributions which will serve to educate
not face the thought of leaving Rwanda alive after so many and inspire practitioners, researchers and students for
people had died. On my travels around the country, whole years to come.
roads and villages were empty, as if theyd been hit by a The Sleep Research Society congratulates Drs. Kryger,
nuclear bomb or the bubonic plague. You could drive for Roth, and Dement on the publication of this very impres-
miles without seeing a single human being or a single sive 5th edition of Principles and Practice of Sleep Medicine
living creature. Everything seemed so dead. and thanks them for undertaking the important service of
From, Romo A. Dallaire, Shake Hands with the Devil: identifying the latest advances in the field and compiling
The Failure of Humanity in Rwanda. Carroll & Graf the body of knowledge represented herein.
Publishers, New York. 2003. pp 499-500.
This is the first medical textbook that focuses on the James K. Walsh, PhD
President, Sleep Research Society
sleep disorders that affect everyone, and that also includes Executive Director and Senior Scientist
the problems of first responders and the military and Sleep Medicine and Research Center
teaches doctors about how post traumatic stress disorder St. Lukes Hospital
impacts sleep. I congratulate the editors. Chesterfield, Missouri

Lt. Gen the Hon. Romo Dallaire, OC, CMM, Clifford B. Saper, MD, PhD
GOQ, MSC, CD, (Retd), Senator, Canada Past-president, Sleep Research Society
James Jackson Putnam Professor of Neurology and Neuroscience,
Harvard Medical School and Beth Israel Deaconess Medical Center
Boston, Massachusetts

xxiv Foreword

The success of any field of medicine is often directly pro- gold standard over the span of 20 years. Our field is deeply
portional to the scope and comprehensiveness of the indebted to their dedication, hard work, and diligence.
knowledge base available to physicians, scientists, trainees,
and the general public. For the field of sleep medicine, we Clete A. Kushida, MD, PhD, RPSGT
President, American Academy of Sleep Medicine
are fortunate in that there continues to be dramatic growth Director, Stanford Center for Human Sleep Research
in this knowledge, derived from both patient care and Stanford University, California
clinical/basic research. It is revealing when we step back
and reflect on how this knowledge base has developed in The American Sleep Apnea Association (ASAA) congratu-
so short a time: It has been less than 60 years since the lates the editors of the Principles and Practice of Sleep
discovery of rapid eye movement (REM) sleep, which initi- Medicine on the publication of the fifth edition of this
ated the organized, scientific study of sleep, and barely 25 invaluable guide for those practicing sleep medicine. As
years since the invention of continuous positive airway the field evolves, it has become a key resource for those
pressure (CPAP), which comprised the first effective treat- responsible for diagnosing and treating those with sleep
ment for obstructive sleep apnea. In this short time, the disorders like sleep apnea. The long-term management of
sleep field has expanded to the point where we have almost patients requires a partnership between the doctor who
1,900 accredited sleep centers and laboratories in the will learn from this volume the scientific basis of the field
United States and over 9,000 members of the American and the patient with a chronic illness who must also learn
Academy of Sleep Medicine. Our field has blossomed to about their condition.
the point where it is truly interdisciplinary, comprising For the patient and their families seeking to understand
specialists from the areas of pulmonary medicine, neurol- sleep apnea prior to diagnosis, and often once they are
ogy, psychiatry, internal and family medicine, pediatrics, treated, they frequently turn to the ASAA, since 1990 the
psychology, otolaryngology, and others. Exciting break- only patient interest organization in the USA dedicated to
throughs in sleep research have impacted other disciplines sleep apnea education, support and advocacy. The ASAA
of science and research as well, and it is not unusual for produces educational materials and includes support
sleep medicine specialists to collaborate with other diverse groups for sleep apnea patients and their families operating
fields of medicine such as cardiology, endocrinology, and under the name A.W.A.K.E. that stands for Alert, Well,
immunology. And Keep Energetic. The network of the more than 300
Despite our amazing growth, there are still many ques- A.W.A.K.E. groups has locations around the United
tions yet to be answered, including the holy grail of our States, Canada and overseas.
field: the function of sleep. To explore these questions, the As elucidated in this book sleep apnea is associated with
field requires a continued supply of dedicated and talented several important comorbidities and thus the ASAA will
researchers in both the clinical and basic sciences. In addi- expand contacts with other patient groups concerned such
tion, funding from the government, industry, and founda- as heart disease and type 2 diabetes.
tions; support from institutions; and strong mentorship by The fifth edition of Practice and Principles of Sleep
experienced investigators are important cornerstones to a Medicine is for the education and support of the caregiver
successful independent research career. As members of the and the American Sleep Apnea Association for the educa-
field, we must collectively strive to ensure that funding, tion and support of the apnea patient and their family.
support, and mentorship continues in order to ensure
success of our field, even in times of economic downturns Edward Grandi
Executive Director
and increased competition from other fields. For without American Sleep Apnea Association
breakthroughs in research, there wont be new diagnostic
tools, medications, or treatments to help us manage the On behalf of the National Sleep Foundation, it is my
nearly 90 different sleep disorders that we have identified pleasure and honor to help introduce the fifth edition of
thus far. Principles and Practice of Sleep Medicine, the latest update to
The growth of our field and the exploration of critical what has over the past 21 years properly come to be
research areas cannot exist without adequate education and regarded as the preeminent reference text on sleep medi-
training of our young clinicians and investigators. We are cine. I congratulate the editors, Drs. Kryger, Roth, and
indeed privileged that we have excellent resources available Dement, for this new edition representing as it does their
that enable trainees to learn more about sleep and sleep long-standing and continuing commitment to the expan-
medicine. For countless numbers of students, Principles sion of the field of sleep medicine and science, and their
and Practice of Sleep Medicine has served as the primary collective talent for presentation and dissemination of
textbook, study material for the sleep medicine board cer- sleep-related knowledge.
tification examination, and/or the basic resource for any When the first edition was published, the discipline of
sleep-related condition or question about sleep. Often sleep medicine was in its infancy. Today, perusing previous
fondly referred to simply as P&P, it continues to rise in editions of PPSM is akin to marveling over the growth of
prominence and demand. Ive had the great pleasure to a child by leafing through old family photo albums:
learn from and collaborate with Drs. Kryger, Roth, and Although it is understood on an intellectual level that
Dement, and not only are they among the top clinicians growth and maturation have been occurring continuously,
and scientists within our field, but they have continued to the intermittent nature of the photographic record serves
produce a sleep medicine reference that has remained the to crystallize those changes in startlingly sharp relief. So it
Foreword xxv

is with PPSM and the history of our field. Each successive a phase of high energy and escalating complexity with a
edition of PPSM has faithfully reflected the progress made few early hints of maturity, but a phase primarily charac-
in the intervening years, in a narrative manner that makes terized by the exhilaration of a future that still seems hori-
clear the process by which high-quality scientific research zonless. For those wishing an understanding of the roots
on sleep and its disorders accrues and is translated into the of sleep science, desiring a comprehensive overview of the
practice of sleep medicine. current state of the art of sleep medicine, or yearning for
Perusal of the current edition will, I think, make it a glimpse of what the future of sleep science holds, just
apparent that our field has now grown past infancy, spurted start turning the pages.
through childhood, and entered its adolescence: to be sure,
Thomas J. Balkin, Ph.D.
Chairman of the Board of Directors
National Sleep Foundation
Washington, DC

It has been a quarter century since we started work on the 400 articles published using the keywords sleep apnea
first edition of Principles and Practice of Sleep Medicine. or sleep apnoea, by 2010 about 4000 articles are being
At that time there was a great deal of concern from col- published per year. Entirely new fields now have important
leagues that there simply wasnt enough information avail- sleep components including geriatrics, occupational
able for a book, and there simply wasnt enough of a health, womens health, and there has been an explosion
market to justify all that effort. In fact, several colleagues of knowledge just in the past few years in the genetics,
whom we respect a great deal tried to talk us out of a molecular biology, and neuroanatomy and neurochemistry
textbook because they felt it would have been a waste of of sleep.
time. At that time there were roughly 2000 members of To ensure that the information is up-to-date, this
the American Sleep Disorders Association (now the Amer- edition has five entirely new sections (Sleep Medicine in
ican Academy of Sleep Medicine). Information about sleep the Elderly, Occupational Sleep Medicine, Physiology
medicine was transmitted and shared via telephone (almost in Sleep, Genetics of Sleep, and Sleep Mechanisms
everybody knew everybody practicing sleep medicine), and Phylogeny). There are more than 50 brand new chap-
articles, scientific meetings and one or two journals that ters and the remaining chapters have all been updated.
regularly published articles about sleep. There are new sleep stage and respiratory event scoring
How things have changed. In the United States of rules and this is perhaps the first major sleep textbook that
America, within about 25 miles of anybodys home there incorporates these rules. A companion volume, Atlas of
are about 5 to 10 accredited sleep disorders centers, and Clinical Sleep Medicine, contains hundreds of examples of
also some unaccredited centers. There are about 1900 polysomnographic data, and videos of interviews with
accredited sleep disorders centers. The American Academy patients with sleep disorders.
of Sleep medicine has approximately 9000 members and Besides the dramatic increase in content, we had to
counting. The number of doctors who practice sleep medi- deal with the changes in the delivery of information.
cine is unknown, but there are 3,445 who have specialty Books were the way to go in the 1980s, content on
certification in sleep awarded by the American Board of CDs were the rage in the 1990s, but web-enabled books
Sleep Medicine. This has been supplanted by a new certi- appear to be what people want beyond 2010. The third
fication exam administered by the American Board of edition almost came out solely on a CD. It turned out
Medical Specialties. The first time the exam was given, 724 that CDs were a passing trend and people hated reading
physicians passed the examination. In addition, doctors large amounts of text on a computer screen. What
from other specialties who may not have had any training doctors want now is a book that they can access in
in sleep medicine (for example those trained in pulmonary, their library, via the internet and even on their handheld
neurology, otolaryngology) frequently evaluate and device and tablet. We have been listening to caregivers
manage patients with sleep disorders. who use this content to help them better care for their
There are now eleven journals that publish articles patients and hope that you will continue to provide
exclusively about sleep. Whereas in 1985 there were about feedback.

Thomas Roth Meir H. Kryger William C. Dement


This is now about the 25th year since the editors started We would also like to thank the staff at Elsevier who
to work on the first edition of the Principles and Practice have been so supportive of sleep medicine. Dolores Meloni
of Sleep Medicine. Twenty-five years is a generation and has been fabulous and supportive of the book. Angela
during that time it is safe to say that thousands of profes- Norton and Anne Snyder shepherded the production of
sionals and editorial staff have worked on this book, and it content for the book and Bob Browne has quietly and
is impossible to thank each and every one, as much as we efficiently managed the web content. We especially wish
would love to. to thank Sarah Wunderly who took over production of the
We would like to thank the current and previous section book at the home stretch and calmly, expertly, and patiently
editors. They are the leaders in their fields and they bring delivered it to the printers.
their vision to the book. They fashioned the content so it We want to thank the book designers and the produc-
is of greatest value and relevance to the reader, and ulti- tion staff for producing such an elegant volume.
mately the patient. We would like to thank all the chapter Finally, we want to thank the many people, readers, who
contributors for their hard work. Producing a new chapter have given us such excellent ideas and feedback.
or refreshing and updating a previous one is not easy
because scientific papers on sleep are published daily and Meir H. Kryger, Thomas Roth,
it is a difficult task to remain completely up-to-date. and William C. Dement

Abbreviations xxxi


AASM: American Academy of Sleep Medicine DSM-IV: Diagnostic and Statistical Manual of Mental
ACC: anterior cingulate cortex Disorders, 4th Edition
Ach: acetylcholine DSPS: delayed sleep phase syndrome
ACTH: adrenocorticotropic hormone DSPT: delayed sleep phase type
AD-ACL: Activation-Deactivation Adjective Check List DTs: delirium tremens
ADHD: attention-deficit-hyperactivity disorder DU: duodenal ulcer
AHI: apnea-hypopnea index ECG: electrocardiogram/electrocardiographic
AIM: ancestry informative marker EDS: excessive daytime sleepiness
AMPA: -amino-3hydroxy-5-methylisozazole-4- EEG: electroencephalogram, electroencephalographic
propionic acid EMG: electromyogram
AMPK: adenosine-monophosphate-activated protein ENS: enteric nervous system
kinase EOG: electrooculogram
AMS: acute mountain sickness EPS: extrapyramidal side effects (s)
ANS: autonomic nervous system EPSP: excitatory postsynaptic potential
ASPS: advanced sleep phase syndrome ERP: event-related potential
ASPT: advanced sleep phase type ESS: Epworth Sleepiness Scale
AW: active wakefulness FAID: Fatigue Audit InterDyne
BA: Brodman area 18
FDG: 2-deoxy-2-[18F]fluoro-d-glucose
BAC: blood alcohol content F-DOPA: 6-[18F]fluoro-l-dopa
BCOPS: Buffalo Cardio-Metabolic occupational Police FEV1: forced expiratory volume in 1 second
Stress FIRST: Ford Insomnia Response to Stress Test
BF: basal forebrain fMRI: functional magnetic resonance imaging
BMAL1: brain and muscle ARNT-like FOQA: flight operations quality assurance
BMI: body mass index FOSQ: Functional Outcomes of Sleep Questionnaire
BNST: bed nucleus of the stria terminalis FRA: Federal Railroad Administration
BPD: biliopancreatic diversion FRC: functional residual capacity
BPDDS: Biliopancreatic diversion with duodenal switch FSIVGTT: frequently sampled intravenous glucose
BzRA: benzodiazepine receptor agonist tolerance test
CAD: coronary artery disease GABA: gamma-aminobutyric acid
CAPS: cyclic alternating pattern sequence(s) GAD: generalized anxiety disorder
CBT: cognitive behavior therapy GAHMS: genioglossus advancement, hyoid myotomy,
CHF: congestive heart failure and suspension
CI: confidence interval GCD: global cessation of dreaming
CPS/HHPRI: Calgary Police Service Health and Human GER: gastroesophageal reflux
Performance Research Initiative GHB: gamma-hydroxybutyrate
COMT: catechol-O-methyltransferase GHRH: growth hormone-releasing hormone
COPD: chronic obstructive pulmonary disease GWA: genome wide association
CPAP: continuous positive airway pressure 5-HIAA: 5-hydroxyindole acetic acid
CRP: C-reactive protein 5-HT: hydroxytryptamine (serotonin)
CRY: cryptochrome HAPE: high-altitude pulmonary edema
CSN: cold-sensitive neuron Hcrt: hypocretin
CYP: cytochrome P-450 HDI: hypnotic-dependent insomnia
DA: dopamine HDL: high density lipoprotein
DAT: dopamine transporter HIF: hypoxia inducible factor
DBP: D-element binding protein HIV: human immunodeficiency virus
DD: constant dark HLA: human leukocyte antigen
DIM: digital integration mode HOMA: homeostasis model assessment
DLMO: dim-light melatonin onset HPA: hypothalamic-pituitary-adrenal axis
DLPFC: dorsolateral prefrontal cortex HRV: heart rate variability
DMD: Duchennes muscular dystrophy HVA: homovanillic acid
DSISD: Duke Structured Interview for Sleep HWHSGPS: Harvard Work Hours and Safety Group
Disorders Police Study

xxxii Abbreviations

ICD: International Classification of Diseases NREM: nonrapid eye movement, non-REM

ICD-9-CM: International Classification of Diseases, 9th OCD: obsessive-compulsive disorder
revision, Clinical Modification OFC: orbitofrontal cortex
ICS-10: International Classification of Diseases, 10th 6-OHDA: 6-hydroxydopamine
revision OHS: obesity-hypoventilation syndrome
ICSD-2: International Classification of Sleep Disorders, OR: odds ratio
Revised OSA: obstructive sleep apnea
ICV: intracerebroventricular OSAHS: obstructive sleep apneahypopnea syndrome
IEG: immediate early gene OSAS: obstructive sleep apnea syndrome
IGL: intergeniculate leaflet PACU: postanesthesia care unit
IL: interleukin PCOS: polycystic ovary syndrome
ILD: interstitial lung disease PEEP: positive end-expiratory pressure
IPSP: inhibitory postsynaptic potential PER: period
ISI: Insomnia Severity Index PET: positron emission tomography
kd: kilodalton PGO: ponto-geniculo-occipital [spike]
KSS: Karolinska Sleepiness Scale PIA: pontine inhibitory area
LAUP: laser-assisted uvulopalatoplasty PLMS: periodic limb movements during sleep (or PLM)
LD: light-dark PMDD: premenstrual dysphoric disorder
LDL: low density lipoprotein PNI: people not having insomnia
l-dopa: l-dihydroxyphenylalanine, levodopa POA: preoptic area
LG: lateral geniculate POMS: Profile of Mood States
LL: constant light POSSR: Patrol Officers Shift Schedule Review
LOC: left outer canthus PR: prevalence ratio
LPA (or LPOA): lateral preoptic area PRC: phaseresponse curve
LSAT: lowest oxyhemoglobin saturation PSG: polysomnography, polysomnographic
LTIH: long-term intermittent hypoxia PSQI: Pittsburgh Sleep Quality Index
MAO: monoamine oxidase PTSD: posttraumatic stress disorder
MAOI: monamine oxidase inhibitor PVN: paraventricular nucleus
MDA: methylenedioxyamphetamine PVT: psychomotor vigilance test
MDD: major depressive disorder PWI: people with insomnia
MDMA: methylenedioxymethamphetamine (ecstasy) PWOP: people who did not report having the medical
MDP-LD: muramyl dipeptide problem
N-actyl-muramyl-l-alanyl-d-isoglutamine PWP: people who reported have the medical problem
MEG: magnetoencephalography QTL: quantitative trait loci (or locus)
MI: myocardial infarction QW: quiet wakefulness
MMC: migrating motor complex RBD: REM sleep behavior disorder
MMO: maxillary and mandibular osteotomy RDC: research diagnostic criteria
MnPN: median preoptic nucleus RDI: respiratory disturbance index
MNSA: muscle nerve sympathetic vasomotor activity REM: rapid eye movement
MPA or (MPOA): medial preoptic area RERA: respiratory effort related arousal
MPA: medroxyprogesterone acetate RFA: radiofrequency ablation
MRA: mandibular repositioning appliance RHT: retinohypothalamic tract
MSLT: Multiple Sleep Latency Test RI: recombinant inbred
MWT: Maintenance of Wakefulness Test Rin: membrane input resistance
NAD: nicotinamide adenine nucleotide RIP: respiratory inductive plethysmography
NAMPT: nicotinamide phosphoribosyltransferase RLS: restless legs syndrome
NASH: nonalcoholic steatohepatitis RMMA: rhythmic masticatory motor activity
NCEP: National Cholesterol Education Program ROC: right outer canthus
NCSDR: National Center on Sleep Disorders Research ROS: reactive oxygen species (sing. and pl.)
NE: norepinephrine RR: risk ratio
NET: norepinephrine transporter RT: reaction time
NFLD: nonalcoholic fatty liver disease RYGB: Roux-en-Y gastric bypass
NFB: nuclear factor kappa B SAFTE: Sleep, Activity, Fatigue, and Task Effectiveness
NHANES: national Health and Nutrition Examination (model)
Survey SCD: stearoyl coenzyme A desaturase
NIH: National Institutes of Health SCID: Structured Clinical Interview for Diagnosis
NIPPV: nasal intermittent positive-pressure ventilation SCN: suprachiasmatic nucleus
NK: natural killer [cell] SCT: sleep compression therapy
NMDA: N-methyl-d-aspartate SDB: sleep-disordered breathing
NO: nitric oxide SE%: sleep efficiency percentage
NPPV: noninvasive positive-pressure ventilation SEMs: small eye movements
NPT: nocturnal penile tumescence SIDS: sudden infant death syndrome
Abbreviations xxxiii

SIT: suggested immobilization test TAT: time above threshold

SND: synucleinopathic disorders TCA: tricyclic antidepressant
SNP: single nucleotide polymorphism THH: terrifying hypnagogic hallucination
SOL: sleep-onset latency TLR: Toll-like receptor
SOREM: sleep-onset REM TMJ: temporomandibular joint
SOREMP: sleep-onset REM period TNF: tumor necrosis factor
SP: sleep paralysis TRD: tongue-retaining device
SPM: statistical parametric mapping UARS: upper airway resistance syndrome
SRE: sleep-related erection UNS: Ullanlinna Narcolepsy Scale
SREBP: sterol regulatory element binding proteinSRED: UPF: uvulopalatal flap
sleep-related eating disorder UPPP: uvulopalatopharyngoplasty
SRT: sleep restriction therapy VIP: vasoactive intestinal peptide
SSEP: somatosensory evoked potential VLDL: very low density lipoprotein
SSS: Stanford Sleepiness Scale VLPO: ventrolateral POA (preoptic area)
SSRI: selective serotonin reuptake inhibitor VMAT2: vascular monoamine transporter-2
SWA: slow wave activity VTA: ventral tegmental area
SWD: shift work disorder/shift work sleep disorder WASO: wake after sleep onset
SWS: slow wave sleep WSN: warm-sensitive neuron
Ta: ambient temperature ZCM: zero crossing mode
Normal Sleep and Section

Its Variations 1
Timothy Roehrs

1 History of Sleep Physiology and Medicine 4 Daytime Sleepiness and Alertness

2 Normal Human Sleep: An Overview 5 Acute Sleep Deprivation
3 Normal Aging 6 Chronic Sleep Deprivation

History of Sleep Physiology Chapter

and Medicine
William C. Dement 1
Abstract more attention from poets and writers. Some of the worlds
There has been great scientific interest in sleep for well over greatest thinkers, such as Aristotle, Hippocrates, Freud, and
a century, with the discoveries of the electrical activity of the Pavlov, have attempted to explain the physiologic and psycho-
brain, the arousal systems, the circadian system, and rapid logical bases of sleep and dreaming. However, it is not the
eye movement sleep. In spite of these discoveries, the field of purpose of this chapter to present a scholarly review across
sleep medicine has existed for only about 4 decades. The evo- the ages about prehistoric, biblical, and Elizabethan thoughts
lution of the field required clinical research, development of and concerns regarding sleep or the history of mans enthrall-
clinical services, and changes in society and public policy that ment with dreams and nightmares. This has been reviewed
recognized the impact of sleep disorders on society. The field by others.1 What is emphasized here for the benefit of the
is still evolving as new disorders are being discovered, new student and the practitioner is the evolution of the key con-
treatments are being delivered, and basic science is helping cepts that define and differentiate sleep research and sleep
us understand the complexity of sleep and its disorders. medicine, crucial discoveries and developments in the forma-
Interest in sleep and dreams has existed since the dawn of tive years of the field, and those principles and practices that
history. Perhaps only love and human conflict have received have stood the test of time.

SLEEP AS A PASSIVE STATE ronment. No real distinction was seen between sleep and
Sleep is the intermediate state between wakefulness and other states of quiescence such as coma, stupor, intoxica-
death; wakefulness being regarded as the active state of all tion, hypnosis, anesthesia, and hibernation.
the animal and intellectual functions, and death as that of The passive to active historical dichotomy is also given
their total suspension.2 great weight by the modern investigator J. Allan Hobson.3
The foregoing is the first sentence of The Philosophy In the first sentence of his book, Sleep, published in 1989,
of Sleep, a book by Robert MacNish, a member of the he stated that more has been learned about sleep in the
faculty of physicians and surgeons of Glasgow; the first past 60 years than in the preceding 6,000. He went on,
American edition was published in 1834 and the Scottish In this short period of time, researchers have discovered
edition somewhat earlier. This sentence exemplifies the that sleep is a dynamic behavior. Not simply the absence
overarching historical conceptual dichotomy of sleep of waking, sleep is a special activity of the brain, controlled
research and sleep medicine, which is sleep as a passive by elaborate and precise mechanisms.
process versus sleep as an active process. Until the discov- Dreams and dreaming were regarded as transient, fleet-
ery of rapid eye movements and the duality of sleep, sleep ing interruptions of this quiescent state. Because dreams
was universally regarded as an inactive state of the brain. seem to occur spontaneously and sometimes in response
With one or two exceptions, most thinkers regarded sleep to environmental stimulation (e.g., the well-known alarm
as the inevitable result of reduced sensory input with the clock dreams), the notion of a stimulus that produces the
consequent diminishment of brain activity and the occur- dream was generalized by postulating internal stimulation
rence of sleep. Waking up and being awake were consid- from the digestive tract or some other internal source.
ered a reversal of this process, mainly as a result of Some anthropologists have suggested that notions of spiri-
bombardment of the brain by stimulation from the envi- tuality and the soul arose from primitive peoples need to

4 PARTI / Section 1 Normal Sleep and Its Variations

explain how their essence could leave the body temporarily demonstrated electrical rhythms in the brains of animals.
at night in a dream and permanently at death. The centennial of his discovery was commemorated at the
In addition to the mere reduction of stimulation, a host 15th annual meeting of the Association for the Psycho-
of less popular theories were espoused to account for the physiological Study of Sleep convening at the site of the
onset of sleep. Vascular theories were proposed from discovery, Edinburgh, Scotland.
the notion that the blood left the brain to accumulate in However, it was not until 1928 when the German psy-
the digestive tract, and from the opposite idea that sleep chiatrist Hans Berger6 recorded electrical activity of the
was due to pressure on the brain by blood. Around the human brain and clearly demonstrated differences in these
end of the 19th century, various versions of a hypnotoxin rhythms when subjects were awake or asleep that a real
theory were formulated in which fatigue products (toxins scientific interest commenced. Berger correctly inferred
and the like) were accumulated during the day, finally that the signals he recorded, which he called electroen-
causing sleep, during which they were gradually elimi- cephalograms, were of brain origin. For the first time, the
nated. It had, of course, been observed since biblical times presence of sleep could be conclusively established without
that alcohol would induce a sleeplike state. More recently, disturbing the sleeper, and, more important, sleep could
these observations included other compounds such as be continuously and quantitatively measured without dis-
opium. Finally, it was noted that caffeine had the power turbing the sleeper.
to prevent sleep. All the major elements of sleep brain wave patterns
The hypnotoxin theory reached its zenith in 1907 when were described by Harvey, Hobart, Davis, and others7-9
French physiologists, Legendre and Pieron,4 did experi- at Harvard University in a series of extraordinary papers
ments showing that blood serum from sleep-deprived dogs published in 1937, 1938, and 1939. Blake, Gerard, and
could induce sleep in dogs that were not sleep deprived. Kleitman10,11 added to this from their studies at the
The notion of a toxin causing the brain to sleep has gradu- University of Chicago. On the human electroencepha-
ally given way to the notion that there are a number of logram (EEG), sleep was characterized by high-amplitude
endogenous sleep factors that actively induce sleep by slow waves and spindles, whereas wakefulness was char-
specific mechanisms. acterized by low-amplitude waves and alpha rhythm. The
In the 1920s, the University of Chicago physiologist image of the sleeping brain completely turned off
Nathaniel Kleitman carried out a series of sleep-depriva- gave way to the image of the sleeping brain engaged
tion studies and made the simple but brilliant observation in slow, synchronized, idling neuronal activity. Although
that individuals who stayed up all night were generally less it was not widely recognized at the time, these studies
sleepy and impaired the next morning than in the middle were some of the most critical turning points in sleep
of their sleepless night. Kleitman argued that this observa- research. Indeed, Hobson3 dated the turning point of
tion was incompatible with the notion of a continual sleep research to 1928, when Berger began his work on
buildup of a hypnotoxin in the brain or blood. In addition, the human EEG. Used today in much the same way as
he felt that humans were about as impaired as they would they were in the 1930s, brain wave recordings with paper
get, that is, very impaired, after about 60 hours of wakeful- and ink, or more recently on computer screens, have
ness, and that longer periods of sleep deprivation would been extraordinarily important to sleep research and sleep
produce little additional change. In the 1939 (first) edition medicine.
of his comprehensive landmark monograph Sleep and The 1930s also saw one series of investigations that
Wakefulness Kleitman5 summed up by saying, It is seemed to establish conclusively both the passive theory of
perhaps not sleep that needs to be explained, but wakeful- sleep and the notion that it occurred in response to reduc-
ness, and indeed, there may be different kinds of wakeful- tion of stimulation and activity. These were the investiga-
ness at different stages of phylogenetic and ontogenetic tions of Frederick Bremer,12,13 reported in 1935 and 1936.
development. In spite of sleep being frequently designated These investigations were made possible by the aforemen-
as an instinct, or global reaction, an actively initiated tioned development of electroencephalography. Bremer
process, by excitation or inhibition of cortical or subcorti- studied brain wave patterns in two cat preparations. One,
cal structures, there is not a single fact about sleep that which Bremer called encphale isol, was made by cutting
cannot be equally well interpreted as a let down of the a section through the lower part of the medulla. The other,
waking activity. cerveau isol, was made by cutting the midbrain just behind
the origin of the oculomotor nerves. The first preparation
permitted the study of cortical electrical rhythms under the
THE ELECTRICAL ACTIVITY influence of olfactory, visual, auditory, vestibular, and
OF THE BRAIN musculocutaneous impulses; in the second preparation, the
As the 20th century got under way, Camillo Golgi and field was narrowed almost entirely to the influence of
Santiago Ramn y Cajal had demonstrated that the nervous olfactory and visual impulses.
system was not a mass of fused cells sharing a common In the first preparation, the brain continued to present
cytoplasm but rather a highly intricate network of discrete manifestations of wakeful activity alternating with phases
cells that had the key property of signaling to one another. of sleep as indicated by the EEG. In the second prepara-
Luigi Galvani had discovered that the nerve cells of animals tion, however, the EEG assumed a definite deep sleep
produce electricity, and Emil duBois-Reymond and character and remained in this condition. In addition, the
Hermann von Helmholtz found that nerve cells use their eyeballs immediately turned downward with a progressive
electrical capabilities for signaling information to one miosis. Bremer concluded that in sleep there occurs a func-
another. In 1875, the Scottish physiologist Richard Caton tional (reversible, of course) deafferentation of the cerebral
CHAPTER 1 History of Sleep Physiology and Medicine 5

cortex. The cerveau isol preparation results in a suppres-

sion of the incessant influx of nerve impulses, particularly
cutaneous and proprioceptive, which are essential for the SLEEP PATHOLOGY
maintenance of the waking state of the telencephalon. Insomnia has been described since the dawn of history and
Apparently, olfactory and visual impulses are insufficient attributed to many causes, including a recognition of the
to keep the cortex awake. It is probably misleading to assert association between emotional disturbance and sleep dis-
that physiologists assumed the brain was completely turned turbance. Scholars and historians have a duty to bestow
off, whatever this metaphor might have meant, because credit accurately. However, many discoveries lie fallow for
blood flow and, presumably, metabolism continued. want of a contextual soil in which they may be properly
However, Bremer and others certainly favored the concept understood and in which they may extend the understand-
of sleep as a reduction of activity-idling, slow, synchro- ing of more general phenomena. Important early observa-
nized, resting neuronal activity. tions were those of von Economo on sleeping sickness
and of Pavlov, who observed dogs falling asleep during
conditioned reflex experiments.
THE ASCENDING RETICULAR Two early observations about sleep research and sleep
SYSTEM medicine stand out. The first is the description in 1880
After World War II, insulated, implantable electrodes of narcolepsy by Jean Baptiste Edouard Glineau (1859-
were developed, and sleep research on animals began in 1906), who derived the name narcolepsy from the Greek
earnest. In 1949, one of the most important and influential words narkosis (a benumbing) and lepsis (to overtake).
studies dealing with sleep and wakefulness was published: He was the first to clearly describe the collection of
Moruzzi and Magouns classic paper Brain Stem Reticular components that constitute the syndrome, although the
Formation and Activation of the EEG.14 These authors term cataplexy for the emotionally induced muscle
concluded that transitions from sleep to wakefulness or weakness was subsequently coined in 1916 by Richard
from the less extreme states of relaxation and drowsiness Henneberg.
to alertness and attention are all characterized by an appar- Obstructive sleep apnea syndrome (OSAS), which may
ent breaking up of the synchronization of discharge of the be called the leading sleep disorder of the 20th century,
elements of the cerebral cortex, an alteration marked in was described in 1836, not by a clinician but by the
the EEG by the replacement of high voltage, slow waves novelist Charles Dickens. In a series of papers entitled
with low-voltage fast activity (p. 455). the Posthumous Papers of the Pickwick Club, Dickens
High-frequency electrical stimulation through elec- described Joe, a boy who was obese and always exces-
trodes implanted in the brainstem reticular formation pro- sively sleepy. Joe, a loud snorer, was called Young Dropsy,
duced EEG activation and behavioral arousal. Thus, EEG possibly as a result of having right-sided heart failure.
activation, wakefulness, and consciousness were at one end Meir Kryger18 and Peretz Lavie19,20 published scholarly
of the continuum; sleep, EEG synchronization, and lack accounts of many early references to snoring and condi-
of consciousness were at the other end. This view, as can tions that were most certainly manifestations of OSAS.
be seen, is hardly different from the statement by MacNish Professor Pierre Passouant21 provided an account of the
quoted at the beginning of this chapter.
The demonstration by Starzl and coworkers15 that sen-
sory collaterals discharge into the reticular formation sug-
gested that a mechanism was present by which sensory
stimulation could be transduced into prolonged activation
of the brain and sustained wakefulness. By attributing an
amplifying and maintaining role to the brainstem core and
the conceptual ascending reticular activating system, it was
possible to account for the fact that wakefulness outlasts,
or is occasionally maintained in the absence of, sensory
Chronic lesions in the brainstem reticular formation
produced persisting slow waves in the EEG and immobil-
ity. The usual animal for this research was the cat because
excellent stereotaxic coordinates of brain structures had
become available in this model.16 These findings appeared
to confirm and extend Bremers observations. The theory
of the reticular activating system was an anatomically
based passive theory of sleep or an active theory of wake-
Figure 1-1 Lateral view of the monkeys brain, showing the
fulness. Figure 1-1 is from the published proceedings of ascending reticular activating system in the brainstem receiving
a symposium entitled Brain Mechanisms and Consciousness, collaterals from direct afferent paths and projecting primarily
which published in 1954 and is probably the first genuine to the associational areas of the hemisphere. (Redrawn from
neuroscience bestseller.17 Horace Magoun had extended Magoun HW: The ascending reticular system and wakefulness.
In Adrian ED, Bremer F, Jasper HH [eds]. Brain mechanisms and
his studies to the monkey, and the illustration represents consciousness. A symposium organized by the Council for
the full flowering of the ascending reticular activating International Organizations of Medical Sciences, 1954. (Cour-
system theory. tesy of Charles C Thomas, Publisher, Springfield, Ill.)
6 PARTI / Section 1 Normal Sleep and Its Variations

life of Glineau and his landmark description of the nar-

colepsy syndrome.


By far the most widespread interest in sleep by health
professionals was engendered by the theories of Sigmund
Freud, specifically about dreams. Of course, the interest
was really in dreaming, with sleep as the necessary con-
comitant. Freud developed psychoanalysis, the technique
of dream interpretation, as part of his therapeutic approach
to emotional and mental problems. As the concept of the
ascending reticular activating system dominated behav-
ioral neurophysiology, so the psychoanalytic theories
about dreams dominated the psychological side of the coin.
Dreams were thought to be the guardians of sleep and to
occur in response to a disturbance in order to obviate
waking up, as exemplified in the classic alarm clock dream.
Freuds concept that dreaming discharged instinctual
energy led directly to the notion of dreaming as a safety
valve of the mind. At the time of the discovery of rapid eye
movements during sleep (circa 1952), academic psychiatry
was dominated by psychoanalysts, and medical students all
over America were interpreting one anothers dreams.
From the vantage point of todays world, the dream
deprivation studies of the early 1960s, engendered and
reified by the belief in psychoanalysis, may be regarded by Figure 1-2 Representation of de Mairans original experiment.
some as a digression from the mainstream of sleep medi- When exposed to sunlight during the day (upper left), the leaves
of the plant were open; during the night (upper right), the
cine. On the other hand, because the medicalpsychiatric leaves were folded. De Mairan showed that sunlight was not
establishment had begun to take dreams seriously, it was necessary for these leaf movements by placing the plant in
also ready to support sleep research fairly generously under total darkness. Even under these constant conditions, the
the guise of dream research. leaves opened during the day (lower left) and folded during
the night (lower right). (Redrawn from Moore-Ede MC, Sulzman
FM, Fuller CA. The clocks that time us: physiology of the cir-
cadian timing system. Cambridge, Mass: Harvard University
CHRONOBIOLOGY Press; 1982. p. 7.)
Most, but not all, sleep specialists share the opinion that
what has been called chronobiology or the study of bio-
logic rhythms is a legitimate part of sleep research and
sleep medicine. The 24-hour rhythms in the activities of nient and widely used method for demonstrating circa-
plants and animals have been recognized for centuries. dian rhythmicity.
These biologic 24-hour rhythms were quite reasonably 2. The favorite animal of sleep research from the 1930s
assumed to be a direct consequence of the periodic envi- through the 1970s was the cat, and neither cats nor dogs
ronmental fluctuation of light and darkness. However, in demonstrate clearly defined circadian activity rhythms.
1729, Jean Jacques dOrtous de Mairan described a helio- 3. The separation between chronobiology and sleep
trope plant that opened its leaves during the day even after research was further maintained by the tendency for
de Mairan had moved the plant so that sunlight could not chronobiologists to know very little about sleep, and for
reach it. The plant opened its leaves during the day and sleep researchers to remain ignorant of such biological
folded them for the entire night even though the environ- clock mysteries as phase response curves, entrainment,
ment was constant. This was the first demonstration of and internal desynchronization.
the persistence of circadian rhythms in the absence of
environmental time cues. Figure 1-2, which represents de
Mairans original experiment, is reproduced from The THE DISCOVERY OF REM SLEEP
Clocks That Time Us by Moore-Ede and colleagues.22 The characterization of rapid eye movement (REM) sleep
Chronobiology and sleep research developed separately. as a discrete organismic state should be distinguished from
The following three factors appear to have contributed to the discovery that rapid eye movements occur during
this: sleep. The historical threads of the discovery of rapid eye
1. The long-term studies commonly used in biologic movements can be identified. Nathaniel Kleitman (Fig.
rhythm research precluded continuous recording of 1-3, Video 1-1), a professor of physiology at the University
brain wave activity. Certainly, in the early days, the of Chicago, had long been interested in cycles of activity
latter was far too difficult and not really necessary. The and inactivity in infants and in the possibility that this cycle
measurement of wheel-running activity was a conve- ensured that an infant would have an opportunity to
CHAPTER 1 History of Sleep Physiology and Medicine 7

great deal of concern that these potentials were electrical

artifacts. With their presence on the EOG as a signal,
however, it was possible to watch the subjects eyes simul-
taneously, and when this was done, the distinct rapid
movement of the eyes beneath the closed lids was extremely
easy to see.
At this point, Aserinsky and Kleitman made two
1. These eye movements represented a lightening of
2. Because they were associated with irregular respira-
tion and accelerated heart rate, they might represent
The basic sleep cycle was not identified at this time,
primarily because the EOG and other physiologic mea-
sures, notably the EEG, were not recorded continuously
but rather by sampling a few minutes of each hour or half-
hour. The sampling strategy was done to conserve paper
(there was no research grant) and because there was not a
clear reason to record continuously. Also, the schedule
made it possible for the researcher to nap between sam-
pling episodes.
Aserinsky and Kleitman initiated a small series of awak-
Figure 1-3 Nathaniel Kleitman (circa 1938), Professor of Physi- enings, both when rapid eye movements were present and
ology, University of Chicago, School of Medicine. when rapid eye movements were not present, for the
purpose of eliciting dream recall. They did not apply
sophisticated methods of dream content analysis, but the
respond to hunger. He postulated that the times infants descriptions of dream content from the two conditions
awakened to nurse on a self-demand schedule would be were generally quite different with REM awakenings
integral multiples of a basic rest-activity cycle. The second yielding vivid complex stories and non-REM (NREM)
thread was Kleitmans interest in eye motility as a possible awakenings often yielding nothing at all or very sparse
measure of depth of sleep. The reasoning for this was accounts. This made it possible to conclude that rapid eye
that eye movements had a much greater cortical represen- movements were associated with dreaming. This was,
tation than did almost any other observable motor activity, indeed, a breakthrough in sleep research.24,25
and that slow, rolling, or pendular eye movements had The occurrence of the eye movements was quite com-
been described at the onset of sleep with a gradual slowing patible with the contemporary dream theories that dreams
and disappearance as sleep deepened.23 occurred when sleep lightened in order to prevent or delay
In 1951, Kleitman assigned the task of observing eye awakening. In other words, dreaming could still be regarded
movement to a graduate student in physiology named as the guardian of sleep. However, it could no longer be
Eugene Aserinsky. Watching the closed eyes of sleeping assumed that dreams were fleeting and evanescent.
infants was tedious, and Aserinsky soon found that it was
easier to designate successive 5-minute epochs as periods
of motility if he observed any movement at all, usually a ALL-NIGHT SLEEP RECORDINGS
writhing or twitching of the eyelids, versus periods of no AND THE BASIC SLEEP CYCLE
motility. The seminal Aserinsky and Kleitman paper was published
After describing an apparent rhythm in eye motility, in 1953. It attracted little attention, and no publications on
Kleitman and Aserinsky decided to look for a similar phe- the subject appeared from any other laboratory until 1959.
nomenon in adults. Again, watching the eyes during the Staying up at night to study sleep remained an undesirable
day was tedious, and at night it was even worse. Casting occupation by anyones standards. In the early 1950s,
about, they came upon the method of electrooculography most previous research on the EEG patterns of sleep, like
and decided (correctly) that this would be a good way to most approaches to sleep physiology generally, had either
measure eye motility continuously and would relieve the equated short periods of sleep with all sleep or relied on
human observer of the tedium of direct observations. intermittent time sampling during the night. The notion
Sometimes in the course of recording electrooculograms of obtaining continuous records throughout typical nights
(EOGs) during sleep, they saw bursts of electrical potential of sleep would have seemed highly extravagant.
changes that were quite different from the slow move- However, motivated by the desire to expand and quan-
ments at sleep onset. tify the description of rapid eye movements, then graduate
When they were observing infants, Aserinsky and student William Dement and Kleitman26 did just this over
Kleitman had not differentiated between slow and rapid a total of 126 nights with 33 subjects and, by means of a
movements. However, on the EOG, the difference between simplified categorization of EEG patterns, scored the
the slow eye movements at sleep onset and the newly dis- paper recordings in their entirety. When they examined
covered rapid motility was obvious. Initially, there was a these 126 records, they found that there was a predictable
8 PARTI / Section 1 Normal Sleep and Its Variations

sequence of patterns over the course of the night, such as EEG pattern during a precise interval of sleep always
had been hinted at by Aserinskys study but entirely over- associated with bursts of REM, which were additionally
looked in all previous EEG studies of sleep. Although this established as periods of vivid dreaming.
sequence of regular variations has now been observed tens Observations of motor activity in both humans and
of thousands of times in hundreds of laboratories, the animals revealed the unique occurrence of an active sup-
original description remains essentially unchanged. pression of spinal motor activity and muscle reflexes.
The usual sequence was that after the onset of sleep, Thus, sleep consists not of one state but rather of two
the EEG progressed fairly rapidly to stage 4, which per- distinct organismic states, as different from one another as
sisted for varying amounts of time, generally about 30 both are from wakefulness. It had to be conceded that sleep
minutes, and then a lightening took place. Whereas the could no longer be thought of as a time of brain inactivity
progression from wakefulness to stage 4 at the beginning and EEG slowing. By 1960, this fundamental change in
of the cycle was almost invariable through a continuum of our thinking about the nature of sleep was well established;
change, the lightening was usually abrupt and coincident it exists as fact that has not changed in any way since that
with a body movement or series of body movements. After time.
the termination of stage 4, there was generally a short The discovery of rapid eye movements during sleep in
period of stage 2 or stage 3 which gave way to stage 1 and humans, plus the all-night sleep recordings that revealed
rapid eye movements. When the first eye movement the regular recurrence of lengthy periods during which
period ended, the EEG again progressed through a con- rapid eye movements occurred and during which brain
tinuum of change to stage 3 or 4, which persisted for a wave patterns resembled light sleep, prepared the way for
time and then lightened, often abruptly, with body move- the discovery of REM sleep in cats, in spite of the extremely
ment to stage 2, which again gave way to stage 1 and the powerful bias that an activated EEG could not be associ-
second rapid eye movement period (see p. 679 of Dement ated with sleep. In the first study of cats, maintaining the
and Kleitman26). insulation and, therefore, the integrity of implanted elec-
Dement and Kleitman found that this cyclical variation trodes had not yet been solved, so an alternative, small pins
of EEG patterns occurred repeatedly throughout the night in the scalp, was used. With this approach, the waking
at intervals of 90 to 100 minutes from the end of one eye EEG was totally obscured by the electromyogram from
movement period to the end of the next. The regular the large temporal muscles of the cat. However, when the
occurrences of REM periods and dreaming strongly sug- cat fell asleep, slow waves could be seen, and the transition
gested that dreams did not occur in response to chance to REM sleep was clearly observed because muscle poten-
disturbances. tials were completely suppressed. The cats rapid eye
At the time of these observations, sleep was still consid- movements and also the twitching of the whiskers and
ered to be a single state. Dement and Kleitman character- paws could be directly observed.
ized the EEG during REM periods as emergent stage 1 It is very difficult today, in 2010, to understand and
as opposed to descending stage 1 at the onset of sleep. appreciate the exceedingly controversial nature of these
The percentage of the total sleep time occupied by REM findings. The following note from a more personal
sleep was between 20% and 25%, and the periods of REM account27 illustrates both the power and the danger of
sleep tended to be shorter in the early cycles of the night. scientific dogma. I wrote them [the findings] up, but the
This pattern of all-night sleep has been seen over and over paper was nearly impossible to publish because it was com-
in normal humans of both sexes, in widely varying environ- pletely contradictory to the totally dominant neurophysi-
ments and cultures, and across the life span. ological theory of the time. The assertion by me that an
activated EEG could be associated with unambiguous
sleep was considered to be absurd. As it turned out, previ-
REM SLEEP IN ANIMALS ous investigators had observed an activated EEG during
The developing knowledge of the nature of sleep with sleep in cats28,29 but simply could not believe it and ascribed
rapid eye movements was in direct opposition to the it to arousing influences during sleep. A colleague who was
ascending reticular activating system theory and consti- assisting me was sufficiently skeptical that he preferred I
tuted a paradigmatic crisis. The following observations publish the paper as sole author. After four or five rejec-
were crucial: tions, to my everlasting gratitude, Editor-in-Chief Herbert
Arousal thresholds in humans were much higher during Jasper accepted the paper without revision for publication
periods of REM sleep that had a low-amplitude, rela- in Electroencephalography and Clinical Neurophysiol-
tively fast (stage 1) EEG pattern than during similar ogy. (see p. 23 of Dement30)
light sleep periods at the onset of sleep. It is notable, however, that I did not appreciate the
Rapid eye movements during sleep were discovered in significance of the absence of muscle potentials during
cats; the concomitant brain wave patterns (low- the REM periods in cats. It remained for Michel Jouvet,
amplitude, fast) were indistinguishable from active working in Lyon, France, to insist on the importance
wakefulness. of electromyographic suppression in his early papers,31,32
By discarding the sampling approach and recording con- the first of which was published in 1959. Hodes and
tinuously, a basic 90-minute cycle of sleep without rapid Dement began to study the H reflex in humans in
eye movements, alternating with sleep with rapid eye 1960, finding complete suppression of reflexes during
movements, was discovered. This basic sleep-cycle char- REM sleep,33 and Octavio Pompeiano and others in Pisa,
acterized all episodes of nocturnal sleep. Continuous Italy, worked out the basic mechanisms of REM atonia
recording also revealed a consistent, low-amplitude in the cat.34
CHAPTER 1 History of Sleep Physiology and Medicine 9

Fishers suggestion, a sleep recording was begun. Within

THE DUALITY OF SLEEP seconds after he fell asleep, the patient was showing the
Even though the basic NREM sleep cycle was well estab- dramatic, characteristic rapid eye movements of REM
lished, the realization that REM sleep was qualitatively sleep, and sawtooth waves as well, in the EEG. The first
different from the remainder of sleep took years to evolve. paper documenting sleep-onset REM periods in a single
Jouvet35 and his colleagues performed an elegant series of patient was published in 1960 by Gerald Vogel.39 In a col-
investigations on the brainstem mechanisms of sleep that laborative study between the University of Chicago and
forced the inescapable conclusion that sleep consists of the Mount Sinai Hospital, nine patients were studied, and
two fundamentally different organismic states. Among his the important sleep-onset REM periods at night were
many early contributions were clarification of the role of described in a 1963 paper.40 Subsequent research showed
pontine brainstem systems as the primary anatomic site for that sleepy patients who did not have cataplexy did not
REM sleep mechanisms and the clear demonstration that have sleep-onset REM periods (SOREMPs), and those
electromyographic activity and muscle tonus are com- with cataplexy always had SOREMPs.41 It was clear that
pletely suppressed during REM periods and only during the best explanation for cataplexy was the normal motor
REM periods. These investigations began in 1958 and inhibitory mechanisms of REM sleep occurring during
were carried out during 1959 and 1960. wakefulness in a precocious or abnormal way.
It is now an established fact that atonia is a fundamental
characteristic of REM sleep and that it is mediated by an
active and highly specialized neuronal system. The pio- THE NARCOLEPSY CLINIC:
neering microelectrode studies of Edward Evarts36 in cats A FALSE START
and monkeys, and observations on cerebral blood flow in In January 1963, after moving to Stanford University,
the cat by Reivich and Kety37 provided convincing evi- Dement was eager to test the hypothesis of an association
dence that the brain during REM sleep is very active. between cataplexy and SOREMPs. However, not a single
Certain areas of the brain appear to be more active in REM narcoleptic patient could be identified. A final attempt was
sleep than in wakefulness. By now, the notion of sleep as made by placing a want ad, a few words about an inch
a passive process was totally demolished although for many high, in a daily newspaper, the San Francisco Chronicle.
years there was a lingering attitude that NREM sleep was More than 100 people responded! About 50 of these
essentially inactive and quiet. By 1960, it was possible to patients were bona fide narcoleptics having both sleepiness
define REM sleep as a completely separate organismic and cataplexy.
state characterized by cerebral activation, active motor The response to the ad was a noteworthy event in the
inhibition, and, of course, an association with dreaming. development of sleep disorders medicine. With one or
The fundamental duality of sleep was an established fact. two exceptions, none of the narcoleptics had ever been
diagnosed correctly. A responsibility for their clinical
management had to be assumed in order to facilitate
PREMONITIONS OF SLEEP MEDICINE their participation in the research. The late Dr. Stephen
Sleep research, which emphasized all-night sleep record- Mitchell, who had completed his neurology training and
ings, burgeoned in the 1960s and was the legitimate was entering a psychiatry residency at Stanford University,
precursor of sleep medicine and particularly of its core joined Dement in creating a narcolepsy clinic in 1964,
clinical test, polysomnography. Much of the research at and they were soon managing well over 100 patients.
this time emphasized studies of dreaming and REM sleep Mostly, this involved seeing the patients at regular inter-
and had its roots in a psychoanalytic approach to mental vals and adjusting their medication. Nonetheless, the seeds
illness which strongly implicated dreaming in the psy- of the typical sleep disorders clinic were sowed because
chotic process. After sufficient numbers of all-night sleep at least one daytime polygraphic sleep recording was
recordings had been carried out in humans to demon- performed on all patients to establish the presence of
strate a highly characteristic normal sleep architecture, SOREMPs, and patients were questioned exhaustively
investigators noted a significantly shortened REM latency about their sleep. If possible, an all-night sleep recording
in association with endogenous depression.38 This phe- was carried out. Unfortunately, most of the patients were
nomenon has been intensively investigated ever since. unable to pay cash to cover their bills, and insurance com-
Other important precursors of sleep medicine were the panies declared that the recordings of narcoleptic patients
following: were experimental. Because the clinic was unable to gener-
1. Discovery of sleep-onset REM periods in patients with ate sufficient income, it was discontinued and most of
narcolepsy the patients were referred back to local physicians with
2. Interest in sleep, epilepsy, and abnormal movement instructions about treatment.
primarily in France
3. Introduction of benzodiazepines and the use of sleep
laboratory studies in defining hypnotic efficacy EUROPEAN INTEREST
In Europe, a genuine clinical interest in sleep problems
had arisen, and it achieved its clearest expression in a 1963
SLEEP-ONSET REM PERIODS symposium held in Paris, organized by Professor H.
AND CATAPLEXY Fischgold, and published as La Sommeil de Nuit Normal
In 1959, a patient with narcolepsy came to the Mount Sinai et Pathologique in 1965.42 The primary clinical emphasis
Hospital in New York City to see Dr. Charles Fisher. At in this symposium was the documentation of sleep-related
10 PARTI / Section 1 Normal Sleep and Its Variations

epileptic seizures and a number of related studies on sleep- in the French discovery of sleep apnea, C. Alberto Tassi-
walking and night terrors. Investigators from France, Italy, nari, joined the Italian neurologist Elio Lugaresi in
Belgium, Germany, and the Netherlands took part. Bologna in 1970. These clinical investigators along with
Giorgio Coccagna and a host of others over the years
performed a crucial series of clinical sleep investigations
BENZODIAZEPINES AND HYPNOTIC and, indeed, provided a complete description of the sleep
EFFICACY STUDIES apnea syndrome, including the first observations of the
Benzodiazepines were introduced in 1960 with the market- occurrence of sleep apnea in nonobese patients, an account
ing of chlordiazepoxide (Librium). This compound offered of the cardiovascular correlates, and a clear identification
a significant advance in terms of safety over barbiturates of the importance of snoring and hypersomnolence as
for the purpose of tranquilizing and sedating. It was quickly diagnostic indicators. These studies are recounted in
followed by diazepam (Valium) and the first benzodiaze- Lugaresis book, Hypersomnia with Periodic Apneas, pub-
pine introduced specifically as a hypnotic, flurazepam lished in 1978.52
(Dalmane). Although a number of studies had been done
on the effects of drugs on sleep, usually to answer theoreti-
cal questions, the first use of the sleep laboratory to evalu- ITALIAN SYMPOSIA
ate sleeping pills was the 1965 study by Oswald and Priest.43 In 1967, Henri Gastaut and Elio Lugaresi (Fig. 1-4) orga-
An important series of studies establishing the role of the nized a symposium, published as The Abnormalities of Sleep
sleep laboratory in the evaluation of hypnotic efficacy in Man,53 which encompassed issues across a full range of
was carried out by Anthony Kales and his colleagues at pathologic sleep in humans. This meeting took place in
the University of California, Los Angeles.44 The group also Bologna, Italy, and the papers presented covered many of
carried out studies of patients with hypothyroidism, what are now major topics in the sleep medicine field:
asthma, Parkinsons disease, and somnambulism.45-48 insomnia, sleep apnea, narcolepsy, and periodic leg move-
ments during sleep. It was an epic meeting from the point
of view of the clinical investigation of sleep; the only major
THE DISCOVERY OF SLEEP APNEA issues not represented were clear concepts of clinical prac-
One of the most important events in the history of sleep tice models and clear visions of the high population preva-
disorders medicine occurred in Europe. Sleep apnea lence of sleep disorders. However, the event that may have
was discovered independently by Gastaut, Tassinari, and finally triggered a serious international interest in sleep
Duron49 in France and by Jung and Kuhlo in Germany.50 apnea syndromes was organized by Lugaresi in 1972 and
Both these groups reported their findings in 1965. As took place in Rimini, a small resort on the Adriatic coast.54
noted earlier, scholars have found references to this pheno
menon in many places, but this was the first clear-cut
recognition and description that had a direct causal conti- BIRTH PANGS
nuity to sleep disorders medicine as we know it today. In spite of all the clinical research, the concept of all-night
Peretz Lavie has detailed the historical contributions made sleep recordings as a clinical diagnostic test did not emerge
by scientists and clinicians around the world in helping to unambiguously. It is worth considering the reasons for this
describe and understand this disorder.20 failure, partly because they continue to operate today as
These important findings were widely ignored in impediments to the expansion of the field, and partly to
America (Video 1-2). What should have been an almost understand the fields long overdue development.
inevitable discovery by either the otolaryngologic surgery The first important reason was the unprecedented
community or the pulmonary medicine community did nature of an all-night diagnostic test, particularly if it
not occur because there was no tradition in either specialty was conducted on outpatients. The cost of all-night
for carefully observing breathing during sleep. The well-
known and frequently cited study of Burwell and col-
leagues51although impressive in a literary sense in its
evoking of the somnolent boy, Joe, from the Posthumous
Papers of the Pickwick Cluberred badly in evaluating
their somnolent obese patients only during waking and
attributing the cause of the somnolence to hypercapnia.
The popularity of this paper further reduced the likeli-
hood of discovery of sleep apnea by the pulmonary com-
munity. To this day, there is no evidence that hypercapnia
causes true somnolence, although, of course, high levels of
PCO2 are associated with impaired cerebral function.
Nonetheless, the term pickwickian became an instant
success as a neologism and may have played a role in
stimulating interest in this syndrome by the European
neurologists who were also interested in sleep.
A small group of French neurologists who specialized in
clinical neurophysiology and electroencephalography were Figure 1-4 Elio Lugaresi, Professor of Neurology, University of
in the vanguard of sleep research. One of the collaborators Bologna, at the 1972 Rimini symposium.
CHAPTER 1 History of Sleep Physiology and Medicine 11

polygraphic recording, in terms of its basic expense, was had extensive knowledge of the European studies of sleep
high enough without adding the cost of hospitalization apnea. Until his arrival, the Stanford group had not rou-
although hospitalization would have legitimized a patients tinely used respiratory and cardiac sensors in their all-
spending the night in a testing facility. To sleep in an night sleep studies. Starting in 1972, these measurements
outpatient clinic for a diagnostic test was a totally unprec- became a routine part of the all-night diagnostic test. This
edented, time-intensive, and labor-intensive enterprise test was given the permanent name of polysomnography
and completely in conflict with the brief time required to in 1974 by Dr. Jerome Holland, a member of the Stanford
go to the chemistry laboratory to give a blood sample, to group. Publicity about narcolepsy and excessive sleepiness
breathe into a pulmonary function apparatus, to undergo resulted in a small flow of referrals to the Stanford sleep
a radiographic examination, and so forth. clinic, usually with the presumptive diagnosis of narco-
A second important barrier was the reluctance of non- lepsy. During the first year or two, the goal for the
hospital clinical professionals to work at night. Although Stanford practice was to see at least five new patients
medical house staff are very familiar with night work, they per week. To foster financial viability, the group did as
do not generally enjoy it; furthermore, clinicians could not much as possible (within ethical limits) to publicize its
work 24-hour days, first seeing patients and ordering tests, services. As a result, there was also a sprinkling of patients,
and then conducting the tests themselves. often self-referred, with chronic insomnia. The diagnosis
Finally, only a very small number of people understood of obstructive sleep apnea in patients with profound
that complaints of daytime sleepiness and nocturnal sleep excessive daytime sleepiness was nearly always completely
disturbance represented something of clinical significance. unambiguous.
Even narcolepsy, which was by the early 1970s fully char- During 1972, the search for sleep abnormality in patients
acterized as an interesting and disabling clinical syndrome with sleep-related complaints continued; an attempt was
requiring sleep recordings for diagnosis, was not recog- made as well to conceptualize the pathophysiologic process
nized in the larger medical community and had too low a both as an entity and as the cause of the presenting
prevalence to warrant creating a medical subspecialty. A symptom. With this approach, a number of phenomena
study carried out in 1972 documented a mean of 15 years seen during sleep were rapidly linked to the fundamental
from onset of the characteristic symptoms of excessive sleep-related presenting complaints.
daytime sleepiness and cataplexy to diagnosis and treat- Toward the end of 1972, the basic concepts and formats
ment by a clinician. The study also showed that a mean of of sleep disorders medicine were sculpted to the extent that
5.5 different physicians were consulted without benefit it was possible to offer a daylong course through Stanford
throughout that long interval.55 Universitys Division of Postgraduate Medicine. The
course was titled The Diagnosis and Treatment of Sleep
THE EARLY DEVELOPMENT OF Disorders. The topics covered were normal sleep archi-
tecture; the diagnosis and treatment of insomnia with
SLEEP MEDICINE CLINICAL drug-dependent insomnia, pseudoinsomnia, central sleep
PRACTICE apnea, and periodic leg movement as diagnostic entities;
The practice of sleep medicine developed in many centers and the diagnosis and treatment of excessive daytime sleep-
in the 1970s, often as a function of the original research iness or hypersomnia, with narcolepsy, NREM narcolepsy,
interests of the center. The sleep disorders clinic at and obstructive sleep apnea as diagnostic entities.
Stanford University is in many ways a microcosm of how The disability and cardiovascular complications of severe
sleep medicine evolved throughout the world. Patients sleep apnea were severe and alarming. Unfortunately, the
complaining of insomnia were enrolled in hypnotic effi- treatment options at this time were limited to usually inef-
cacy research studies. This brought the Stanford group fective attempts to lose weight and chronic tracheostomy.
into contact with many insomnia patients and demolished The dramatic results of chronic tracheostomy in amelio-
the notion that the majority of such patients had psychiat- rating the symptoms and complications of obstructive
ric problems. An early question was how reliable the sleep apnea had been reported by Lugaresi and cowork-
descriptions of their sleep by these patients were. The ers56 in 1970. However, the notion of using such a treat-
classic all-night sleep recording gave an answer and yielded ment was strongly resisted at the time by the medical
a great deal of information. Throughout the second half community, both in the Stanford University medical com-
of the 1960s, as a part of their research, the Stanford group munity and elsewhere. One of the first patients referred to
continued to manage patients with narcolepsy. As the the Stanford sleep clinic for investigation of this severe
groups reputation for expertise in narcolepsy grew, it somnolence and who eventually had a tracheostomy was a
began to receive referrals for evaluation from physicians 10-year-old boy. The challenges that were met to secure
all over the United States. Although sleep apnea had not the proper management of this patient can be seen in this
yet been identified (or treated) as a frequent cause of severe account by Christian Guilleminault (personal communica-
daytime sleepiness, it was clear that a number of patients tion, 1990).
referred with the presumptive diagnosis of narcolepsy cer- In addition to medical skepticism, a major obstacle to
tainly did not possess narcolepsys two cardinal signs, the practice of sleep disorders medicine was the retroactive
SOREMPs and cataplexy, and actually suffered from denial of payment by insurance companies, including the
obstructive sleep apnea. True pickwickians were an infre- largest one in the United States. A 3-year period of edu-
quent referral at this time. cational efforts directed toward third-party carriers finally
In January 1972, Christian Guilleminault, a French neu- culminated in the recognition of polysomnography as a
rologist and psychiatrist, joined the Stanford group. He reimbursable diagnostic test in 1974. Another issue was
12 PARTI / Section 1 Normal Sleep and Its Variations

that outpatient clinics that offered overnight testing in The choices of a 20-minute duration of a single test and a
polysomnographic testing bedrooms needed to obtain 2-hour interval between tests were essentially arbitrary and
state licensure in order to avoid the licensing requirements dictated by the practical demands of that study. This test
of hospitals. This, too, was finally accomplished in 1974. was then formally applied to the clinical evaluation of
sleepiness in patients with narcolepsy61 and, later, in
patients with OSAS.62
CLINICAL SIGNIFICANCE OF Carskadon and her colleagues then undertook a monu-
EXCESSIVE DAYTIME SLEEPINESS mental study of sleepiness in children by following them
Christian Guilleminault, in a series of studies, had clearly longitudinally across the second decade of life, which
shown that excessive daytime sleepiness was a major pre- happens to also be the decade of highest risk for the devel-
senting complaint of several sleep disorders as well as a opment of narcolepsy. Using the new MSLT measure, she
pathologic phenomenon unto itself.57 However, it was rec- found that 10-year-old children were completely alert in
ognized that methods to quantify this symptom and the the daytime, but by the time they reached sexual maturity,
underlying condition were not adequate to document the they were no longer fully alert even though they obtained
degree of improvement as a result of treatment. The sub- almost the same amount of sleep at night as at age ten.
jective Stanford Sleepiness Scale, developed by Hoddes Results of this remarkable decade of work and other studies
and colleagues,58 did not give reliable results. The problem are summarized in an important review.63
was not a crisis, however, because patients with severe Early MSLT research established the following impor-
apnea and overwhelming daytime sleepiness improved dra- tant advances in thinking:
matically after tracheostomy, and the reduction in daytime 1. Daytime sleepiness and nighttime sleep are an inter
sleepiness was unambiguous. Nonetheless, the less urgent active continuum, and the adequacy of nighttime sleep
need to document the pharmacologic treatment of narco- absolutely cannot be understood without a complemen-
lepsy and the objective improvement of sleepiness in tary measurement of the level of daytime sleepiness or
patients with less severe sleep apnea continued to be a its antonym, alertness.
problem. 2. Excessive sleepiness, also known as impaired alertness,
The apparent lack of interest in daytime sleepiness by was sleep medicines most important symptom.
individuals who were devoting their careers to the investi-
gation of sleep at that time has always been puzzling.
There is no question but that the current active investiga-
tion of this phenomenon is the result of the early interest SLEEP MEDICINE
of sleep disorders specialists. The early neglect of sleepi- As the decade of the 1970s drew to a close, the consolida-
ness is all the more difficult to understand because it is now tion and formalization of the practice of sleep disorders
widely recognized that sleepiness and the tendency to fall medicine was largely completed. What is now the
asleep during the performance of hazardous tasks is one of American Academy of Sleep Medicine was formed and
the most important problems in our society. A number of provided a home for professionals interested in sleep and,
reasons have been put forward. One is that sleepiness and particularly, in the diagnosis and treatment of sleep disor-
drowsiness are negative qualities. A second is that the soci- ders. This organization began as the Association of Sleep
etal failure to confront the issue was fostered by language Disorders Centers with five members in 1975. The orga-
ambiguities in identifying sleepiness. A third is that the nization then was responsible for the initiation of the
early sleep laboratory studies focused almost exclusively on scientific journal Sleep, and it fostered the setting of stan-
REM sleep and other nighttime procedures with little dards through center accreditation and an examination for
concern for the daytime except for psychopathology. practitioners by which they were designated Accredited
Finally, the focus with regard to sleep deprivation was on Clinical Polysomnographers.
performance from the perspective of human factors rather The first international symposium on narcolepsy took
than on sleepiness as representing a homeostatic response place in the French Languedoc in the summer of 1975,
to sleep reduction. immediately after the Second International Congress of
An early attempt to develop an objective measure of the Association for the Physiological Study of Sleep in
sleepiness was that of Yoss and coworkers59 who observed Edinburgh. The former meeting, in addition to being sci-
pupil diameter directly by video monitoring and described entifically productive, had landmark significance because
changes in sleep deprivation and narcolepsy. Subsequently it produced the first consensus definition of a specific sleep
designated pupillometry, this technique has not been disorder,64 drafted, revised, and unanimously endorsed
widely accepted. Dr. Mary Carskadon deserves most of the by 65 narcoleptologists of international reputation. The
credit for the development of the latter-day standard first sleep disorders patient volunteer organization, the
approach to the measurement of sleepiness called the American Narcolepsy Association, was also formed in
Multiple Sleep Latency Test (MSLT). She noted that sub- 1975. The ASDC/APSS Diagnostic Classification of Sleep
jective ratings of sleepiness made before a sleep recording and Arousal Disorders was published in fall 1979 after 3
not infrequently predicted the sleep latency. In the spring years of extraordinary effort by a small group of dedicated
of 1976, she undertook to establish sleep latency as an individuals who composed the nosology committee
objective measurement of the state of sleepiness-alertness chaired by Howard Roffwarg.65
by measuring sleep tendency before, during, and after 2 Before the 1980s, the only effective treatment for severe
days of total sleep deprivation.60 The protocol designed for OSAS was chronic tracheostomy. This highly effective but
this study has become the standard protocol for the MSLT. personally undesirable approach was replaced by two new
CHAPTER 1 History of Sleep Physiology and Medicine 13

proceduresone surgical,66 the other mechanical.67 The person today knows a great deal more about sleep and its
first was uvulopalatopharyngoplasty, which is giving way disorders than the average person did at the end of the
to more complex and effective approaches. The second 1980s.
was the widely used and highly effective continuous posi-
tive nasal airway pressure technique introduced by the
Australian pulmonologist Colin Sullivan (Video 1-3). The THE TURN OF THE CENTURY
combination of the high prevalence of OSAS and effective AND BEYOND
treatments fueled a strong expansion of centers and indi- Chapter 62 of this volume deals with public policy and
viduals offering the diagnosis and treatment of sleep dis- public health issues. From todays vantage, the greatest
orders to patients. challenge for the future is the cost-effective expansion of
The decade of the 1980s was capped by the publication sleep medicine so that its benefits will be readily available.
of sleep medicines first textbook, Principles and Practice of The major barrier to this availability is the continuing
Sleep Medicine.68 For many years there was only one medical failure of sleep research and sleep medicine to effectively
journal devoted to sleep; by 2004 there were seven: Sleep, penetrate the educational system at any level. As a conse-
Journal of Sleep Research, Sleep and Biological Rhythms, Sleep quence, the majority of individuals remain unaware of
& Breathing, Sleep Medicine, Sleep Medicine Reviews, and important facts of sleep and wakefulness, biologic rhythms,
Sleep Research Online. Articles about sleep are now rou- and sleep disorders, and particularly of the symptoms that
tinely published in the major pulmonary, neurology, and suggest a serious pathologic process. The management of
psychiatric journals. sleep deprivation and its serious consequences in the work-
The 1990s saw an acceleration in the acceptance of sleep place, particularly in those industries that depend on sus-
medicine throughout the world.69 In spite of that, adequate tained operations, continue to need increased attention.
sleep medicine services are still not readily available every- Finally, the education and training of all health profes-
where.70 In the United States, the National Center on sionals, including nurses, has far to go. This situation was
Sleep Disorders Research (NCSDR) was established by highlighted by the recently published report of the Insti-
statute as part of the National Heart, Lung, and Blood tute of Medicine.74 Take heart! These problems are grand
Institute of the National Institutes of Health.71,72 The opportunities. Sleep medicine has come into its own
mandate of NCSDR is to support research, promote edu- (Videos 1-4 and 1-5). It has made concern for health a truly
cational activities, and coordinate sleep-related activities 24-hours-a-day enterprise, and it has energized a new
throughout various branches of the U.S. government. This effort to reveal the secrets of the healthy and unhealthy
initiative led to the development of large research projects sleeping brain.
dealing with various aspects of sleep disorders and the
establishment of awards to develop educational materials Clinical Pearl
at all levels of training.
The 1990s also saw the establishment of the National Recent advances in sleep science, sleep medicine,
Sleep Foundation73 as well as other organizations for public policy, and communications will foster an edu-
patients. This foundation points out to the public the cated public that will know a great deal about sleep
dangers of sleepiness, and sponsors the annual National and its disorders. Clinicians should expect that their
patients may have already learned about their sleep
Sleep Awareness Week. As the Internet increases exponen-
disorders from the information sources that are
tially in size, so does the availability of sleep knowledge for readily available.
physicians, patients, and the general public. The average

Case History
Raymond M. was a 10 12 -year-old boy referred to the was not paying attention and was hyperactive and
pediatrics clinic in 1971 for evaluation of unexplained aggressive. His mother confirmed that he had been a
hypertension, which had developed progressively over very loud snorer since he was very young, at least since
the preceding 6 months. There was a positive family age 2, perhaps before.
history of high blood pressure, but never so early in Physical examination revealed an obese boy with a
life. Raymond was hospitalized and had determination short neck and a very narrow airway. I recommended
of renin, angiotensin, and aldosterone, renal function a sleep evaluation, which was accepted. An esophageal
studies including contrast radiographs, and extensive balloon and measurement of end-tidal CO2 was added
cardiac evaluation. All results had been normal except to the usual array. His esophageal pressure reached
that his blood pressure oscillated between 140-170/90- 80 to 120 cm H2O, he had values of 6% end-tidal
100. It was noticed that he was somnolent during CO2 at end of apnea, apneic events lasted between 25
the daytime and Dr. S. suggested that I see him for and 65 seconds, and the apnea index was 55. His
this unrelated symptom. SaO2 [oxygen saturation, arterial] was frequently below
I reviewed Raymonds history with his mother. 60%.
Raymond had been abnormally sleepy all his life. I called the pediatric resident and informed him that
However, during the past 2 to 3 years, his schoolteach- the sleep problem was serious. I also suggested that
ers were complaining that he would fall asleep in class the sleep problem might be the cause of the unex-
and was at times a behavioral problem because he plained hypertension. The resident could not make
14 PARTI / Section 1 Normal Sleep and Its Variations

sense of my information and passed it to the attending pletely unacceptable in a child. However, they did
physician. I was finally asked to present my findings at concede that if no improvement was achieved by
the pediatric case conference, which was led by Dr. S. medical management, Raymond would be reinvesti-
I came with the recordings, showed the results, and gated, including sleep studies.
explained why I believed that there was a relationship That was spring 1972. In the fall, he was, if any-
between the hypertension and the sleep problem. There thing, worse in spite of vigorous medical treatment. At
were a lot of questions. They simply could not believe the end of 1972, Raymond finally had his tracheos-
it. I was asked what treatment I would recommend, and tomy. His blood pressure went down to 90/60 within
I suggested a tracheostomy. I was asked how many 10 days, and he was no longer sleepy. During the 5
patients had this treatment in the United States, and years we were able to follow Raymond, he remained
how many children had ever been treated with trache- normotensive and alert, but I had to fight continuously
ostomy. When I had to answer zero to both questions, to prevent outside doctors from closing his tracheos-
the audience was somewhat shocked. It was decided tomy. I do not know what has happened to him since
that such an approach was doubtful at best, and com- then.

24. Aserinsky E, Kleitman N. Regularly occurring periods of eye motil-

REFERENCES ity, and concomitant phenomena, during sleep. Science 1953;118:
1. Thorpy M. History of sleep and man. In: Thorpy M, Yager J, editors. 273-274.
The encyclopedia of sleep and sleep disorders. New York: Facts on 25. Aserinsky E, Kleitman N. Two types of ocular motility occurring in
File; 1991. sleep. J Appl Physiol 1955;8:11-18.
2. MacNish R. The philosophy of sleep. New York: D Appleton; 1834. 26. Dement W, Kleitman N. Cyclic variations in EEG during sleep and
3. Hobson J. Sleep. New York: Scientific American Library; 1989. their relation to eye movements, body motility, and dreaming. Elec-
4. Legendre R, Pieron H. Le probleme des facteurs du sommeil: resul- troencephalogr Clin Neurophysiol 1957;9:673-690.
tats dinjections vasculaires et intracerebrales de liquides insom- 27. Dement W. A personal history of sleep disorders medicine. J Clin
niques. C R Soc Biol 1910;68:1077-1079. Neurophysiol 1990;1:17-47.
5. Kleitman N. Sleep and wakefulness. Chicago: University of Chicago 28. Derbyshire AJ, Rempel B, Forbes A, et al. The effects of anesthetics
Press; 1939. on action potentials in the cerebral cortex of the cat. Am J Physiol
6. Berger H. Ueber das Elektroenkephalogramm des Menschen. 1936;116:577-596.
J Psychol Neurol 1930;40:160-179. 29. Hess R, Koella WP, Akert K. Cortical and subcortical recordings in
7. Davis H, Davis PA, Loomis AL, et al. Changes in human brain natural and artificially induced sleep in cats. Electroencephalogr Clin
potentials during the onset of sleep. Science 1937;86:448-450. Neurophysiol 1953;5:75-90.
8. Davis H, Davis PA, Loomis AL, et al. Human brain potentials during 30. Dement W. The occurrence of low voltage, fast electroencephalo-
the onset of sleep. J Neurophysiol 1938;1:24-38. gram patterns during behavioral sleep in the cat. Electroencephalogr
9. Harvey EN, Loomis AL, Hobart GA. Cerebral states during sleep Clin Neurophysiol 1958;10:291-296.
as studied by human brain potentials. Science 1937;85:443-444. 31. Jouvet M, Michel F, Courjon J. Sur un stade dactivite electrique
10. Blake H, Gerard RW. Brain potentials during sleep. Am J Physiol cerebrale rapide au cours du sommeil physiologique. C R Soc Biol
1937;119:692-703. 1959;153:1024-1028.
11. Blake H, Gerard RW, Kleitman N. Factors influencing brain poten- 32. Jouvet M, Mounier D. Effects des lesions de la formation reticulaire
tials during sleep. J Neurophysiol 1939;2:48-60. pontique sur le sommeil du chat. C R Soc Biol 1960;154:2301-2305.
12. Bremer F. Cerveau isol et physiologie du sommeil. C R Soc Biol 33. Hodes R, Dement W. Depression of electrically induced reflexes
1935;118:1235-1241. (H-reflexes) in man during low voltage EEG sleep. Electroen-
13. Bremer F. Cerveau. Nouvelles recherches sur le mecanisme du cephalogr Clin Neurophysiol 1964;17:617-629.
sommeil. C R Soc Biol 1936;122:460-464. 34. Pompeiano O. Mechanisms responsible for spinal inhibition during
14. Moruzzi G, Magoun H. Brain stem reticular formation and activa- desynchronized sleep: Experimental study. In: Guilleminault C,
tion of the EEG. Electroencephalogr Clin Neurophysiol Dement WC, Passouant P, editors. Advances in Sleep Research, vol
1949;1:455-473. 3: Narcolepsy. New York: Spectrum; 1976. p. 411-449.
15. Starzl TE, Taylor CW, Magoun HW. Collateral afferent excitation 35. Jouvet M. Recherches sur les structures nerveuses et les mecanismes
of reticular formation of brain stem. J Neurophysiol 1951;14:479. responsales des differentes phases du sommeil physiologique. Arch
16. Jasper H, Ajmone-Marsan C. A Stereotaxic Atlas of the Diencepha- Ital Biol 1962;100:125-206.
lon of the Cat. Ottawa, Ontario, Canada: The National Research 36. Evarts E. Effects of sleep and waking on spontaneous and evoked
Council of Canada; 1954. discharge of single units in visual cortex. Fed Proc 1960;4
17. Magoun HW. The ascending reticular system and wakefulness. In: (Suppl.):828-837.
Adrian ED, Bremer F, Jasper HH, editors. Brain mechanisms and 37. Reivich M, Kety S. Blood flow and metabolism in the sleeping brain.
consciousness: a symposium organized by the council for interna- In: Plum F, editor. Brain Dysfunction in Metabolic Disorders. New
tional organizations of medical sciences. Springfield, Ill: Charles C York: Raven Press; 1968. p. 125-140.
Thomas; 1954. 38. Kupfer D, Foster F. Interval between onset of sleep and rapid eye
18. Kryger MH. Sleep apnea: From the needles of Dionysius to continu- movement sleep as an indicator of depression. Lancet 1972;2:684-
ous positive airway pressure. Arch Intern Med 1983;143:2301-2308. 686.
19. Lavie P. Nothing new under the moon: historical accounts of sleep 39. Vogel G. Studies in psychophysiology of dreams, III: The dream of
apnea syndrome. Arch Intern Med 1986;144:2025-2028. narcolepsy. Arch Gen Psychiatry 1960;3:421-428.
20. Lavie P. Restless Nights: Understanding Snoring and Sleep Apnea. 40. Rechtschaffen A, Wolpert E, Dement W, et al. Nocturnal sleep
New Haven, Conn: Yale University Press; 2003. of narcoleptics. Electroencephalogr Clin Neurophysiol 1963;15:
21. Passouant P. Doctor Glineau (1828-1906): narcolepsy centennial. 599-609.
Sleep 1981;3:241-246. 41. Dement W, Rechtschaffen A, Gulevich G. The nature of the narco-
22. Moore-Ede M, Sulzman F, Fuller C. The clocks that time us: Physi- leptic sleep attack. Neurology 1966;16:18-33.
ology of the circadian timing system. Cambridge, Mass: Harvard 42. Fischgold H, editor. La Sommeil de Nuit Normal et Pathologique:
University Press; 1982. Etudes Electroencephalographiques. Paris, France: Masson et Cie;
23. de Toni G. I movimenti pendolari dei bulbi oculari dei bambini 1965.
durante il sonno fisiologico, ed in alcuni stati morbosi. Pediatria 43. Oswald I, Priest R. Five weeks to escape the sleeping pill habit. Br
1933;41:489-498. Med J 1965;2:1093-1095.
CHAPTER 1 History of Sleep Physiology and Medicine 15

44. Kales A, Malmstrom EJ, Scharf MB, et al. Psychophysiological and 59. Yoss R, Moyer N, Hollenhorst R. Pupil size and spontaneous pupil-
biochemical changes following use and withdrawal of hypnotics. In: lary waves associated with alertness, drowsiness, and sleep. Neurol-
Kales A, editor. Sleep: Physiology and Pathology. Philadelphia: JB ogy 1970;20:545-554.
Lippincott; 1969. p. 331-343. 60. Carskadon M, Dement W. Effects of total sleep loss on sleep ten-
45. Kales A, Beall GN, Bajor GF, et al. Sleep studies in asthmatic adults: dency. Percept Mot Skills 1979;48:495-506.
Relationship of attacks to sleep stage and time of night. J Allergy 61. Richardson G, Carskadon M, Flagg W, et al. Excessive daytime
1968;41:164-173. sleepiness in man: Multiple sleep latency measurements in narcolep-
46. Kales A, Heuser G, Jacobson A, et al. All night sleep studies in tic and control subjects. Electroencephalogr Clin Neurophysiol
hypothyroid patients, before and after treatment. J Clin Endocrinol 1978;45:621-627.
Metab 1967;27:1593-1599. 62. Dement W, Carskadon M, Richardson G. Excessive daytime sleepi-
47. Kales A, Ansel RD, Markham CH, et al. Sleep in patients with Par- ness in the sleep apnea syndromes. In: Guilleminault C, Dement W,
kinsons disease and normal subjects prior to and following levodopa editors. Sleep Apnea Syndromes. New York: Alan R Liss. 1978. p.
administration. Clin Pharmacol Ther 1971;12:397-406. 23-46.
48. Kales A, Jacobson A, Paulson NJ, et al. Somnambulism: Psychophysi- 63. Carskadon M, Dement W. Daytime sleepiness: Quantification of a
ological correlates, I: All-night EEG studies. Arch Gen Psychiatry behavioral state. Neurosci Biobehav Rev 1987;11:307-317.
1966;14:586-594. 64. Guilleminault C, Dement W, Passouant P, editors. Narcolepsy. New
49. Gastaut H, Tassinari C, Duron B. Etude polygraphique des manifes- York: Spectrum; 1976.
tations episodiques (hypniques et respiratoires) du syndrome de Pick- 65. Sleep Disorders Classification Committee. Diagnostic classification
wick. Rev Neurol 1965;112:568-579. of sleep and arousal disorders, 1st ed. Association of Sleep Disorders
50. Jung R, Kuhlo W. Neurophysiological studies of abnormal night Centers and the Association for the Psychophysiological Study of
sleep and the pickwickian syndrome. Prog Brain Res 1965;18: Sleep. Sleep 1979;2:1-137.
140-159. 66. Fujita S, Conway W, Zorick F, et al. Surgical correction of anatomic
51. Burwell CS, Robin ED, Whaley RD, et al. Extreme obesity associ- abnormalities in obstructive sleep apnea syndrome: Uvulopalatopha-
ated with alveolar hypoventilation: a pickwickian syndrome. Am J ryngoplasty. Otolaryngol Head Neck Surg 1981; 89:923-934.
Med 1956;21:811-818. 67. Sullivan CE, Issa FG, Berthon-Jones M, et al. Reversal of obstructive
52. Lugaresi E, Coccagna G, Mantovani M. Hypersomnia with Periodic sleep apnea by continuous positive airway pressure applied through
Apneas. New York: Spectrum; 1978. the nares. Lancet 1981;1:862-865.
53. Gastaut H, Lugaresi E, Berti-Ceroni G, et al, editors. The Abnor- 68. Kryger M, Roth T, Dement WC. Principles and Practice of Sleep
malities of Sleep in Man. Bologna, Italy: Aulo Gaggi Editore; Medicine. Philadelphia: WB Saunders; 1989.
1968. 69. International Sleep Medicine Societies. Available at http://dev.ersnet.
54. Gastaut H, Lugaresi E, Berti-Ceroni G, et al. Pathophysiological, org/uploads/Document/15/WEB_CHEMIN_743_11654189
clinical, and nosographic considerations regarding hypersomnia 26.pdf, accessed September 2010.
with periodic breathing. Bull Physiopathol Resp 1972;8:1249- 70. Flemons WW, Douglas NJ, Kuna ST, et al. Access to diagnosis and
1256. treatment of patients with suspected sleep apnea. Am J Respir Crit
55. Dement W, Guilleminault C, Zarcone V, et al. The narcolepsy syn- Care Med 2004;169:668-672.
drome. In: Conn H, Conn R, editors. Current diagnosis, vol. 2. 71. Lefant C, Kiley JP. Sleep research: Celebration and opportunity.
Philadelphia: WB Saunders; 1974. p. 917-921. Sleep 1998;21:665-669.
56. Lugaresi E, Coccagna G, Mantovani M, et al. Effects de la trachoto- 72. National Heart, Lung, and Blood Institute. Sleep Disorders Informa-
mie dans les hypersomnies avec respiration periodique. Rev Neurol tion. Available at http://www.nhlbi.nih.gov/about/ncsdr/. Accessed
1970;123:267-268. September 2010.
57. Guilleminault C, Dement W. 235 cases of excessive daytime sleepi- 73. National Sleep Foundation. Available at http://www.sleepfound
ness: Diagnosis and tentative classification. J Neurol Sci 1977;31: ation.org. Accessed September 2010.
13-27. 74. Colten H, Altevogt B, editors. Sleep Disorders and Sleep Depriva-
58. Hoddes E, Zarcone V, Smythe H, et al. Quantification of sleepiness: tion: An Unmet Public Health Problem. Washington DC: National
A new approach. Psychophysiology 1973;10:431-436. Academies Press; 2006.
Normal Human Sleep: Chapter
An Overview
Mary A. Carskadon and William C. Dement 2
Abstract minutes); coherent sleep stages emerge as the brain matures
during the first year. At birth, active sleep is approximately
Normal human sleep comprises two statesrapid eye move- 50% of total sleep and declines over the first 2 years to
ment (REM) and nonREM (NREM) sleepthat alternate cycli- approximately 20% to 25%. NREM sleep slow waves are not
cally across a sleep episode. State characteristics are well present at birth but emerge in the first 2 years. Slow-wave
defined: NREM sleep includes a variably synchronous cortical sleep (stages 3 and 4) decreases across adolescence by 40%
electroencephalogram (EEG; including sleep spindles, K- from preteen years and continues a slower decline into old
complexes, and slow waves) associated with low muscle tonus age, particularly in men and less so in women. REM sleep as
and minimal psychological activity; the REM sleep EEG is a percentage of total sleep is approximately 20% to 25%
desynchronized, muscles are atonic, and dreaming is typical. across childhood, adolescence, adulthood, and into old age
A nightly pattern of sleep in mature humans sleeping on a except in dementia. Other factors predictably alter sleep, such
regular schedule includes several reliable characteristics: as previous sleep-wake history (e.g., homeostatic load), phase
Sleep begins in NREM and progresses through deeper NREM of the circadian timing system, ambient temperature, drugs,
stages (stages 2, 3, and 4 using the classic definitions, or and sleep disorders.
stages N2 and N3 using the updated definitions) before the A clear appreciation of the normal characteristics of sleep
first episode of REM sleep occurs approximately 80 to 100 provides a strong background and template for understand-
minutes later. Thereafter, NREM sleep and REM sleep cycle ing clinical conditions in which normal characteristics are
with a period of approximately 90 minutes. NREM stages 3 altered, as well as for interpreting certain consequences of
and 4 (or stage N3) concentrate in the early NREM cycles, and sleep disorders. In this chapter, the normal young adult sleep
REM sleep episodes lengthen across the night. pattern is described as a working baseline pattern. Normative
Age-related changes are also predictable: Newborn humans changes due to aging and other factors are described with
enter REM sleep (called active sleep) before NREM (called quiet that background in mind. Several major sleep disorders are
sleep) and have a shorter sleep cycle (approximately 50 highlighted by their differences from the normative pattern.

SLEEP DEFINITIONS sleep is usually associated with minimal or fragmentary

According to a simple behavioral definition, sleep is a mental activity. A shorthand definition of NREM sleep is
reversible behavioral state of perceptual disengagement a relatively inactive yet actively regulating brain in a
from and unresponsiveness to the environment. It is also movable body.
true that sleep is a complex amalgam of physiologic and REM sleep, by contrast, is defined by EEG activation,
behavioral processes. Sleep is typically (but not necessarily) muscle atonia, and episodic bursts of rapid eye movements.
accompanied by postural recumbence, behavioral quies- REM sleep usually is not divided into stages, although
cence, closed eyes, and all the other indicators one com- tonic and phasic types of REM sleep are occasionally dis-
monly associates with sleeping. In the unusual circumstance, tinguished for certain research purposes. The distinction
other behaviors can occur during sleep. These behaviors of tonic versus phasic is based on short-lived events such
can include sleepwalking, sleeptalking, teeth grinding, as eye movements that tend to occur in clusters separated
and other physical activities. Anomalies involving sleep by episodes of relative quiescence. In cats, REM sleep
processes also include intrusions of sleepsleep itself, phasic activity is epitomized by bursts of ponto-geniculo-
dream imagery, or muscle weaknessinto wakefulness, for occipital (PGO) waves, which are accompanied peripher-
example (Box 2-1). ally by rapid eye movements, twitching of distal muscles,
Within sleep, two separate states have been defined on middle ear muscle activity, and other phasic events that
the basis of a constellation of physiologic parameters. correspond to the phasic event markers easily measurable
These two states, rapid eye movement (REM) and in human beings. As described in Chapter 141, PGO waves
non-REM (NREM), exist in virtually all mammals and are not usually detectable in human beings. Thus, the most
birds yet studied, and they are as distinct from one another commonly used marker of REM sleep phasic activity in
as each is from wakefulness. human beings is, of course, the bursts of rapid eye move-
NREM (pronounced non-REM) sleep is convention- ments (Fig. 2-2); muscle twitches and cardiorespiratory
ally subdivided into four stages defined along one measure- irregularities often accompany the REM bursts. The
ment axis, the electroencephalogram (EEG). The EEG mental activity of human REM sleep is associated with
pattern in NREM sleep is commonly described as synchro- dreaming, based on vivid dream recall reported after
nous, with such characteristic waveforms as sleep spindles, approximately 80% of arousals from this state of sleep.1
K-complexes, and high-voltage slow waves (Fig. 2-1). The Inhibition of spinal motor neurons by brainstem mecha-
four NREM stages (stages 1, 2, 3, and 4) roughly parallel nisms mediates suppression of postural motor tonus in
a depth-of-sleep continuum, with arousal thresholds gen- REM sleep. A shorthand definition of REM sleep, there-
erally lowest in stage 1 and highest in stage 4 sleep. NREM fore, is an activated brain in a paralyzed body.

CHAPTER 2 Normal Human Sleep: An Overview 17

Box 2-1 Sleep Medicine Methodology

and Nomenclature Stage 1

In 2007, the American Academy of Sleep Medicine

(AASM) published a new manual (see reference 50)
for scoring sleep and associated events. This manual Stage 2
recommends alterations to recording methodology
and terminology that the Academy will demand of
clinical laboratories in the future. Although specifica-
tion of arousal, cardiac, movement, and respiratory Stage 3
rules appear to be value added to the assessment of
sleep-related events, the new rules, terminology, and
technical specifications for recording and scoring
sleep are not without controversy. Stage 4
The current chapter uses the traditional terminol-
ogy and definitions, upon which most descriptive
and experimental research has been based since the 100 V
1960s.17 Hence, where the AASM terminology uses
the term N for NREM sleep stages and R for REM sleep 5 sec
stages, N1 and N2 are used instead of stage 1 and
stage 2; N3 is used to indicate the sum of stage 3 Figure 2-1 The stages of nonrapid eye movement sleep. The
and stage 4 (often called slow-wave sleep in human four electroencephalogram tracings depicted here are from a
literature); R is used to name REM sleep. Another 19-year-old female volunteer. Each tracing was recorded from
change is to the nomenclature for the recording a referential lead (C3/A2) recorded on a Grass Instruments Co.
placements. Hence, calling the auricular placements (West Warwick, R.I.) Model 7D polygraph with a paper speed of
M1 and M2 (rather than A1 and A2) is unnecessary 10mm/sec, time constant of 0.3 sec, and 12 -amplitude high-
frequency setting of 30Hz. On the second tracing, the arrow
and places the sleep EEG recording terminology
indicates a K-complex and the underlining shows two sleep
outside the pale for EEG recording terminology in spindles.
other disciplines. Although these are somewhat
trivial changes, changes in nomenclature can result
in confusion when attempting to compare to previ-
ous literature and established data sets and are of C3/A2
concern for clinicians and investigators who com-
municate with other fields. ROC/A1
Of greater concern are changes to the core record-
ing and scoring recommendations that the AASM LOC/A2
manual recommends. For example, the recommended 50 V 3 sec
scoring montage requires using a frontal (F3 or F4) CHIN EMG
EEG placement for use with visual scoring of the
recordings, rather than the central (C3 or C4) EEG
placements recommended in the standard manual. Figure 2-2 Phasic events in human rapid eye movement (REM)
The rationale for the change is that the frontal place- sleep. On the left side is a burst of several rapid eye movements
(out-of-phase deflections in right outer canthus [ROC]/A1 and
ments pick up more slow-wave activity during sleep.
left outer canthus [LOC]/A2). On the right side, there are addi-
The consequence, however, is that sleep studies per- tional rapid eye movements as well as twitches on the electro-
formed and scored with the frontal EEG cannot be myographic (EMG) lead. The interval between eye movement
compared to normative or clinical data and the frontal bursts and twitches illustrates tonic REM sleep.
placements also truncate the ability to visualize sleep
spindles. Furthermore, developmental changes to the
regional EEG preclude the universal assumption that sleep through REM sleep can be a diagnostic sign in adult
sleep slow-wave activity is a frontal event. patients with narcolepsy.
Other issues are present in this new AASM approach
to human sleep; however, this is not the venue for a Definition of Sleep Onset
complete description of such concerns. In summary,
The precise definition of the onset of sleep has been a topic
the AASM scoring manual has not yet become the uni-
versal standard for assessing human sleep and might of debate, primarily because there is no single measure that
not achieve that status in its current form. Specifica- is 100% clear-cut 100% of the time. For example, a change
tions for recording and scoring sleep are not without in EEG pattern is not always associated with a persons
controversy.51-56 perception of sleep, yet even when subjects report that they
are still awake, clear behavioral changes can indicate the
presence of sleep. To begin a consideration of this issue,
let us examine the three basic polysomnographic measures
SLEEP ONSET of sleep and how they change with sleep onset. The elec-
The onset of sleep under normal circumstances in normal trode placements are described in Chapter 141.
adult humans is through NREM sleep. This fundamental
principle of normal human sleep reflects a highly reliable Electromyogram
finding and is important in considering normal versus The electromyogram (EMG) may show a gradual diminu-
pathologic sleep. For example, the abnormal entry into tion of muscle tonus as sleep approaches, but rarely does
18 PARTI / Section 1 Normal Sleep and Its Variations

C3/A2 C3/A2

O2/A1 O2/A1

50 V 3 sec 50 V 3 sec

Figure 2-3 The transition from wakefulness to stage 1 sleep. Figure 2-4 A common wake- to-sleep transition pattern. Note
The most marked change is visible on the two electroencepha- that the electroencephalographic pattern changes from wake
lographic (EEG) channels (C3/A2 and O2/A1), where a clear (rhythmic alpha) to stage 1 (relatively low-voltage, mixed-
pattern of rhythmic alpha activity (8 cps) changes to a relatively frequency) sleep twice during this attempt to fall asleep. EMG,
low-voltage, mixed- frequency pattern at about the middle of electromyogram; LOC, left outer canthus; ROC, right outer
the figure. The level of electromyographic (EMG) activity does canthus.
not change markedly. Slow eye movements (right outer canthus
[ROC]/left outer canthus [LOC]) are present throughout this
episode, preceding the EEG change by at least 20 seconds. In
general, the change in EEG patterns to stage 1 as illustrated
here is accepted as the onset of sleep. C3/A2

a discrete EMG change pinpoint sleep onset. Further- SEMs Asleep SEMs
more, the presleep level of the EMG, particularly if the LOC/A2
person is relaxed, can be entirely indistinguishable from
that of unequivocal sleep (Fig. 2-3). SAT Gap

Electrooculogram Gap
As sleep approaches, the electrooculogram (EOG) shows
Figure 2-5 Failure to perform a simple behavioral task at the
slow, possibly asynchronous eye movements (see Fig. 2-3) onset of sleep. The volunteer had been deprived of sleep over-
that usually disappear within several minutes of the EEG night and was required to tap two switches alternately, shown
changes described next. Occasionally, the onset of these as pen deflections of opposite polarity on the channel labeled
slow eye movements coincides with a persons perceived SAT. When the electroencephalographic (EEG; C3/A2) pattern
changes to stage 1 sleep, the behavior stops, returning when
sleep onset; more often, subjects report that they are still the EEG pattern reverts to wakefulness. LOC, left outer canthus;
awake. ROC, right outer canthus; SEMs, slow eye movements. (From
Carskadon MA, Dement WC. Effects of total sleep loss on sleep
Electroencephalogram tendency. Percept Mot Skills 1979;48:495-506. Perceptual
In the simplest circumstance (see Fig. 2-3), the EEG and Motor Skills, 1979.)
changes from a pattern of clear rhythmic alpha (8 to 13
cycles per second [cps]) activity, particularly in the occipi-
tal region, to a relatively low-voltage, mixed-frequency that another EEG sleep pattern does not intervene. One
pattern (stage 1 sleep). This EEG change usually occurs might not always be able to pinpoint this transition to the
seconds to minutes after the start of slow eye movements. millisecond, but it is usually possible to determine the
With regard to introspection, the onset of a stage 1 EEG change reliably within several seconds.
pattern may or may not coincide with perceived sleep
onset. For this reason, a number of investigators require Behavioral Concomitants of Sleep Onset
the presence of specific EEG patternsthe K-complex or Given the changes in the EEG that accompany the onset
sleep spindle (i.e., stage 2 sleep)to acknowledge sleep of sleep, what are the behavioral correlates of the wake-to-
onset. Even these stage 2 EEG patterns, however, are not sleep transition? The following material reviews a few
unequivocally associated with perceived sleep.2 A further common behavioral concomitants of sleep onset. Keep in
complication is that sleep onset often does not occur all at mind that different functions may be depressed in differ-
once; instead, there may be a wavering of vigilance before ent sequence and to different degrees in different subjects
unequivocal sleep ensues (Fig. 2-4). Thus, it is difficult and on different occasions (p. 35).3
to accept a single variable as marking sleep onset. As Davis
and colleagues3 wrote many years ago (p. 35): Simple Behavioral Task
Is falling asleep a unitary event? Our observations In the first example, volunteers were asked to tap two
suggest that it is not. Different functions, such as sensory switches alternately at a steady pace. As shown in Figure
awareness, memory, self-consciousness, continuity of logi- 2-5, this simple behavior continues after the onset of slow
cal thought, latency of response to a stimulus, and altera- eye movements and may persist for several seconds after
tions in the pattern of brain potentials all go in parallel in the EEG changes to a stage 1 sleep pattern.4 The behavior
a general way, but there are exceptions to every rule. Nev- then ceases, usually to recur only after the EEG reverts to
ertheless, a reasonable consensus exists that the EEG a waking pattern. This is an example of what one may think
change to stage 1, usually heralded or accompanied by slow of as the simplest kind of automatic behavior pattern.
eye movements, identifies the transition to sleep, provided Because such simple behavior can persist past sleep onset
CHAPTER 2 Normal Human Sleep: An Overview 19

and as one passes in and out of sleep, it might explain how sleep, for example, ones own name spoken softly will
impaired, drowsy drivers are able to continue down the produce an arousal; a similarly applied nonmeaningful
highway. stimulus will not. Similarly, a sleeping mother is more
likely to hear her own babys cry than the cry of an
Visual Response unrelated infant.
A second example of behavioral change at sleep onset Williams and colleagues10 showed that the likelihood of
derives from an experiment in which a bright light is placed an appropriate response during sleep was improved
in front of the subjects eyes, and the subject is asked to when an otherwise nonmeaningful stimulus was made
respond when a light flash is seen by pressing a sensitive meaningful by linking the absence of response to pun-
microswitch taped to the hand.5 When the EEG pattern ishment (a loud siren, flashing light, and the threat of
is stage 1 or stage 2 sleep, the response is absent more than an electric shock).
85% of the time. When volunteers are queried afterward, From these examples and others, it seems clear that
they report that they did not see the light flash, not that sensory processing at some level does continue after the
they saw the flash but the response was inhibited. This is onset of sleep. Indeed, one study has shown with functional
one example of the perceptual disengagement from the magnetic resonance imaging that regional brain activation
environment that accompanies sleep onset. occurs in response to stimuli during sleep and that differ-
ent brain regions (middle temporal gyrus and bilateral
Auditory Response orbitofrontal cortex) are activated in response to meaning-
In another sensory domain, the response to sleep onset is ful (persons own name) versus nonmeaningful (beep)
examined with a series of tones played over earphones to stimuli.11
a subject who is instructed to respond each time a tone is
heard. One study of this phenomenon showed that reac- Hypnic Myoclonia
tion times became longer in proximity to the onset of stage What other behaviors accompany the onset of sleep? If you
1 sleep, and responses were absent coincident with a awaken and query someone shortly after the stage 1 sleep
change in EEG to unequivocal sleep.6 For responses in EEG pattern appears, the person usually reports the mental
both visual and auditory modalities, the return of the experience as one of losing a direct train of thought and
response after its sleep-related disappearance typically of experiencing vague and fragmentary imagery, usually
requires the resumption of a waking EEG pattern. visual.12 Another fairly common sleep-onset experience is
hypnic myoclonia, which is experienced as a general or
Olfactory Response localized muscle contraction very often associated with
When sleeping humans are tasked to respond when they rather vivid visual imagery. Hypnic myoclonias are not
smell something, the response depends in part on sleep pathologic events, although they tend to occur more com-
state and in part on the particular odorant. In contrast to monly in association with stress or with unusual or irregu-
visual responses, one study showed that responses to lar sleep schedules.
graded strengths of peppermint (strong trigeminal stimu- The precise nature of hypnic myoclonias is not clearly
lant usually perceived as pleasant) and pyridine (strong understood. According to one hypothesis, the onset of
trigeminal stimulant usually perceived as extremely sleep in these instances is marked by a dissociation of REM
unpleasant) were well maintained during initial stage 1 sleep components, wherein a breakthrough of the imagery
sleep.7 As with other modalities, the response in other component of REM sleep (hypnagogic hallucination)
sleep stages was significantly poorer: Peppermint simply occurs in the absence of the REM motor inhibitory com-
was not consciously smelled in stages 2 and 4 NREM sleep ponent. A response by the individual to the image, there-
or in REM sleep; pyridine was never smelled in stage 4 fore, results in a movement or jerk. The increased frequency
sleep, and only occasionally in stage 2 NREM and in REM of these events in association with irregular sleep schedules
sleep.7 On the other hand, a tone successfully aroused the is consistent with the increased probability of REM sleep
young adult participants in every stage. One conclusion of occurring at the wake-to-sleep transition under such con-
this report was that the olfactory system of humans is not ditions (see later). Although the usual transition in adult
a good sentinel system during sleep. human beings is to NREM sleep, the REM portal into
sleep, which is the norm in infancy, can become partially
Response to Meaningful Stimuli opened under unusual circumstances.
One should not infer from the preceding studies that the
mind becomes an impenetrable barrier to sensory input at Memory Near Sleep Onset
the onset of sleep. Indeed, one of the earliest modern What happens to memory at the onset of sleep? The tran-
studies of arousability during sleep showed that sleeping sition from wake to sleep tends to produce a memory
human beings were differentially responsive to auditory impairment. One view is that it is as if sleep closes the gate
stimuli of graded intensity.8 Another way of illustrating between short-term and long-term memory stores. This
sensory sensitivity is shown in experiments that have phenomenon is best described by the following experi-
assessed discriminant responses during sleep to meaningful ment.13 During a presleep testing session, word pairs were
versus nonmeaningful stimuli, with meaning supplied in a presented to volunteers over a loudspeaker at 1-minute
number of ways and response usually measured as evoked intervals. The subjects were then awakened either 30
K-complexes or arousal. The following are examples. seconds or 10 minutes after the onset of sleep (defined as
A person tends to have a lower arousal threshold for his EEG stage 1) and asked to recall the words presented
or her own name versus someone elses name.9 In light before sleep onset. As illustrated in Figure 2-6, the
20 PARTI / Section 1 Normal Sleep and Its Variations

40 Wake
Correct recall percentage

30 S2


24 1 2 3 4 5 6 7
10 9 8 7 6 5 4 3 2 1 Figure 2-7 The progression of sleep stages across a single
Sleep night in a normal young adult volunteer is illustrated in this
Word presentation minutes onset sleep histogram. The text describes the ideal or average pattern.
before sleep onset
This histogram was drawn on the basis of a continuous over-
night recording of electroencephalogram, electrooculogram,
Subjects awakened 30 seconds after sleep onset and electromyogram in a normal 19-year-old man. The record
Subjects awakened 10 minutes after sleep onset was assessed in 30-second epochs for the various sleep stages.
REM, rapid eye movement.
Figure 2-6 Memory is impaired by sleep, as shown by the
study results illustrated in this graph. See text for explanation.
Learning and Sleep
In contrast to this immediate sleep-related forgetting,
30-second condition was associated with a consistent level the relevance for sleep to human learningparticularly for
of recall from the entire 10 minutes before sleep onset. consolidation of perceptual and motor learningis of
(Primacy and recency effects are apparent, although not growing interest.14,15 The importance of this association
large.) In the 10-minute condition, however, recall paral- has also generated some debate and skepticism.16 Never-
leled that in the 30-second group for only the 10 to 4 theless, a spate of recent research is awakening renewed
minutes before sleep onset and then fell abruptly from that interest in the topic, and mechanistic studies explaining the
point until sleep onset. roles of REM and NREM sleep more precisely are under
In the 30-second condition, therefore, both longer-term examination (see Chapter 29).
(4 to 10 minutes) and shorter-term (0 to 3 minutes)
memory stores remained accessible. In the 10-minute con- PROGRESSION OF SLEEP ACROSS
dition, by contrast, words that were in longer-term stores THE NIGHT
(4 to 10 minutes) before sleep onset were accessible,
whereas words that were still in shorter-term stores (0 to Pattern of Sleep in a Normal Young Adult
3 minutes) at sleep onset were no longer accessible; that The simplest description of sleep begins with the ideal
is, they had not been consolidated into longer-term case, the normal young adult who is sleeping well and on
memory stores. One conclusion of this experiment is that a fixed schedule of about 8 hours per night (Fig. 2-7). In
sleep inactivates the transfer of storage from short- to general, no consistent male versus female distinctions have
long-term memory. Another interpretation is that encod- been found in the normal pattern of sleep in young adults.
ing of the material before sleep onset is of insufficient In briefest summary, the normal human adult enters sleep
strength to allow recall. The precise moment at which this through NREM sleep, REM sleep does not occur until 80
deficit occurs is not known and may be a continuing minutes or longer thereafter, and NREM sleep and REM
process, perhaps reflecting anterograde amnesia. Never- sleep alternate through the night, with an approximately
theless, one may infer that if sleep persists for approxi- 90-minute cycle (see Chapter 141 for a full description of
mately 10 minutes, memory is lost for the few minutes sleep stages).
before sleep. The following experiences represent a few
familiar examples of this phenomenon: First Sleep Cycle
Inability to grasp the instant of sleep onset in your The first cycle of sleep in the normal young adult begins
memory. with stage 1 sleep, which usually persists for only a few (1
Forgetting a telephone call that had come in the middle to 7) minutes at the onset of sleep. Sleep is easily dis
of the night. continued during stage 1 by, for example, softly calling a
Forgetting the news you were told when awakened in persons name, touching the person lightly, quietly closing
the night. a door, and so forth. Thus, stage 1 sleep is associated with
Not remembering the ringing of your alarm clock. a low arousal threshold. In addition to its role in the initial
Experiencing morning amnesia for coherent wake-to-sleep transition, stage 1 sleep occurs as a transi-
sleeptalking. tional stage throughout the night. A common sign of
Having fleeting dream recall. severely disrupted sleep is an increase in the amount and
Patients with syndromes of excessive sleepiness can experi- percentage of stage 1 sleep.
ence similar memory problems in the daytime if sleep Stage 2 NREM sleep, signaled by sleep spindles or
becomes intrusive. K-complexes in the EEG (see Fig. 2-1), follows this brief
CHAPTER 2 Normal Human Sleep: An Overview 21

episode of stage 1 sleep and continues for approximately first one third); REM sleep episodes are longest in the last
10 to 25 minutes. In stage 2 sleep, a more intense stimulus one third of the night. Brief episodes of wakefulness tend
is required to produce arousal. The same stimulus that to intrude later in the night, usually near REM sleep tran-
produced arousal from stage 1 sleep often results in an sitions, and they usually do not last long enough to be
evoked K-complex but no awakening in stage 2 sleep. remembered in the morning. The preferential distribution
As stage 2 sleep progresses, high-voltage slow-wave of REM sleep toward the latter portion of the night in
activity gradually appears in the EEG. Eventually, this normal human adults is thought to be linked to a circadian
activity meets the criteria17 for stage 3 NREM sleep, that oscillator, which can be gauged by the oscillation of
is, high-voltage (at least 75V) slow-wave (2 cps) activity body temperature.18,19 The preferential distribution of
accounting for more than 20% but less than 50% of the SWS toward the beginning of a sleep episode is not
EEG activity. Stage 3 sleep usually lasts only a few minutes thought to be mediated by circadian processes but shows
in the first cycle and is transitional to stage 4 as more and a marked response to the length of prior wakefulness.20
more high-voltage slow-wave activity occurs. Stage 4 The SWS pattern reflects the homeostatic sleep system,
NREM sleepidentified when the high-voltage slow- highest at sleep onset and diminishing across the night as
wave activity comprises more than 50% of the record sleep pressure wanes. Thus, these aspects of the normal
usually lasts approximately 20 to 40 minutes in the first sleep pattern highlight features of the two-process model
cycle. An incrementally larger stimulus is usually required of sleep as elaborated on in Chapter 37.
to produce an arousal from stage 3 or 4 sleep than from
stage 1 or 2 sleep. (Investigators often refer to the com- Length of Sleep
bined stages 3 and 4 sleep as slow-wave sleep [SWS], delta The length of nocturnal sleep depends on a great number
sleep, or deep sleep.) of factorsof which volitional control is among the most
A series of body movements usually signals an ascent significant in human beingsand it is thus difficult to
to lighter NREM sleep stages. A brief (1- or 2-minute) characterize a normal pattern. Most young adults report
episode of stage 3 sleep might occur, followed by perhaps sleeping approximately 7.5 hours a night on weekday
5 to 10 minutes of stage 2 sleep interrupted by body move- nights and slightly longer, 8.5 hours, on weekend nights.
ments preceding the initial REM episode. REM sleep in The variability of these figures from person to person and
the first cycle of the night is usually short-lived (1 to 5 from night to night, however, is quite high. Sleep length
minutes). The arousal threshold in this REM episode is also depends on genetic determinants,21 and one may think
variable, as is true for REM sleep throughout the night. of the volitional determinants (staying up late, waking by
Theories to explain the variable arousal threshold of REM alarm, and so on) superimposed on the background of a
sleep have suggested that at times, the persons selective genetic sleep need. Length of prior waking also affects how
attention to internal stimuli precludes a response or that much sleeps, although not in a one-for-one manner.
the arousal stimulus is incorporated into the ongoing Indeed, the length of sleep is also determined by processes
dream story rather than producing an awakening. Certain associated with circadian rhythms. Thus, when one sleeps
early experiments examining arousal thresholds in cats helps to determine how long one sleeps. In addition, as sleep
found highest thresholds in REM sleep, which was then is extended, the amount of REM sleep increases, because
termed deep sleep in this species. Although this terminology REM sleep depends on the persistence of sleep into the
is still often used in publications about sleep in animals, it peak circadian time in order to occur.
should not be confused with human NREM stages 3 plus
4 sleep, which is also often called deep sleep. One should Generalizations about Sleep in
also note that SWS is sometimes used (as is synchronized the Normal Young Adult
sleep) as a synonym for all of NREM sleep in other species A number of general statements can be made regarding
and is thus distinct from SWS (stages 3 plus 4 NREM) in sleep in the normal young adult who is living on a conven-
human beings. tional sleep-wake schedule and who is without sleep
NREM-REM Cycle Sleep is entered through NREM sleep.
NREM sleep and REM sleep continue to alternate through NREM sleep and REM sleep alternate with a period
the night in cyclic fashion. REM sleep episodes usually near 90 minutes.
become longer across the night. Stages 3 and 4 sleep SWS predominates in the first third of the night and is
occupy less time in the second cycle and might disappear linked to the initiation of sleep and the length of time
altogether from later cycles, as stage 2 sleep expands to awake.
occupy the NREM portion of the cycle. The average REM sleep predominates in the last third of the night
length of the first NREM-REM sleep cycle is approxi- and is linked to the circadian rhythm of body
mately 70 to 100 minutes; the average length of the second temperature.
and later cycles is approximately 90 to 120 minutes. Across Wakefulness in sleep usually accounts for less than 5%
the night, the average period of the NREM-REM cycle is of the night.
approximately 90 to 110 minutes. Stage 1 sleep generally constitutes approximately 2% to
5% of sleep.
Distribution of Sleep Stages across Stage 2 sleep generally constitutes approximately 45%
the Night to 55% of sleep.
In the young adult, SWS dominates the NREM portion Stage 3 sleep generally constitutes approximately 3% to
of the sleep cycle toward the beginning of the night (the 8% of sleep.
22 PARTI / Section 1 Normal Sleep and Its Variations

Stage 4 sleep generally constitutes approximately 10% sity of slow-wave activity early in the night. A similar,
to 15% of sleep. although less profound qualitative difference distinguishes
NREM sleep, therefore, is usually 75% to 80% SWS occurring in the first and later cycles of the night in
of sleep. a given person. The quantitative change in SWS may best
REM sleep is usually 20% to 25% of sleep, occurring in be seen across adolescence, when SWS decreases by nearly
four to six discrete episodes. 40% during the second decade, even when length of noc-
turnal sleep remains constant.23 Feinberg24 hypothesized
Factors Modifying Sleep that the age-related decline in nocturnal SWS might paral-
Stage Distribution lel loss of cortical synaptic density. By midadolescence,
Age youngsters no longer typically skip their first REM, and
The strongest and most consistent factor affecting the their sleep resembles that described earlier for young
pattern of sleep stages across the night is age (Fig. 2-8). adults. By age 60 years, SWS might no longer be present,
The most marked age-related differences in sleep from the particularly in men. Women appear to maintain SWS later
patterns described earlier are found in newborn infants. into life than men.
For the first year of life, the transition from wake to sleep REM sleep as a percentage of total sleep is maintained
is often accomplished through REM sleep (called active well into healthy old age; the absolute amount of REM
sleep in newborns). The cyclic alternation of NREM-REM sleep at night has been correlated with intellectual func-
sleep is present from birth but has a period of approxi- tioning25 and declines markedly in the case of organic brain
mately 50 to 60 minutes in the newborn compared with dysfunctions of the elderly.26
approximately 90 minutes in the adult. Infants also only Arousals during sleep increase markedly with age.
gradually acquire a consolidated nocturnal sleep cycle, Extended wake episodes of which the individual is aware
and the fully developed EEG patterns of the NREM and can report, as well as brief and probably unremem-
sleep stages are not present at birth but emerge over the bered arousals, increase with aging.27 The latter type of
first 2 to 6 months of life. When brain structure and func- transient arousals may occur with no known correlate but
tion achieve a level that can support high-voltage slow- are often associated with occult sleep disturbances, such as
wave EEG activity, NREM stages 3 and 4 sleep become periodic limb movements during sleep (PLMS) and sleep-
prominent. related respiratory irregularities, which also become more
SWS is maximal in young children and decreases mark- prevalent in later life.28,29
edly with age. The SWS of young children is both quali- Perhaps the most notable finding regarding sleep in the
tatively and quantitatively different from that of older elderly is the profound increase in interindividual vari-
adults. For example, it is nearly impossible to wake young- ability,30 which thus precludes generalizations such as
sters in the SWS of the nights first sleep cycle. In one those made for young adults.
study,22 a 123-dB tone failed to produce any sign of arousal
in a group of children whose mean age was 10 years. In Prior Sleep History
addition, children up to midadolescence often skip their A person who has experienced sleep loss on one or more
first REM episode, perhaps due to the quantity and inten- nights shows a sleep pattern that favors SWS during recov-
ery (Fig. 2-9). Recovery sleep is also usually prolonged and
deeperthat is, having a higher arousal threshold through-
outthan basal sleep. REM sleep tends to show a rebound
on the second or subsequent recovery nights after an
600 episode of sleep loss. Therefore, with total sleep loss, SWS
tends to be preferentially recovered compared with REM
500 Sleep latency sleep, which tends to recover only after the recuperation
400 WASO
of SWS.
Cases in which a person is differentially deprived of
300 REM or SWSeither operationally, by being awakened
each time the sleep pattern occurs, or pharmacologically
200 (see later)show a preferential rebound of that stage of
Stage 2
sleep when natural sleep is resumed. This phenomenon has
Stage 1 particular relevance in a clinical setting, in which abrupt
0 withdrawal from a therapeutic regimen can result in mis-
5 10 15 25 35 45 55 65 75 85
leading diagnostic findings (e.g., sleep-onset REM periods
[SOREMPs] as a result of a REM sleep rebound) or could
conceivably exacerbate a sleep disorder (e.g., if sleep apneas
tend to occur preferentially or with greater intensity in the
Figure 2-8 Changes in sleep with age. Time (in minutes) for
sleep latency and wake time after sleep onset (WASO) and for rebounding stage of sleep).
rapid eye movement (REM) sleep and non-REM (NREM) sleep Chronic restriction of nocturnal sleep, an irregular
stages 1, 2, and slow wave sleep (SWS). Summary values are sleep schedule, or frequent disturbance of nocturnal sleep
given for ages 5 to 85 years. (Ohayon M, Carskadon MA, Guil- can result in a peculiar distribution of sleep states, most
leminault C, etal. Meta-analysis of quantitative sleep parame-
ters from childhood to old age in healthy individuals: developing
commonly characterized by premature REM sleep, that
normative sleep values across the human lifespan. Sleep is, SOREMPs. Such episodes can be associated with hyp-
2004;27:1255-1273.) nagogic hallucinations, sleep paralysis, or an increased
CHAPTER 2 Normal Human Sleep: An Overview 23

circadian body temperature phase position shows that

Wake sleep onset is likeliest to occur on the falling limb of the
S1 temperature cycle. A secondary peak of sleep onsets, cor-
S2 responding to afternoon napping, also occurs; the offset of
sleep occurs most often on the rising limb of the circadian
body temperature curve.34
movement Extremes of temperature in the sleeping environment tend
23 24 1 2 3 4 5 6 7 8
to disrupt sleep. REM sleep is commonly more sensitive
Wake to temperature-related disruption than is NREM sleep.
S1 Accumulated evidence from human beings and other
species suggests that mammals have only minimal ability
to thermoregulate during REM sleep; in other words, the
control of body temperature is virtually poikilothermic in
S4 REM sleep.35 This inability to thermoregulate in REM
REM sleep probably affects the response to temperature extremes
and suggests that such conditions are less of a problem
23 24 1 2 3 4 5 6 7 8 early during a night than late, when REM sleep tends to
predominate. It should be clear, as well, that such responses
Figure 2-9 The upper histogram shows the baseline sleep as sweating or shivering during sleep under ambient tem-
pattern of a normal 14-year-old female volunteer. The lower
histogram illustrates the sleep pattern in this volunteer for the
perature extremes occur in NREM sleep and are limited
first recovery night after 38 hours without sleep. Note that the in REM sleep.
amount of stage 4 sleep on the lower graph is greater than on
baseline, and the first rapid eye movement (REM) sleep episode Drug Ingestion
is markedly delayed. The distribution of sleep states and stages is affected by
many common drugs, including those typically prescribed
in the treatment of sleep disorders as well as those not
specifically related to the pharmacotherapy of sleep dis
incidence of hypnic myoclonia in persons with no organic orders and those used socially or recreationally. Whether
sleep disorder. changes in sleep stage distribution have any relevance to
Although not strictly related to prior sleep history, the health, illness, or psychological well-being is unknown;
first night of a laboratory sleep evaluation is commonly however, particularly in the context of specific sleep dis
associated with a disruption of the normal distribution orders that differentially affect one sleep stage or another,
of sleep states, characterized chiefly by a delayed onset such distinctions may be relevant to diagnosis or treat-
of REM sleep.31 Often this delay takes the form of skip- ment. A number of generalizations regarding the effects of
ping the first REM episode of the night. In other words, certain of the more commonly used compounds on sleep
the NREM sleep stages progress in a normal fashion, stage distribution can be made.
but the first cycle ends with an episode of stage 1 or a Benzodiazepines tend to suppress SWS and have no
brief arousal instead of the expected brief REM sleep consistent effect on REM sleep.
episode. In addition, REM sleep episodes are often dis- Tricyclic antidepressants, monoamine oxidase inhibi-
rupted, and the total amount of REM sleep on the first tors, and certain selective serotonin reuptake inhibitors
night in the sleep laboratory is also usually reduced from tend to suppress REM sleep. An increased level of motor
the normal value. activity during sleep occurs with certain of these com-
pounds, leading to a pattern of REM sleep without
Circadian Rhythms motor inhibition or an increased incidence of PLMS.
The circadian phase at which sleep occurs affects the dis- Fluoxetine is also associated with rapid eye movements
tribution of sleep stages. REM sleep, in particular, occurs across all sleep stages (Prozac eyes).
with a circadian distribution that peaks in the morning Withdrawal from drugs that selectively suppress a stage
hours coincident with the trough of the core body tem- of sleep tends to be associated with a rebound of that
perature rhythm.18,19 Thus, if sleep onset is delayed until sleep stage. Thus, acute withdrawal from a benzodiaz-
the peak REM phase of the circadian rhythmthat is, the epine compound is likely to produce an increase of
early morningREM sleep tends to predominate and can SWS; acute withdrawal from a tricyclic antidepressant
even occur at the onset of sleep. This reversal of the or monoamine oxidase inhibitor is likely to produce an
normal sleep-onset pattern is commonly seen in a normal increase of REM sleep. In the latter case, this REM
person who acutely undergoes a phase shift, either as a rebound could result in abnormal SOREMPs in the
result of a work shift change or as a change resulting from absence of an organic sleep disorder, perhaps leading to
jet travel across a number of time zones. Studies of persons an incorrect diagnosis of narcolepsy.
sleeping in environments free of all cues to time have Acute presleep alcohol intake can produce an increase
shown that the timing of sleep onset and the length of in SWS and REM sleep suppression early in the night,
sleep occur as a function of circadian phase.32,33 Under which can be followed by REM sleep rebound in the
these conditions, sleep distribution with reference to the latter portion of the night as the alcohol is metabolized.
24 PARTI / Section 1 Normal Sleep and Its Variations

Wake Wake

S1 S1

S2 S2



24 1 2 3 4 5 6 23 24 1 2 3 4 5 6
Pretreatment CPAP
Recording date: 5/14/85 Total sleep time: 302.0 minutes = Lights out Recording date: 6/3/85 Total sleep time: 432.50 minutes = Lights out
Subjects age: 64 years REM percentage: 10.27% = End of night Subjects age: 64 years REM percentage: 38.61% = End of night
Subjects gender: M Subjects gender: M

Figure 2-10 These sleep histograms depict the sleep of a 64-year-old male patient with obstructive sleep apnea syndrome.
The let graph shows the sleep pattern before treatment. Note the absence of slow-wave sleep, the preponderance of stage 1
(S1), and the very frequent disruptions. The right graph shows the sleep pattern in this patient during the second night of
treatment with continuous positive airway pressure (CPAP). Note that sleep is much deeper (more SWS) and more consolidated,
and rapid eye movement (REM) sleep in particular is abnormally increased. The pretreatment REM percentage of sleep was only
10%, versus nearly 40% with treatment. (Data supplied by G. Nino-Murcia, Stanford University Sleep Disorders Center, Stanford,

Low doses of alcohol have minimal effects on sleep have indicated a relatively high prevalence of REM sleep
stages, but they can increase sleepiness late at night.36,37 onsets in young adults40 and in adolescents with early rise
Acute effects of marijuana (tetrahydrocannabinol times.41 In the latter, the REM sleep onsets on morning
[THC]) include minimal sleep disruption, characterized (8:30 am and 10:30 am) naps were related to a delayed
by a slight reduction of REM sleep. Chronic ingestion circadian phase as indicated by later onset of melatonin
of THC produces a long-term suppression of SWS.38 secretion.


Sleep disorders, as well as other nonsleep problems, have Sleep apnea syndromes may be associated with suppression
an impact on the structure and distribution of sleep. As of SWS or REM sleep secondary to the sleep-related
suggested before, these distinctions appear to be more breathing problem. Successful treatment of this sleep dis-
important in diagnosis and in the consideration of treat- order, as with nocturnal continuous positive airway pres-
ments than for any implications about general health or sure, can produce large rebounds of SWS or REM sleep
illness resulting from specific sleep stage alterations. A (Fig. 2-10).
number of common sleep-stage anomalies are commonly
associated with sleep disorders. S LEEP FRAGMENTATION
Fragmentation of sleep and increased frequency of arousals
N ARCOLEPSY occur in association with a number of sleep disorders as
Narcolepsy is characterized by an abnormally short delay well as with medical disorders involving physical pain or
to REM sleep, marked by SOREMPs. This abnormal discomfort. PLMS, sleep apnea syndromes, chronic fibro-
sleep-onset pattern occurs with some consistency, but not sitis, and so forth may be associated with tens to hundreds
exclusively; that is, NREM sleep onset can also occur. of arousals each night. Brief arousals are prominent in
Thus, the preferred diagnostic test consists of several such conditions as allergic rhinitis,42,43 juvenile rheumatoid
opportunities to fall asleep across a day (see Chapter 143). arthritis,44 and Parkinsons disease.45 In upper airway resis-
If REM sleep occurs abnormally on two or more such tance syndrome,46 EEG arousals are important markers
opportunities, narcolepsy is extremely probable. The because the respiratory signs of this syndrome are less
occurrence of this abnormal sleep pattern in narcolepsy is obvious than in frank obstructive sleep apnea syndrome,
thought to be responsible for the rather unusual symptoms and only subtle indicators may be available.47 In specific
of this disorder. In other words, dissociation of compo- situations, autonomic changes, such as transient changes
nents of REM sleep into the waking state results in of blood pressure,48 can signify arousals; Lofaso and col-
hypnagogic hallucinations, sleep paralysis, and, most dra- leagues49 indicated that autonomic changes are highly
matically, cataplexy. correlated with the extent of EEG arousals. Less well
Other conditions in which a short REM sleep latency studied is the possibility that sleep fragmentation may be
can occur include infancy, in which sleep-onset REM sleep associated with subcortical events not visible in the corti-
is normal; sleep reversal or jet lag; acute withdrawal from cal EEG signal. These disorders also often involve an
REM-suppressant compounds; chronic restriction or dis- increase in the absolute amount of and the proportion of
ruption of sleep; and endogenous depression.39 Reports stage 1 sleep.
CHAPTER 2 Normal Human Sleep: An Overview 25

15. Stickgold R, Hobson JA, Fosse R, et al. Sleep, learning and dreams:
Acknowledgments Off-line memory reprocessing. Science 2001;294:1052-1057.
16. Siegel J. The REM sleep-memory consolidation hypothesis. Science
The authors thank Joan Mancuso for preparing the figures. 2001;294:1058-1063.
17. Rechtschaffen A, Kales A, editors. A manual of standardized termi-
nology, techniques and scoring system for sleep stages of human
subjects. Los Angeles: UCLA Brain Information Service/Brain
Clinical Pearls Research Institute; 1968.
18. Czeisler CA, Zimmerman JC, Ronda JM, et al. Timing of REM sleep
The clinician should expect to see less slow-wave is coupled to the circadian rhythm of body temperature in man. Sleep
sleep (stages 3 and 4) in older persons, particularly 1980;2:329-346.
men. 19. Zulley J. Distribution of REM sleep in entrained 24 hour and free-
Clinicians or colleagues might find themselves running sleep-wake cycles. Sleep 1980;2:377-389.
denying mid-night communications (nighttime calls) 20. Weitzman ED, Czeisler CA, Zimmerman JC, et al. Timing of REM
because of memory deficits that occur for events and stages 3 + 4 sleep during temporal isolation in man. Sleep
proximal to sleep onset. This phenomenon might 1980;2:391-407.
21. Karacan I, Moore CA. Genetics and human sleep. Psychiatr Ann
also account for memory deficits in excessively
sleepy patients. 22. Busby K, Pivik RT. Failure of high intensity auditory stimuli to affect
Many medications (even if not prescribed for sleep) behavioral arousal in children during the first sleep cycle. Pediatr Res
can affect sleep stages, and their use or discontinu- 1983;17:802-805.
ation alters sleep. Thus, REM-suppressant medica- 23. Carskadon MA, Dement WC. Sleepiness in the normal adolescent.
tions, for example, can result in a rebound of REM In: Guilleminault C, editor. Sleep and its disorders in children. New
sleep when they are discontinued. York: Raven Press; 1987. p. 53-66.
Certain patients have sleep complaints (insomnia, 24. Feinberg I. Schizophrenia: caused by a fault in programmed synaptic
hypersomnia) that result from attempts to sleep or elimination during adolescence? J Psychiatr Res 1983; 17:319-334.
25. Prinz P. Sleep patterns in the healthy aged: relationship with intel-
be awake at times not in synchrony with their circa-
lectual function. J Gerontol 1977;32:179-186.
dian phase. 26. Prinz PN, Peskind ER, Vitaliano PP, et al. Changes in the sleep and
Patients who wake with events early in the night waking EEGs of nondemented and demented elderly subjects. J Am
might have a disorder affecting NREM sleep; patients Geriatr Soc 1982;30:86-93.
who wake with events late in the night may have a 27. Carskadon MA, Brown ED, Dement WC. Sleep fragmentation in
disorder affecting REM sleep. the elderly: relationship to daytime sleep tendency. Neurobiol Aging
When using sleep restriction to build sleep pres- 1982;3:321-327.
sure, treatment will be more effective if sleep is 28. Ancoli-Israel S, Kripke DF, Mason W, et al. Sleep apnea and noctur-
scheduled at the correct circadian phase. The problem nal myoclonus in a senior population. Sleep 1981;4:349-358.
29. Carskadon MA, Dement WC. Respiration during sleep in the aging
of napping in patients with insomnia is that naps
human. J Gerontol 1981;36:420-423.
diminish the homeostatic drive to sleep. 30. Williams RL, Karacan I, Hursch CJ. EEG of human sleep: clinical
applications. New York: John Wiley & Sons; 1974.
31. Agnew HW, Webb WB, Williams RL. The first-night effect: an
EEG study of sleep. Psychophysiology 1966;2:263-266.
REFERENCES 32. Czeisler CA, Weitzman ED, Moore-Ede MC, et al. Human sleep:
1. Dement W, Kleitman N. The relation of eye movements during Its duration and organization depend on its circadian phase. Science
sleep to dream activity: an objective method for the study of dream- 1980;210:1264-1267.
ing. J Exp Psychol 1957;53:339-346. 33. Zulley J, Wever R, Aschoff J. The dependence of onset and duration
2. Agnew HW, Webb WB. Measurement of sleep onset by EEG crite- of sleep on the circadian rhythm of rectal temperature. Pflugers Arch
ria. Am J EEG Technol 1972;12:127-134. 1981;391:314-318.
3. Davis H, Davis PA, Loomis AL, et al. Human brain potentials during 34. Strogatz SH. The mathematical structure of the human sleep-wake
the onset of sleep. J Neurophysiol 1938;1:24-38. cycle. New York: Springer-Verlag; 1986.
4. Carskadon MA, Dement WC. Effects of total sleep loss on sleep 35. Parmeggiani PL. Temperature regulation during sleep: a study in
tendency. Percept Mot Skills 1979;48:495-506. homeostasis. In Orem J, Barnes CD, editors. Physiology in sleep.
5. Guilleminault C, Phillips R, Dement WC. A syndrome of hypersom- New York: Academic Press; 1980. p. 98-143.
nia with automatic behavior. Electroencephalogr Clin Neurophysiol 36. Van Reen E, Jenni O, Carskadon MA. Effects of alcohol on sleep and
1975;38:403-413. the sleep electroencephalogram in healthy young women. Alcohol
6. Ogilvie RD, Wilkinson RT. The detection of sleep onset: behavioral Clin Exp Res 2006;30(6):974-981.
and physiological convergence. Psychophysiology 1984;21:510-520. 37. Rupp TL, Acebo C, Van Reen E, Carskadon MA. Effects of a moder-
7. Carskadon MA, Herz R. Minimal olfactory perception during sleep: ate evening dose of alcohol. I. Sleepiness. Alcohol Clin Exp Res
why odor alarms will not work for humans. Sleep 2004;27:402-405. 2007;31(8):1358-1364.
8. Williams HL, Hammack JT, Daly RL, et al. Responses to auditory 38. Freemon FR. The effect of chronically administered delta-9-tetra-
stimulation, sleep loss and the EEG stages of sleep. Electroencepha- hydrocannabinol upon the polygraphically monitored sleep of
logr Clin Neurophysiol 1964;16:269-279. normal volunteers. Drug Alcohol Depend 1982;10:345-353.
9. Oswald I, Taylor AM, Treisman M. Discriminative responses to 39. Kupfer DJ. REM latency: a psychobiologic marker for primary
stimulation during human sleep. Brain 1960;83:440-453. depressive disease. Biol Psychiatry 1976;11:159-174.
10. Williams HL, Morlock HC, Morlock JV. Instrumental behavior 40. Bishop C, Rosenthal L, Helmus T, et al. The frequency of multiple
during sleep. Psychophysiology 1966;2:208-216. sleep onset REM periods among subjects with no excessive daytime
11. Portas CM, Krakow K, Allen P, et al. Auditory processing across the sleepiness. Sleep 1996;19:727-730.
sleep-wake cycle: simultaneous EEG and fMRI monitoring in 41. Carskadon MA, Wolfson AR, Acebo C, et al. Adolescent sleep pat-
humans. Neuron 2000;28:991-999. terns, circadian timing, and sleepiness at a transition to early school
12. Foulkes D. The psychology of sleep. New York: Charles Scribners days. Sleep 1998;21:871-881.
Sons; 1966. 42. Lavie P, Gertner R, Zomer J, et al. Breathing disorders in sleep
13. Wyatt JK, Bootzin RR, Anthony J, et al. Does sleep onset produce associated with microarousals in patients with allergic rhinitis. Acta
retrograde amnesia? Sleep Res 1992;21:113. Otolaryngol 1981;92:529-533.
14. Maquet P. The role of sleep in learning and memory. Science 43. Craig TJ, Teets S, Lehman EB, et al. Nasal congestion secondary to
2001;294:1048-1052. allergic rhinitis as a cause of sleep disturbance and daytime fatigue
26 PARTI / Section 1 Normal Sleep and Its Variations

and the response to topical nasal corticosteroids. J Allergy Clin 1st ed. Westchester, Ill: American Academy of Sleep Medicine,
Immunol 1998;101:633-637. 2007.
44. Zamir G, Press J, Tal A, et al. Sleep fragmentation in children with 51. Moser D, Anderer P, Gruber G, et al. Sleep classification according
juvenile rheumatoid arthritis. J Rheumatol 1998;25:1191-1197. to AASM and Rechtschaffen & Kales: effects on sleep scoring param-
45. Stocchi F, Barbato L, Nordera G, et al. Sleep disorders in Parkinsons eters. Sleep 2009;32:139-149.
disease. J Neurol 1998;245(Suppl. 1):S15-S18. 52. Danker-Hopfe H, Anderer P, Zeitlhofer J, et al. Interrater reliability
46. Guilleminault C, Stoohs R, Clerk A, et al. From obstructive sleep for sleep scoring according to the Rechtschaffen & Kales and the new
apnea syndrome to upper airway resistance syndromeconsistency AASM standard. J Sleep Res 2009;18:74-84.
of daytime sleepiness. Sleep 1992;15(6 Suppl.):S13-S16. 53. Parrino L, Ferri R, Zucconi M, Fanfulla F. Commentary from the
47. Hosselet JJ, Norman RG, Ayappa I, et al. Detection of flow limitation Italian Association of Sleep Medicine on the AASM manual for the
with a nasal cannula/pressure transducer system. Am J Respir Crit scoring of sleep and associated events: for debate and discussion.
Care Med 1998;157:1461-1467. Sleep Med 2009;10:799-808.
48. Pitson DJ, Stradling JR. Autonomic markers of arousal during sleep 54. Grigg-Damberger MM. The AASM scoring manual: a critical
in patients undergoing investigation for obstructive sleep apnoea, appraisal. Curr Opin Pulm Med 2009;15:540-549.
their relationship to EEG arousals, respiratory events and subjective 55. Novelli L, Ferri R, Bruni O. Sleep classification according to AASM
sleepiness. J Sleep Res 1998;7:53-59. and Rechtschaffen and Kales: effects on sleep scoring parameters of
49. Lofaso F, Goldenberg F, Dortho MP, et al. Arterial blood pressure children and adolescents. J Sleep Res 2010;19:238-247.
response to transient arousals from NREM sleep in nonapneic 56. Miano S, Paolino MC, Castaldo R, Villa MP. Visual scoring of sleep:
snorers with sleep fragmentation. Chest 1998;113:985-991. A comparison between the Rechtschaffen and Kales criteria and the
50. Iber C, Ancoli-Israel S, Quan SF, for the American Academy of American Academy of Sleep Medicine criteria in a pediatric popula-
Sleep Medicine. The AASM manual for the scoring of sleep and tion with obstructive sleep apnea syndrome. Clin Neurophysiol
associated events: rules, terminology, and technical specifications. 2010;121:39-42.
Normal Aging
Donald L. Bliwise
Abstract demonstrate age-dependence and contribute to sleep prob-
Numerous factors challenge the integrity of sleep in the older lems. Descriptive data continue to accrue to suggest that sleep
human. Intrinsic lightening and fragmentation of sleep, disturbance in late life might carry its own morbidity and
reflecting changes in circadian and, even more so, homeo- should not be dismissed by the sleep medicine specialist. New
static processes characterize aging, as do numerous medical data suggest that the breakdown of sleep in the aged organ-
and psychiatric comorbidities. Specific disorders such as ism might reflect physiologic age and reflect alterations in
sleep-disordered breathing and restless legs syndrome can function present at the genomic level.

As the populations of industrialized societies age, knowl- lence of dementing illnesses is high in late life, determina-
edge of defining how sleep is affected by age will assume tion of the number of normal elderly persons who may
greater importance. Within the United States, the average be in incipient stages of dementia has seldom been
current life expectancy of 77 years means that 80% of resi- addressed. Additionally, recognition of mental impair-
dents now live to be at least 65; the fastest growing segment ments in the more limited domains of memory, executive
of the population is those who are 85 years and older. function, language, attention, and visuospatial ability char-
These huge numbers force the sleep medicine specialist to acterized as of lesser severity has led to the use of an
confront the definition of what is normal. Researchers intermediate diagnostic category termed mild cognitive
often use the term to connote a variety of meanings. In impairment (MCI).2 Few sleep studies of normal aging rely
sleep medicine, confusion often occurs because the term is on extensive diagnostic work to eliminate persons in the
used descriptively, to indicate representativeness, as well earliest stages of mental impairment, and polysomno-
as clinically, to indicate absence of disease. graphic studies in well-defined MCI patients have yet to
Aging is also subject to semantic confusion. Chronologic occur.
age has been shown repeatedly only to approximate physi- The point here is not to dismiss all that is known about
ologic (biological) age. The decline in slow-wave sleep, for sleep patterns in normal aging as inadequate but rather to
example, can occur at a chronological age (at least in men) point out the complexities of defining normal aging.
far earlier than most age-related declines in other biologi- Normal aging can never be defined without some arbitrary
cal functions. Some researchers in gerontology have noted criteria. Throughout this chapter I will refer to aging
that distance from death may be a far better approximation across several species, encompassing both what in humans
of the aging process, but too few longitudinal sleep studies may be considered middle-aged (approximately 40 to 65
in humans exist to yield these types of findings. However, years) and elderly (older than 65 years). We recognize
studies of invertebrates and have shed new light on rela- fully the otherwise arbitrary nature of these verbal and
tionships between physiologic age and sleep that can affect numeric descriptors of processes that are most assuredly
the functional significance of age-dependent changes (see gradual, are continuous, and vary widely across individuals.
Basic Science Considerations, later). It is also important to recognize that the age-dependent
In addition to the issue of physiologic age, subjective alterations in sleep may simply be secondary manifesta-
age must be considered. Because the practice of sleep dis- tions of senescence.
orders medicine in geriatrics relies heavily on the increased
self-reports of sleep disturbance seen in aging, subjective
appraisal of the older persons symptoms must be consid- SLEEP ARCHITECTURE
ered. Whether an aged person views 75% sleep efficiency Although age-dependent alterations in sleep architecture
as insomnia or merely accepts this as a normal part of have been described for many years,3 only recently have
aging may depend largely on that persons perspective on attempts been made to summarize this large body of cross-
growing old and what that means to him or her. It has sectional data using meta-analytic techniques.4,5 Results
been reported that older people are more likely to perceive from the first of these analyses4 indicated that although
themselves as having sleep problems if they have difficulty sleep efficiency showed clear age-dependent declines up
falling asleep rather than staying asleep, even though the to and beyond age 90 years, the vast majority of age-
latter continues to be a generally more commonly endorsed dependent changes in sleep architecture occurred before
symptom.(See reference 1 for review) In addition, some have sug- age of 60 years, with few changes in slow-wave sleep (SWS,
gested that self-reports of sleep measured by polysomnog- now referred to as N3 sleep in the revised American
raphy (PSG) are inherently less accurate and valid in older Academy of Sleep Medicine [AASM] nomenclature6; see
relative to younger subjects, although evidence for such later), rapid eye movement (REM) sleep, and stage 1 per-
age differences in other studies is decidedly mixed and centage (N1) occurring after that.4 Some variables (total
varies according to the variables under consideration or sleep time, REM) appeared best characterized as linear
the subjects sex. decline, whereas others (SWS, wake after sleep onset) fol-
Finally, normal aging must be viewed in counterpoint lowed a more exponential course. Sleep latency showed no
to pathologic aging (see Chapter 136). Although the preva- clear age effect after age 60 years, although it increased up

28 PARTI / Section 1 Normal Sleep and Its Variations

Table 3-1 Sleep Architecture as a Function of Age

Percentage of Time Spent in StageMean (95% CI)



37-54 5.8 4.6 61.4 58.5 11.2 14.2 19.5 20.9

(5.2-6.5) (4.1-5.3) (60.0-62.8) (57.1-60.0) (9.9-12.6) (12.7-15.9) (18.8-20.2) (20.0-21.8)
55-60 6.3 5.0 64.5 56.2 8.2 17.0 19.1 20.2
(5.6-7.0) (4.4-5.7) (63.2-65.9) (54.5-57.8) (7.1-9.5) (15.2-18.9) (18.4-19.8) (19.3-21.1)
61-70 7.1 5.0 65.2 57.3 6.7 16.7 18.4 19.3
(6.4-7.9) (4.4-5.7) (63.9-66.5) (55.7-58.9) (5.7-7.7) (14.8-18.6) (17.8-19.1) (18.4-20.2)
>70 7.6 4.9 66.5 57.1 5.5 17.2 17.8 18.8
(6.8-8.5) (4.3-5.6) (65.1-67.8) (55.6-58.7) (4.5-6.5) (15.5-19.1) (17.1-18.5) (18.0-19.6)
CI, confidence interval; REM, rapid eye movement.
From Redline S, Kirchner HL, Quan SF, etal. The effects of age, sex, ethnicity, and sleep-disordered breathing on sleep architecture. Arch
Intern Med 2004;164:406-418.

to that point. A second meta-analysis focused only on divergence when comparing women and men. Most
REM percentage and noted a cubic trend, with REM notable in this regard is percentage of time spent in sleep
apparently increasing after age 75 years and then demon- stages 3 plus 4, which shows enormous gender differences
strating an even steeper drop after age 90 years.5 The at every age and, in fact, shows no appreciable decline with
meaning of the latter data is unclear and raises many ques- aging in women, relative to men.
tions as to the extent of the precision of chronologic age Men demonstrate a marked cross-sectional decline with
to capture biological processes in these upper age ranges. aging, as well as huge individual differences in every age
Published population-based longitudinal data on sleep group. In fact, the extent of these individual differences is
architecture would assist in addressing many of these emphasized by the fact that even within men as a group,
uncertainties. coefficients of variation (ratio of variance to mean) in per-
Although meta-analyses can provide cumulative infor- centage of time spent in sleep stages 3 plus 4 far exceeded
mation on age-dependent values across many laboratories, those for all other sleep variables in both men and women.
enormous variability in parameter values exists across Although gender differences in SWS have been noted pre-
studies,4 and much of the sleep architecture was not scored viously (see reference 3 for review), the fact that the age-
blindly to the patients chronologic age or sex. This might dependent decline may be confined only to men suggests
limit the value of meta-analytic approaches for extrapola- a more limited utility of this often-characterized aging
tion of readily usable, age-dependent laboratory norms. By biomarker for women.
contrast, the systematically collected, rigorously acquired In contrast to the results of the SHHS, these gender
data derived from the centralized scoring center for the differences in SWS were not confirmed meta-analytically.4
Sleep Heart Health Study (SHHS), although subject to At least one study has proposed that gender differences in
survivor effects and based on single-night data derived delta activity are more likely to be a function of overall
from composite cohorts, offer detailed appreciation of how lower EEG amplitude in men relative to women.9 When
comorbidities, demographics, and sleep-disordered breath- corrected for overall amplitude, the decreased growth
ing (SDB) can affect observed sleep architecture values hormone secretion seen in postmenopausal women was
employing traditional Rechtschaffen and Kales rules.7 accompanied by lower delta amplitude than in comparably
Some have viewed the SHHS sleep architecture data as aged men.10 A decline in the amplitude and the incidence
broadly representative of the elderly population generally of the evoked K-complex over the age range of 19 to 78
because persons with a wide variety of medical conditions years has been reported in both women and men, suggest-
were not excluded.8 ing that similar deficits in delta synchronization processes
operate equally in both sexes.11 Gender did not play a
Percentage of Time Spent in Each significant role when assessed as the homeostatic response
Sleep Stage of nighttime delta power to daytime napping in either
Table 3-1 provides sleep architecture values for 2685 young or elderly subjects.12
SHHS participants ages 37 to 92 years, excluding persons Percentage of time spent in sleep stage 1 also showed
who use psychotropic medications and who have high similar gender-related effects in SHHS, and age-
alcohol intake, restless legs syndrome symptoms and sys- dependent increases in this sleep stage, usually considered
temic pain conditions. About a third of these participants to represent a feature of fragmented, transitional sleep,
were hypertensive and about 10% had histories of cardio- were confined only to men. By contrast, percentage of time
vascular disease or chronic pulmonary disease. Results spent in REM sleep showed a modest decline with age,
clearly show that although age effects are apparent in some but the effect was detected in both men and women. REM
measures, gender occupied a far more dramatic role in percentages of 18% to 20% in 75- to 85-year-olds were
sleep architecture, in some cases showing considerable derived from curve smoothing in another meta-analysis
CHAPTER 3 Normal Aging 29

focused on only REM sleep measures in normal aging,5

Table 3-2 Brief Arousal Index in Elderly Subjects as
which were slightly lower than, but essentially similar to, a Function of Sleep-Disordered Breathing
SHHS data (see Table 3-1). In SHHS, sleep efficiency also
declined with age, with mean values of 85.7 (standard Arousal Index: Brief Arousals per
deviation [SD] = 8.3) in the 37- to 54-year-old group, Hour of Sleep (SD)
83.3 (SD = 8.9) in the 55- to 60-year-old group, 80.6
(SD = 11.7) in the 61- to 70-year-old group, and 79.2
(SD = 10.1), in the over-70-year-old group, but without 5 16.7 (7.7) 14.7 (7.1)
differential effects of gender, findings corroborated meta- >5 to 15 20.5 (8.7) 17.9 (7.8)
analytically in persons older than 60 years.4 However, the >15 to 30 25.2 (10.3) 23.2 (10.4)
declines in percentage of time spent in REM sleep and the >30* 39.4 (14.7) 29.7 (13.6)
(male-specific) increases in percentage of time spent in
sleep stage 1 seen in SHHS were not confirmed meta- *Estimated weighted values.
RDI, respiratory disturbance index (apneas plus hypopneas per
analytically in persons older than 60 years.4 The density
hour of sleep), a measure of sleep-disordered breathing; SD,
of eye movements in REM is reduced with aging,13 standard deviation.
but lack of standardization across laboratories precludes From Redline S, Kirchner HL, Quan SF, etal. The effects of age, sex,
examination of this aspect of REM using meta-analytic ethnicity, and sleep-disordered breathing on sleep architecture.
techniques. Arch Intern Med 2004;164:406-418.

Arousals during Sleep

Brief arousals during sleep, representing one component
of the microarchitecture of sleep, continues to attract con- Comorbidities
siderable interest as a metric, with particular relevance for Insofar as comorbidities are concerned, SHHS sleep archi-
the aged population. When examined in the laboratory, tecture data showed substantial convergence with meta-
healthy older persons wake up from sleep more frequently analytically derived data. In SHHS, selected medical
than younger persons do, regardless of circadian phase, but comorbidities (a positive history of cardiovascular disease,
they have no greater difficulty falling back to sleep.14 hypertension, and stroke) were associated with disturbed
Failure to maintain continuous sleep has, as its basic sleep architecture, as was smoking, although, curiously,
science counterpart, short bout lengths, a feature highly these results were not seen with chronic obstructive pul-
characteristic of sleep in many aged mammalian species.3 monary disease. Consistent with results suggesting that
In elderly humans without SDB, arousal indices from 18 reduced sleep amounts or quality might predispose to
to 27 events per hour have been reported.15 Among the the metabolic syndrome in old age, diabetic patients had
predominantly elderly subjects (mean age, 61 years) in smaller percentages of time in stages 3 plus 4 sleep, lower
SHHS, the mean (SD) Arousal Index showed significant sleep efficiencies, and higher numbers of brief arousals
but relatively small increases with age: 16.0 (8.2) for 37- to and percentage of time spent in sleep stage 1. In most
54-year-olds, 18.4 (10.0) for 55- to 61-year-olds, 20.3 cases, however, these effects appeared to less salient (i.e.,
(10.5) for 62- to 70-year-olds, and 21.0 (11.6) for subjects predicted less variance) for sleep architecture than demo-
older than 70 years.7 Values approximating these have been graphic variables such as gender, age (to a lesser extent),
reported16 in another group of subjects without sleep apnea and, in some cases, ethnicity,7 except for the arousal index,
or periodic leg movements, thus further corroborating where RDI was by far the single most powerful predictor.
these SHHS values. Less-disease-specific moderator effects from meta-analytic
Other phasic events of non-REM (NREM) sleep, such approaches also suggested that across the entire life span,
as K-complex and spindle density, also decrease with age effects were reduced substantially when persons with
age.17 Spindle density is thought to reflect, at least par- medical and psychiatric conditions were included.4 The
tially, the corticothalamic functional integrity of gamma- inclusion of persons with sleep apnea showed some evi-
aminobutyric acidergic (GABAergic) systems. dence of reducing the effects of age in sleep efficiency,
Although, like other metrics of impaired sleep quality, wake after sleep onset, and SWS when considered across
brief arousals show a male predominance (also seen meta- the entire adult life span,4 data that are compatible
analytically using Wake after Sleep Onset4), the influences with SHHS.
of age and gender are not as pronounced as the effects of
breathing events (Table 3-2). In fact, when accounting for Slow-Wave Sleep
the presence of brief arousals in the elderly, the Respira- The gender differences in SWS reported by SHHS not-
tory Disturbance Index (RDI) predicts 10-fold more vari- withstanding, several aspects of these data must be viewed
ance than age and 5-fold more variance than gender. in the context of prior literature on age-dependent changes
Higher levels of RDI were also associated with slightly in architecture. When analyzed with period-amplitude
lower percentage of time spent in REM sleep in both men analyses, the major change in SWS ascribed to aging has
and women and with lower percentages of time spent in been a decline in delta wave amplitude rather than wave-
sleep stages 3 plus 4 in men. The latter result is consistent length (Fig. 3-1).(See reference 3 for review) The decrease in delta
with the hypothesis that at least some SDB in both elderly amplitude simply may be a more readily identifiable visual
men and women might reflect ventilatory control instabil- change of the sleep EEG, which is present at frequencies
ity and that SWS may be protective (see the later section up to about 10Hz, though it is difficult to see above
on SDB). this.18 When scored visually using central derivations and
30 PARTI / Section 1 Normal Sleep and Its Variations

therapeutic importance of such a revisionary approach

remain obscure. Much the same effect could be obtained
by adopting alternative scoring thresholds of less than
75V for defining delta wave activity. Such proposals
were put forth in the 1990s(See reference 3 for review) but have not
led to enhanced understanding of the age-dependent
changes in SWS. Eventually, digitized indices of delta
Delta activity of a 15-year-old male activity (e.g., fast Fourier transform, zero-crossing, or
hybrid techniques) might come to replace such conven-
tional measures; however, considerable controversy
regarding filtering, sampling rates, and data-storage for-
matting leaves formal adoption of such approaches dubious
Well-preserved delta activity, 65-year-old male
for routine clinical purposes at this time,20 though such
efforts at signal processing are yielding important new
clues regarding the significance of sleep-related delta activ-
ity for aging.
Typical delta activity of older men (age 64) 50 V
1 sec
Slow-wave activity during sleep is thought to represent
synaptic downscaling, which is viewed as critical for
Figure 3-1 Age differences in delta activity. The top tracing forming memories.21 Broadly viewed, given at least some
shows typically abundant high-amplitude delta in an adoles- data suggesting decreased SWS with age, such findings
cent. The middle tracing shows particularly well-preserved might fit with mild impairments in cognition that charac-
delta in an older man. Note the marked decrease in amplitude
relative to the adolescent. The bottom tracing is a more typical
terize normal aging. Extremely low frequency (<1.0Hz)
example of delta activity in an older man. Note the number of slow wave activity in NREM sleep has been thought to
waves failing to meet the 75-V amplitude criterion. (From hold particular significance as a more immediate reflection
Zepelin H. Normal age related change in sleep. In: Chase MH, of cellular processes than conventionally defined delta
Weitzman ED, editors. Sleep disorders: basic and clinical activity.22 In dementia, the integrity of the normal auditory
research. New York: Spectrum; 1983. pp. 431-445.)
evoked response (K-complex) may be impaired,23 whereas
in normal aging, there is some suggestion that spectral
power at frequencies below 0.7Hz might demonstrate
employing a 75-V threshold, typical figures for the fewer age differences than at between 0.7 and 3.0Hz.24
amount stages 3 plus 4 sleep in the elderly have often been Other attempts to examine NREM sleep in old age using
considered to fall in the 5% to 10% range. Thus, the nonlinear, dynamic EEG approaches yield broadly com-
figures reported by SHHS, particularly for women, are patible findings. The sleep EEG of healthy older adults in
somewhat higher than these conventionally accepted the first NREM period may resemble patterns of compen-
figures. Whether these values represent a more precise satory activation similar to those seen in young adults
rendering of delta activity within sleep, perhaps engen- during sleep subsequent to sleep deprivation.25 Whether
dered by the visual analyses of EEG waveforms on digital such phenomenological parallels of altered functional con-
display or the simultaneous availability of precise calibra- nectivity in sleep deprivation and aging hold prognostic or
tion of the 75 microvolt criterion for delta waves stipulated practical significance at the individual case level is unknown
by the Rechtschaffen and Kales guidelines, is unclear. but certainly plausible. Nonlinearity of the NREM sleep
Nonetheless, the strictly controlled and exquisitely refined EEG increases with normal aging25 and sample entropy, a
visual analyses conducted by SHHS are likely to represent novel measure reflecting dynamic probabilities of state,26
a standard of polysomnographic technology aspired to by also changes with age.
the field of sleep medicine, thus arguing that these newly
published metrics may well replace existing data and sup-
plant our current understanding of how sleep architecture CIRCADIAN RHYTHMS IN AGING
measures should be benchmarked. In humans, most descriptive data collected under entrained
Given the new AASM guidelines for sleep stage scoring,6 conditions have suggested that the amplitude of the sleep-
much of the foregoing normative data on sleep architec- wake rhythm, body temperature (Fig. 3-2),27 and some hor-
ture may have limited relevance for laboratories that elect mones decrease with aging. Sex differences in such
to adopt such changes. For example, slow-wave activity phenomena have also been reported (see reference 28 for
has higher amplitude when recorded from frontal deriva- review). However, in exceptionally healthy older adults
tions relative to central derivations. This would be expected such differences might not always appear,29 and a study of
to result in increased levels of visually scored slow-wave centenarians indicated relatively robust neuroendocrine
(i.e., N3) sleep. Indeed a recent study comparing record- profiles.30 A phase advance of aging has often been ascribed
ings scored with both revised AASM and traditional to the timing of sleep patterns in older adults. Many sleep
Rechtschaffen and Kales criteria have shown a number of medicine specialists use this designation as an abbreviation
significant differences in resulting measures.19 Predictably, for indicating that older persons go to bed earlier in the
particularly in older persons, the revised scoring system evening and wake up earlier in the morning than younger
resulted in higher percentages in N3 sleep. Beyond creat- persons, findings corroborated in age differences in the
ing the need to establish new normative data, the mecha- timing of bedtimes and wakeup times in dozens of cross-
nistic and functional significance or the diagnostic and sectional surveys(See reference 3 for review) and even longitudinally.31
CHAPTER 3 Normal Aging 31

nounced late in the sleep period (corresponding to early

37.0 morning awakening) and least pronounced just after sleep
onset (corresponding to early evening sleepiness). This
Mean oral body temperature (C)

study also analyzed sleep structure immediately before

36.8 awakenings from consolidated periods of sleep. With cir-
cadian phase controlled, older subjects were far less likely
36.6 to awaken from stage 1 and far more likely to awaken from
stage 2 than were young subjects, suggesting that awaken-
ings in the older subjects probably represented abrupt
transitions from NREM sleep rather than gradual lighten-
ing of sleep.35 These findings can be interpreted as an
36.2 indication of a reduced homeostatic pressure for continu-
ous bouts of sleep independent of circadian phase in the
36.0 sleep of older persons.
09:30 13:30 17:30 21:30 01:30 05:30 09:30 A well-described feature of the circadian system in the
aged organism is the relative impairment in the ability to
phase shift. This may in part reflect loss of rhythmic func-
Figure 3-2 Oral temperatures in young (open circles) and old tion within the suprachiasmatic nucleus.36 In humans, the
(dark circles) subjects, showing apparent decreased amplitude sleep-wake system appears particularly vulnerable to such
and earlier phase in body temperature cycle as a function of changes in phase shifting37 and might account for the some
aging. Data were obtained under entrained conditions. (From
Richardson GS, Carskadon MA, Orav EJ. Circadian variation of
of the apparent self-selection out of shift work typically
sleep tendency in elderly and young adult subjects. Sleep seen in older people (see reference 38 for review). In
1982;5[Suppl. 2]:S82-S94.) rodents, such impairments in phase shifting have also been
described.39 Although both photic and nonphotic influ-
ences on impaired phase-shifting ability have been
However, the reader should be aware that fundamental described in aging,(See reference 38 for review) the ability to phase
changes in phase relationships in human circadian rhythms shift and entrain to light in old age might be expected to
might not always substantiate the notion of such an be particularly impaired because of challenges to the visual
advance. For example, in the constant routine protocol, system that occur as a part of human aging (e.g., cataracts,
older subjects typical earlier bedtimes and wakeup times macular degeneration). Perhaps as a consequence, in epi-
relative to younger subjects actually were more phase demiologic studies, elderly persons with visual impair-
delayed, rather than phase advanced, relative to their peaks ments were 30% to 60% more likely to have impaired
in melatonin.32 The implication of this finding is that the nighttime sleep relative to visually unimpaired elderly sub-
earlier bedtimes and wakeup times may be due to homeo- jects.40 In the laboratory setting, however, disagreement
static factors. Studies using 28-hour forced desynchrony, exists as to whether, among persons without visual impair-
an experimental protocol that requires subjects to sleep ment, responsiveness of the circadian system, as indexed
on a 2:1 wake-to-sleep ratio outside the limits of entrain- by melatonin rhythms, is decreased41 or unchanged42 by
ment of the circadian system, have shown absence of age light exposure. Finally, another study examined the ability
differences in estimates of tau, the endogenous period of a nonphotic stimulus (evening exercise) to phase shift
length of the human core temperature rhythm.33 Again, melatonin onset in young and elderly humans and found
these results implicate factors other than circadian ones comparable phase delays subsequent to exercise in both
that may be responsible for the earlier bedtimes of older groups, arguing for the importance of entraining factors
humans. Further evidence of interaction between circadian other than bright light to affect rhythms in the elderly.43
and homeostatic factors comes from a study of age dif- Invertebrate models in studies of sleep and aging have
ferences in the melatonin rhythm under the constant been broadly confirmatory for many of the changes
routine protocol, during which younger subjects showed described in mammals, such as the decline in nocturnal
elevated melatonin levels during the higher homeostatic sleep durations.44 Several studies in the fruit fly (Drosophila
pressure induced by this procedure, whereas older subjects melanogaster)45 and in the honeybee (Apis mellifera)46 have
did not.34 indicated a dispersal of sleep around the 24-hour day
The interaction between homeostatic and circadian occurring with aging. Perhaps more noteworthy is that
influences in older humans has been described elegantly in these studies have also implied that total amount of sleep
the forced desynchrony protocol, which has shown that per 24 hours does not merely redistribute (via shorter bout
throughout the assigned (9.33-hour) sleep period, elderly length and increased number of bouts) but might also
subjects awakened more frequently than younger subjects, increase in quantity with advanced physiologic age. These
regardless of circadian phase.35 The duration of awaken- changes appeared particularly pronounced in male fruit
ings was virtually identical in young and old subjects, but flies45 and female honeybees46 near the end of life. Because
the largest differences in the frequency of awakenings an avowed function of the mammalian master clock,
between young and older subjects were detected early, located in the suprachiasmatic nucleus, is to provide clock-
rather than late, in the sleep period. These results appear dependent alerting, these data are consistent with early
incompatible with the broadly defined phase-advanced studies that demonstrated such a role for the suprachias-
hypothesis of sleep in the elderly, in which differences in matic nucleus in primates using lesion models.47 Although
sleep consolidation would be predicted to be most pro- studies involving neurodegenerative disease in aged
32 PARTI / Section 1 Normal Sleep and Its Variations

humans have shown increased sleep per 24 hours consis- elderly. Psychiatric conditions have always been consid-
tent with level of dementia (see Chapter 136), the inverte- ered to play a major role in the insomnia of old age. A
brate data imply that increased rather than decreased study of more than 3000 older men reported that depres-
sleep-durations hours would be likely to occur in aged sive symptoms were associated with difficulties in falling
humans normatively on a population-wide basis. In this asleep, but not lower sleep efficiency or total sleep time.57
regard, it is noteworthy that the largest single compilation Among women, although the caregiving role per se
of human sleep durations ever published (more than 1 was not associated with poor sleep, for women who
million subjects), shows that reported sleep durations were depressed, caregiving was associated with reported
increase, rather than decrease, with age.48 Taken together, sleep problems.58 Regardless of caregiving status, anxiety
these data broadly suggest that increased sleep durations appeared to play a larger role in predicting poor sleep
in aged humans might fundamentally represent biological quality in these older women, even relative to depressed
rather than sociocultural factors. mood per se.59 However, poor sleep, left untreated, appears
Among lower vertebrates, aged zebrafish (Danio rerio) to be a risk factor for incident depression in the elderly.60
have been shown to demonstrate reduced rest-activity Limitations in mobility, visual impairment, lack of
rhythm and shorter nocturnal sleep durations across the regular exercise, alcohol use, and smoking all contribute to
lifespan (from 1 to 4 years), and these were associated with declining sleep quality in older persons, as do chronic pain
decreased brain melatonin during the dark period.49 Mela- conditions, such as arthritis, hip fracture, fibromyalgia,
tonin administration was shown to partially restore sleep headache, and back pain; cardiovascular diseases such as
durations and simple measures of learning in this species. hypertension, myocardial infarction, stroke, congestive
Zebrafish aging was also shown to be associated with gene heart failure, and angina; respiratory conditions such as
expression in several clock genes (Bmal1, PER 1),49 although asthma and bronchitis; and other systemic diseases such as
the translation of this finding back to older humans might diabetes, gastroesophageal reflux, and duodenal ulcer (see
not be clear.50 Additionally, in rodents, age-effects in the reference 61 for review). When persons with such comor-
expression of several clock genes may occur equally in both bidities are eliminated from consideration, the resulting
wild-type and mutant lines.51 insomnia prevalence in elderly populations may be only
1% to 3%,62 reiterating the decreased significance of
CAUSES AND CONSEQUENCES OF chronologic age per se in predicting poor sleep. In women,
POOR SLEEP IN OLD AGE some evidence suggests that menopause may also be asso-
ciated with declining sleep quality (see Chapter 140). Addi-
Causes tionally, in both women and men, the otherwise mundane
The prevalence of insomnia in the aged varies across occurrence of nocturia (nightly awakenings to void) appears
studies, but figures between approximately 20% and 40% to be associated with poor sleep,63 even when other factors
are typically reported. In one of the largest surveys of an such as pain and medical comorbidity are taken into
American population (the Established Populations for account. In fact, nocturia may be the single most common
Epidemiologic Studies of the Elderly [EPESE], with more factor associated with poor sleep in the elderly.63
than 9000 participants), 29% of the over-65 population The role of SDB as a cause for insomnia in older adults
had difficulty maintaining sleep.52 Relative to sleep main- has been debated over the years. Although a carefully per-
tenance, sleep latency is less likely to be problematic for formed case-control study suggested that SDB frequency
the elderly population, with prevalence figures on the was no higher (and might even have been lower) in older
order of 10% to 19% typically seen, although one study53 persons with poor sleep than in those who slept well, there
reported a relatively high prevalence of sleep latency prob- was evidence that for persons who demonstrated both SDB
lems (36.7%) in a largely rural aged population. In nearly and poor-quality sleep, the impact on day-to-day function
all studies, elderly women have a greater probability of was quite substantial,64 even exceeding those in persons
sleep complaints and sedative-hypnotic use than elderly with comparable levels of SDB but no insomnia.
men do (see reference 1 for review). Conflicting data exist Several longitudinal studies examining the incidence
on racial differences in sleep complaints,54 though health (development of new cases) of insomnia over periods of up
disparities in sleep durations did not appear to depend on to 10 years have been reported. The single best predictor
age.55 Figures for regular use of sedative-hypnotics in of insomnia continuing longer than 10 years was insomnia
elderly populations range from as low as 5% per year to as at a previous time, although cardiovascular and pulmonary
high as 16% over 4.5 years, 29% over 8 years, 34% over comorbidities conferred risk as well in the over-65 popula-
1 year and 62% over 3 years.(See reference 1 for review) Although tion.65 Reported remissions were less likely in older sub-
subjective reports of poor sleep invariably increased with jects than in younger ones.66 The EPESE data indicated a
age and numerous changes in sleep architecture have been yearly incidence of insomnia complaints in the aged popu-
frequently noted (see earlier), the relationship between lation of about 5%, with a spontaneous remission rate of
these two domains are significant, though relatively weak, about 50% over 3 years.67 In these data, incident insomnia
with women showing slightly higher levels of association was related to heart disease, stroke, hip fracture, and new-
relative to men.56 onset depression. Spontaneous remission of insomnia was
Disturbed sleep is thought to both contribute to and related to the resolution of depression, physical illness, and
reflect allostatic load of illness in old age and might reflect physical disability affecting activities of daily living,67
physiologic age of the organism. When other causes of whereas in the Cardiovascular Health Study, persistence
poor sleep are taken into account, chronological age might of insomnia was associated with unresolved depression.68
explain little of the observed higher prevalence in the Important from the standpoint of prevention, another
CHAPTER 3 Normal Aging 33

study reported that higher levels of physical activity were women were noted to have scored lower on the Mini
protective for incident insomnia over an 8-year period.69 Mental State Examination, a general examination of mental
function and orientation, and take more time to complete
Potential Consequences the Trail Making Test, part B, a paper-and-pencil test of
A major question regarding the frequent complaints of psychomotor speed, as sleep efficiencies decreased and
poor sleep among the elderly involves whether these have sleep latencies increased.83 In another report from that
an impact on the health of older persons. If the poor sleep same population, cognitive decline over 15 years was more
of old age, although annoying and distressing for many, likely to be associated with sleep efficiency of less than
represents primarily a quality-of-life issue, albeit one mod- 70% at follow-up.84 In the Nurses Health Study cohort,
ifiable by medical or behavioral interventions, it might cast poorer performances on a more-comprehensive battery
a different perspective on this problem, than for a medical of cognition, including immediate and delayed verbal
disorder such as SDB, where negative outcomes may be memory, category fluency, and attention, were associated
better defined and quantified. There is no question that with sleep durations of less than 5 hours.85 In an older
almost universally, poor nocturnal sleep is distressing and French population, the likelihood of cognitive problems
related to lower quality of life of many older persons. This derived on a self-reported telephone-derived question-
has been demonstrated in elderly populations in the United naire was increased by 50% for those reporting 4.5 to 6.0
States,70 in Canada,71 across Europe72 and in Asia.73 Inter- hours of sleep, and it more than doubled in respondents
estingly, some have questioned the extent to which such reporting less than 4.5 hours of sleep.86
observations may be tempered by older persons qualitative Relative to poor-quality sleep, at least some data suggest
perceptions of sleep satisfaction74 or sleep insufficiency,75 adverse outcomes associated with short sleep durations in
which might not change with aging. older populations. Gangwisch and colleagues87 reported
Most work relating poor sleep quality or duration to that sleep durations of less than 5 hours were associated
putative adverse outcomes has been observational, which, with higher rates of all-cause mortality in subjects 60 years
although often provocative, lacks the definitive element of and older, a finding that was not present in persons ages
proof of causation that is afforded by randomized clinical 32 to 59 years. Sleep durations of less than 5 hours per
trials. Unfortunately, with few exceptions, most pharma- night in even older populations (67 to 99 years old) were
cologic and nonpharmacologic randomized clinical trials associated with obesity as well,88 a finding otherwise well
attempting to treat poor sleep in elderly persons seldom acknowledged in populations younger than 65 years in
rely on outcomes other than conventional subjective and women in one study89 and in men and women in another.90
polysomnographic measures of nocturnal sleep per se. Hypertension has also been associated with short sleep
Rare exceptions to the latter have been several insomnia durations in elderly persons,91 but other studies of older
treatment studies among older adults that have demon- populations indicated that neither short sleep durations92
strated increases in selected quality-of-life measures such nor complaints of poor sleep93 were associated with this
as the SF-3676 or decreased daytime napping.77 Interven- morbidity. Diabetes and impaired glycemic control were
tional data relevant for other medical outcomes in old associated with sleep durations of less than 6 hours (but
age (e.g., hypertension, insulin resistance) have yet to be not insomnia complaints) across the age range 53 to 93
published. years94 and were independent of age across an even broader
Among observational studies, the association between age range.95 As mentioned earlier, these are all observa-
nocturia and insomnia has led to speculation that the more tional studies, which, although impressive by size of the
likely a person is to rise from bed during the night to use samples studied and control over confounding variables,
the bathroom, the more likely the person is to fall.63 Con- did not manipulate sleep quality or sleep duration to dem-
siderable evidence for this association exists at the popula- onstrate improvement in any of these putative adverse
tion level, where studies have shown associations between outcomes in older persons.
insomnia and falls.78 Sleep durations of less than 5 hours
were associated with an increased risk for falls of more than RESTLESS LEGS SYNDROME
50%.79 Risk for increased falls with short duration of sleep
or poor quality of sleep (or both) is also consistent with AND PERIODIC LIMB
data suggesting that insomnia is associated with impaired MOVEMENTS IN SLEEP
physical function. For example, lower sleep efficiencies One specific cause of insomnia in the elderly is restless legs
were associated with lower grip strength and slower syndrome (RLS) (see Chapter 90). This condition, charac-
walking speed in a population of elderly men.80 Short sleep terized by an urge to move the legs, which is usually
durations were associated with a slightly different set of accompanied by sensations of discomfort, aggravation of
markers of physical impairment in elderly women, primar- symptoms by rest and temporary relief of symptoms by
ily consisting of chair-to-stand speed.81 Although the movement, and worsening during the evening or nocturnal
increased risk for falls in the older populations has been hours, is exceedingly common in elderly populations. Esti-
typically ascribed to psychotropic and sedative-hypnotic mates vary, but the condition appears to be more prevalent
medications, reanalyses of some of these databases have in northern European96 relative to Asian populations,97 and
suggested that poor sleep per se may be a more relevant several genotypes have been identified (see Chapter 90).
predictive factor.82 Peak prevalence was noted in the group ages 60 to 69 years
Population-based studies also have suggested that poor for women (16.3%) and 50 to 59 years for men (7.8%),96
sleep may be linked to lower cognition in old age. In the though the population sampled included persons up to age
Study of Osteoporotic Fractures, a population of older 90 years. Another European study including subjects up to
34 PARTI / Section 1 Normal Sleep and Its Variations

night to night. A 15-night study suggested that estimated

prevalence for PLMS might stabilize only after multiple
nights of measurement.107 The discrepancy between the
higher prevalence of PLMS relative to RLS and the failure
Number of cases

of a number of studies to show associations between their

n ce
ale presence and specific symptoms suggests that in many
Combined prev
elderly persons, PLMS may be an incidental finding.106
The worsening of RLS and PLMS with aging suggests
SDB as an SDB as an that this syndrome may be associated with other conditions
age-related age-dependent known to be common in older populations. Given the
disorder disorder likelihood of anemia among the elderly, current attention
has focused largely on iron transport and storage deficien-
cies.108 Elderly RLS patients with serum ferritin levels of
40 50 60 70 80 less than 45mg/mL showed subjective improvement fol-
Age lowing use of ferrous sulfate, although their total iron
levels were no different.108 These findings were later rep-
Figure 3-3 Heuristic model suggesting that sleep-disordered licated by the same research group.109 Because iron repre-
breathing (SDB) is both an age-related and an age-dependent
condition with potential overlap of distributions in the 60- to sents a key component of production of dopamine, it could
70-year-old age range. Cross-sectionally, note that the number play a role in presence of RLS in some elderly subjects.
of cases observed can remain high and increase with age, One population-based study could not confirm the ferritin
despite a presumed decrease in age-related SDB. finding,110 although another report indicated that higher
serum-soluble transferrin-receptor levels (often character-
istic of early-stage anemia) and lower serum iron were
associated with RLS.96 An interesting perspective on iron
their 80s also showed similar gender differences (14.7% in metabolism and aging was based on examination of ferritin
women; 6.8% in men) with peak prevalence for both levels in the cerebrospinal fluid of elderly RLS patients.
genders in the 50- to 59-year-old range.98 Thus, in some Older patients had higher CSF ferritin levels than did
respects, RLS prevalence appears to be more age related younger patients; however, for elderly patients whose RLS
than age dependent (see later and Fig. 3-3), although at had been long-standing, lower levels were associated with
least one study of a Dutch population noted that preva- a more-severe condition.111
lence was highest in the oldest subjects (80 to 100 years
of age).99
Periodic limb movements in sleep (PLMS) are stereo- SLEEP-DISORDERED BREATHING
typic, repetitive, nonepileptiform movements of the legs Specific considerations related to diagnosis and treatments
usually consisting of dorsiflexion of the ankle but occasion- of SDB in the elderly are covered in Chapter 134.
ally limited to flexion of the great toe or incorporating This section deals with more general issues involving
flexion at the level of knee or hip. They often, but not age-dependence.
invariably, occur in conjunction with RLS. The inter- The previously proposed heuristic model for SDB
movement interval has been reported to decrease with age (Fig. 3-3) posits that SDB represents both an age-related
from about 24 to 28 seconds before the age of 55 years to phenomenon (with a specific vulnerability confined to
about 14 to 16 seconds after the age of 65 years.100 Age- middle age) and an age-dependent phenomenon (with a
dependent increases in the occurrence of PLMS have been prevalence that steadily increases throughout the human
noted cross-sectionally in series without a drop in the life course).1 The articulation and differentiation of these
oldest (e.g., older than 80 years) groups.101 Curiously, lon- two presumably separate, but chronologically overlapping,
gitudinal follow-up of elderly subjects did not show distributions represents a major challenge to clinicians.
increases consistently,102 perhaps owing to inherent vari- Practically, if the health consequences of SDB in elderly
ability in PLMS. Prevalence, defined as a PLM Index of populations are diminished, the necessity to treat the
15 movements or more per hour, has been estimated as enormous numbers of elderly persons who have the condi-
high as 52% in a population of older women.103 When tion is reduced. Age dependence implies that SDB risk
measured with wrist actigraphy in that population, the factors might be best considered markers of physiologic
presence of PLMS was associated with sleep durations of or biological age.112 Chronologic age may thus serve only
less than 5 hours of sleep,104 though earlier population- as a proxy for other risk factors that are themselves
based studies have presented conflicting data as to whether age-dependent.
PLMS, in the absence of frank RLS symptoms, were asso-
ciated with poor sleep.101 One study of men ages 40 to 60 Risk Factors
years noted poorer sleep quality when the PLM index Risk factors for SDB in the older population may differ to
exceeded 10.105 Other studies demonstrating the mixed some extent from those in middle-aged populations. In
pattern of results correlating PLMS with sleep complaints SHHS, several markers of obesity that were significant
have been reviewed elsewhere.106 cross-sectional predictors of SDB in middle-aged popula-
One possible explanation for the variability of results tions (neck circumference and waist-to-hip ratio), were no
across studies is that PLMS may vary considerably from longer significant predictors by age 70 years and 80 years,
CHAPTER 3 Normal Aging 35

respectively,113 although body mass index continued to be and morbidities related to other potential end-organ
correlated with SDB, even past age 80 years, albeit with a damage (see Fig. 3-4). In the absence of large-scale pro-
somewhat diminished effect. Although the male predomi- spective randomized clinical trials specifically targeting
nance in SDB is thought to equalize in old age, this was SDB treatments in geriatric populations, definitive causal
not the case within SHHS.113 Other cohort studies includ- associations with adverse outcomes in the aged remain
ing older subjects suggested roughly equal prevalence in uncertain. The prevailing viewpoint in sleep medicine has
elderly men and women.114,115 been that SDB demonstrates weakened associations with
The prevailing view for many decades was that most morbidities in elderly, relative to middle-aged, persons.
SDB in the elderly consisted of central (i.e., diaphrag- Offered here is a brief description of studies in older popu-
matic) events, whereas in the middle-aged population lations suggesting otherwise.
obstructive events predominated; however this is unsub- In the elderly, SDB has been associated cross-sectionally
stantiated by both descriptive studies, showing the pre- with clinically defined hypertension,125 a nondipping blood
dominance of obstructive apneas, and by pathophysiologic pressure pattern,126 composite cardiovascular disease
studies, which show increased tendency for upper airway history (in men),127 stroke,128 reduced kidney function (in
collapse with aging (see reference 116 for review). Upper men),129 poorer physical function (in men),80 nocturia,130
airway resistance has been reported to be higher in both overactive bladder,131 and impaired cognition (in women).132
REM and NREM sleep in older men relative to younger Longitudinal data have shown a relationship between
men,117 and closing pressures during sleep were higher in declining mental status test scores and the development of
older subjects in N2 sleep relative to younger subjects.118 SDB.133 Additionally, higher health care costs were associ-
Aging has been associated with lengthening of the soft ated with sleep apnea in both middle-aged and elderly
palate and with upper airway fat pad deposition, both of persons.134 An important association between SDB and
which may contribute to oropharygeal collapse during frailty has also been noted in older women,135 which is
sleep.119 Lower lung volumes have been shown to predict particularly important given the fact that, as a well-
incident SDB in elderly persons over time,120 perhaps by acknowledged geriatric syndrome, frailty is highly predic-
providing less caudal traction on the trachea and hastening tive of other morbidities and of mortality.136
upper airway collapse during sleep. In older animals, the In SHHS, when subjects with prevalent cardiovascular
pharyngeal muscles appear to have a worse profile for disease were excluded, relationships between SDB (as
endurance relative to the diaphragm, which may enhance measured by quartiles of the apnea-hypopnea index) and
susceptibility to collapse,121 and a shift from type IIa to various morbidities (including diabetes and hyperlipid-
IIb fibers occurred in the genioglossus in 24-month-old emia) were clearly lower in the over-65-year-old popula-
rats, a finding interpreted as conferring susceptibility to tion than in the under-65-year-old population, but only
fatigue.122 in men, not women, where the associations were similar.137
Predisposing influences on SDB in elderly human By contrast, Haas and colleagues138 reported that isolated
populations are not limited to neuromuscular factors. Ven- systolic hypertension was unrelated to SDB in any age
tilatory control instability,123 which may be accentuated range, but that systolic and diastolic hypertension were
by the decrease of N3 sleep with age, might also predispose related to SDB in only those younger than 60 years.
to SDB in the elderly, though not all studies report In another report examining associations between multi
high loop gain in older subjects.124 These and other poten- ple measures of SDB and more broadly defined car-
tial age-dependent risk factors for SDB are shown in diovascular disease (including coronary heart disease,
Figure 3-4. congestive heart failure, and stroke), relationships with
SDB, although reduced to some extent by age, were
Outcomes still age-independent.139
Potential outcomes relevant to SDB in old age include Despite this suggestive evidence, other studies continue
mortality, cardiovascular and neurobehavioral morbidities, to minimize the significance of SDB for elderly popula-
tions. For example, it has been contended that sleep apnea
has little impact upon quality of life in the elderly,140 and
others have argued that ischemic preconditioning essen-
POTENTIAL POTENTIAL tially renders the SDB of old age innocuous, because
AGE-DEPENDENT AGE-DEPENDENT some component of protective adaptation is likely to have
RISK FACTORS OUTCOMES occurred.141 Goff and colleagues142 have shown that cardio-
Body weight Mortality vascular responses (elevations in heart rate and blood pres-
Lung capacity Neurobehavioral sure) to auditory stimulation are reduced in older relative
Ventilatory control morbidity
Upper airway
to younger persons, and they have interpreted this as con-
as a marker
collapsibility of physiologic morbidity sistent with the reduced associations between SDB and
Muscular endurance aging Other end- systemic hypertension that has been reported in older
Thyroid function organ damage
Sleep fragmentation (e.g., renal) persons. Because the study was performed in older persons
Slow-wave sleep without SDB, the results are difficult to reconcile with
the ischemic preconditioning hypothesis, and it remains
Figure 3-4 Sleep-disordered breathing (SDB) in older adults as
unclear whether such presumed sympathetic downregula-
an age-dependent condition. Other potentially associated age- tion might underlie the presumably reduced impact of
dependent risk factors and outcomes are shown. SDB in the elderly. The consequences of SDB in older
36 PARTI / Section 1 Normal Sleep and Its Variations

populations thus remain an area of considerable contro- function,150 do not adversely affect nocturnal sleep,151 and
versy, and the sleep medicine specialist should be cogni- might even be associated with longer sleep duration the
zant of these issues. For further discussion, the reader is previous night.152 Other studies suggest that naps and
directed elsewhere(See references 1 and 116 for review) and to other hypersomnolence portend mortality153 or ischemic heart
chapters in this volume (Chapters 134 and 136). disease154 and that daytime fatigue or anergy predict diverse
morbidities155 or all-cause mortality.156 Again, however,
not all population-based studies concur, and some suggest
WHY DO OLDER PEOPLE NAP? absence of excess mortality risk associated with napping,157
A time-honored question, asked by both professionals and particularly in elderly persons.149 Clearly, the many reasons
the lay public, involves the significance of napping in old for daytime napping and sleepiness in older populations
age. From the laypersons perspective the question is most continue to be elusive and outcomes associated with the
typically: Is it normal to nap? or Are naps good or bad phenomenon are disparate.(See reference 1 for review) Additionally,
for my health? The sleep medicine specialist may ask the methodologic issues involved in defining napping are
fundamentally similar, though more diagnostically inclined substantial and undoubtedly effect lack of comparability
questions such as: What is the probability that daytime across studies.158
naps in a 75-year-old indicate SDB? Does excessive
sleepiness during the day in an older person portend
dementia? or To what extent does daytime napping BASIC SCIENCE CONSIDERATIONS
adversely effect sleep at night? These are highly relevant Beyond the aforementioned rich description of the complex
questions that are made even more difficult to answer by and interrelated changes in human sleep with advancing
cultural issues related to napping, the complexities in years outlined in this chapter, the ultimate significance of
relying on self-reports to derive estimates of the physio- such alterations remains enigmatic. If one views such
logic tendency of sleep during the daytime hours, and the changes in sleep merely as epiphenomenal to components
fact that, overarching all other issues, sleeping during the of the aging process, they may be merely a consequence of
daytime hours in old age is most assuredly a multi- more fundamental changes in the biology of the organism
determined phenomenon. operating at the system, cellular, or molecular levels. On
Elsewhere I have reviewed the complex matrix of results the other hand, might age-dependent alterations in sleep
that suggest that napping is both a beneficial and poten- and rhythms themselves be potential influences on physi-
tially protective event in the life of an older person as ologic aging per se? If that is indeed the case, then manipu-
well as an identifiable risk factor for numerous morbidi- lations or interventions that alter sleep might modify
ties and even mortality.1 Perhaps the most direct approach disease course, change fundamental processes of aging, or
to answer the question of why older persons are sleepy perhaps even contribute to the longevity of the organism.
during the day is a case-control design that compares Research in this exciting area is only just beginning, but
otherwise demographically well-matched, community- certain provocative clues are emerging, particularly as we
dwelling elderly persons, some of whom are sleepy during learn more about sleep function.
the day and some of whom are not.143 Key to such a Invertebrate models have proved invaluable in under-
design is the incorporation of a diverse array for variables standing more about how sleep is related to the aging
that have been thought to account for sleepiness in older process. The relative strength of the sleepwake cycle and
persons, including subjective and polysomnographically the length of sleep bouts have been shown to breakdown
defined measures of nocturnal sleep fragmentation, careful with age in Drosophila, and the extent of the disruption was
assessment of comorbid medical disease, psychopathology, magnified under higher temperatures (29C) relative to
medication use that might induce sleepiness, alcohol use, more moderate (25C) or cooler (18C) temperatures,45
smoking status, measurements of SDB and PLMS, and an important observation because Drosophila are known to
specific conditions known to disrupt nocturnal sleep, such have a longer life span in cooler temperatures. Addition-
as nocturia and physical pain. Results from such a study ally, incorporation of paraquat, an herbicide known to
indicated that male sex, poor sleep quality, sleep inter- induce oxidative stress, into the flies food supply produced
ruptions due to nocturnal pain or bathroom trips, and similar results.
medications known to induce sleepiness differentiated Examination of sleep in Drosophila mutants with exceed-
cases and controls. Only severe SDB (more than 30 events ingly short sleep durations and rapid physiologic aging
per hour) predicted sleepiness, but PLMS did not.143 (i.e., shortened life span) also suggested that sleep may
Several unanticipated findings (positive relationship to indeed play a vital role in survival.158a In the first of these,
percentage of time spent in REM sleep, negative rela- the minisleep (MNS) line, derived from an exhaustive search
tionship to alcohol consumption) were also noted. Taken of 9000 mutant lines, homozygous flies showed reductions
together, these findings suggest there are many factors in sleep durations of 37% of female flies and 32% of male
that predict why an older person may be sleepy during flies relative to wild-type flies.159 Genotyping suggested
the day. that the MNS line was characterized by a point mutation
Napping has also been associated with falls144 and incipi- in the Shaker gene, thought to regulate potassium channel
ent cognitive decline84 in a population of older women repolarization and broadly defined neuronal excitability.
and with depression,145 nocturia,146 diabetes,147 and lower Perhaps most importantly from the standpoint of aging,
quality of life148 in both men and women. On the other survivorship to very old age in these flies (defined in this
hand, evidence continues to accrue that naps may be pro- study as 56 days), was substantially lower than in flies with
tective for cardiovascular events,149 might improve daytime other Shaker locus mutations or in wild-type controls; in
CHAPTER 3 Normal Aging 37

some cases the effects approached a fivefold reduction in necessity to attend to the myriad number of sleep problems
life span.159 A different fly line, called sleepless (sss), which in old age.163
overexpresses another Shaker-related protein involved in
the downregulation of potassium influx, has been shown
to be associated with even more dramatic reductions in SUMMARY
sleep amount (85% for males, 80% for females).160 In this Defining normality in elderly populations remains a
mutant, reductions in survivorship were even more pro- challenging task. However, an ever-increasing database
found, with virtually none of the sss flies surviving beyond informs the sleep medicine specialist about potential mor-
50 days (median about 30 days) relative to the median bidities that may be associated with poor sleep and various
survival of control flies (about 70 days).160 These data cer- sleep disorders in old age. Multiple factors contribute to
tainly imply that absence of sleep is associated with more poor sleep at night and excessive sleepiness during the day
rapid aging, though they do not suggest what function in the aged human. Although most studies have been
ascribed to sleep may be related to acceleration of such observational and descriptive, the examination of sleep in
aging processes. By contrast, studies of sleep deprivation lower animals might allow greater understanding of age-
in mammals might afford a broader perspective on such dependent changes in sleep as an indicator of biological or
issues. physiologic age and sheds new light on the importance of
Sleep deprivation studies in humans over the last 25 treating sleep problems in human aging.
years have shown repeatedly that sleep loss interacts with
aging in several key ways. Homeostatic pressure for sleep Acknowledgment
may be somewhat diminished in humans and in some This work is supported by National Institute on Aging
rodent species, particularly when recovery sleep is char- grants AG-020269 and AG-025688 and National Institute
acterized by changes in both sleep duration and delta of Neurological Disorders and Stroke grant NS-050595.
activity pressure. In humans, the behavioral consequences
of sleep loss, framed in terms of both greater daytime
sleepiness and more impaired waking performances,
appear diminished, rather than enhanced, with aging (see Clinical Pearls
references 1 and 28 for reviews of this area). These facts
In old age, in particular, the sleep medicine specialist
imply, but cannot prove, that sleep might have less sig-
should always remember that there are numerous
nificance and, as a corollary, loss of sleep may be of less overlapping and contributing reasons why the elderly
consequenceor at very least, no more consequenceas patient has disrupted sleep at night or may be sleepy
the organism ages. The data presented throughout this during the day. Napping should never be assumed to
chapter notwithstanding, such results might be interpreted be innocuous, nor should excessive sleepiness during
hastily as suggesting less of a need to intervene in the the day be attributed unequivocally to active or sub-
sleep of older persons. clinical disease. Sleep-disordered breathing may be
By contrast, important new molecular evidence suggests associated with adverse outcomes in elderly popula-
that at the subcellular level, sleep deprivation may be more tions as well as middle-aged populations and might
detrimental for function in aging animals than in young warrant treatment. Poor sleep might not only reduce
the quality of life of the older person, it might also
animals. More specifically, the unfolded protein response portend adverse health consequences.
within the endoplasmic reticulum occurring after sleep
deprivation161 may be modified substantially by aging.
Protein aggregation is a well-acknowledged feature of
many neurodegenerative conditions in late life and appears
to occur as one of the earliest signs of impending neuronal REFERENCES
death. The unfolded protein response, which consists of a 1. Bliwise DL. Normal aging. In: Kryger MH, Roth T, Dement WC,
host of molecular changes involving attenuation of transla- editors. Principles and practice of sleep medicine. 4th ed. Philadel-
phia: Elsevier; 2005. pp. 24-38.
tion of adverse proteins, degradation of misfolded pro- 2. Petersen RC, Doody R, Kurz A, et al. Current concepts in mild
teins, and promotion of factors (chaperones) protective for cognitive impairment. Arch Neurol 2001;58:1985-1992.
normal function of the endoplasmic reticulum, shows syn- 3. Bliwise DL. Sleep in normal aging and dementia. Sleep 1993;
ergistic age and sleep-deprivation effects. For example, 16:40-81.
expression of a major chaperone, BiP/GRP78, is known to 4. Ohayon MM, Carskadon MA, Guilleminault C, et al. Meta-analysis
of quantitative sleep parameters from childhood to old age in
be downregulated in both brain and liver of aged animals, healthy individuals: developing normative sleep values across the
but it showed major upregulation following sleep depriva- human lifespan. Sleep 2004;27:1255-1273.
tion in young, but not old, mice.161,162 Perhaps even more 5. Floyd JA, Janisse JJ, Jenuwine ES, et al. Changes in REM-sleep
relevant were changes in proapoptotic proteins (e.g., percentage over the adult lifespan. Sleep 2007;30:829-836.
6. Silber MH, Ancoli-Israel S, Bonnet MH, et al. The visual scoring
caspase-12). Aging was associated with greater expression of sleep in adults. J Clin Sleep Med 2007;3:121-131.
of these markers of apoptosis, and sleep deprivation in the 7. Redline S, Kirchner HL, Quan SF, et al. The effects of age, sex,
mouse activated this pathway as well. Strikingly, sleep ethnicity, and sleep-disordered breathing on sleep architecture.
deprivation in the older animal further accentuated such Arch Intern Med 2004;164:406-418.
markers of preprogrammed cell death beyond those seen 8. Vitiello MV. Sleep in normal aging. Sleep Med Clin 2006;1:
in normal aging and beyond those in sleep deprivation.162 9. Latta F, Leproult R, Tasali E, et al. Sex differences in delta and alpha
If these findings are in any way translatable to the human EEG activities in healthy older adults. Sleep 2005;28:
condition, they would suggest a greater, rather than lesser, 1525-1534.
38 PARTI / Section 1 Normal Sleep and Its Variations

10. Latta F, Leproult R, Tasali E, et al. Sex differences in nocturnal 36. Nygard M, Hill RH, Wikstrom MA, et al. Age-related changes in
growth hormone and prolactin secretion in healthy older electrophysiological properties of the mouse suprachiasmatic
adults: relationships with sleep EEG variables. Sleep 2005;28: nucleus in vitro. Brain Res Bull 2005;65:149-154.
1519-1524. 37. Carrier J, Monk TH, Buysse DJ, et al. Inducing a 6-hour phase
11. Colrain IM, Crowley KE, Nicholas CL, et al. Sleep evoked advance in the elderly: effects on sleep and temperature rhythms.
delta frequency responses show a linear decline in amplitude across J Sleep Res 1996;5:99-105.
the adult lifespan. Neurobiol Aging 2008; July 26, epub ahead of 38. Bliwise DL. Sleep and circadian rhythm disorders in aging and
print. dementia. In: Turek F, Zee P, editors. Regulation of sleep
12. Campbell IG, Feinberg I. Homeostatic sleep response to naps is and circadian rhythms. New York: Marcel Dekker; 1999.
similar in normal elderly and young adults. Neurobiol Aging pp. 487-525.
2005;26:135-144. 39. Zee P, Rosenberg RS, Turek FW. Effects of aging on entrainment
13. Darchia N, Campbell IG, Feinberg I. Rapid eye movement density and rate of resynchronization of circadian locomotor activity. Am J
is reduced in the normal elderly. Sleep 2003;26:973-977. Physiol 1992;263:R1099-R1103.
14. Klerman EB, Davis JB, Duffy JF, et al. Older people awaken more 40. Asplund R. Sleep, health, and visual impairment in the elderly. Arch
frequently but fall back asleep at the same rate as younger people. Gerontol Geriatr 2000;30:7-15.
Sleep 2004;27:793-798. 41. Duffy JF, Zeitzer JM, Czeisler CA. Decreased sensitivity to phase-
15. Boselli M, Parrino L, Smerieri A, et al. Effect of age on EEG arousal delaying effects of moderate intensity light in older subjects. Neu-
in normal sleep. Sleep 1998;21:351-357. robiol Aging 2007;28:799-807.
16. Bonnet MH, Arand DL. EEG arousal norms by age. J Clin Sleep 42. Benloucif S, Green K, LHermite-Baleriaux M, et al. Responsive-
Med 2007;3:271-274. ness of the aging circadian clock to light. Neurobiol Aging
17. Crowley K, Trinder J, Kim Y, et al. The effects of normal aging on 2006;27:1870-1879.
sleep spindle and K-complex production. Clin Neurophysiol 43. Baehr EK, Eastman CI, Revelle W, et al. Circadian phase-shifting
2002;113:1615-1622. effects of nocturnal exercise in older compared with younger
18. Tan X, Campbell IG, Feinberg I. Internight reliability and bench- adults. Am J Physiol Regul Integr Comp Physiol 2003;284:R1542-
mark values for computer analyses of non-rapid eye movement R1550.
(NREM) and REM EEG in normal young adult and elderly sub- 44. Shaw PJ, Cirelli C, Greenspan RJ, et al. Correlates of sleep
jects. Clin Neurophysiol 2001;112:1540-1552. and waking in Drosophila melanogaster. Science 2000;287:1834-
19. Moser D, Anderer P, Gruber G, et al. Sleep classification according 1837.
to AASM and Rechtschaffen & Kales: effects on sleep scoring 45. Koh K, Evans JM, Hendricks JC, et al. A Drosophila model for age-
parameters. Sleep 2009;32:139-149. associated changes in sleep:wake cycle. Proc Natl Acad Sci U S A
20. Penzel T, Hirshkowitz M, Harsh J, et al. Digital analysis and techni- 2006;103:13843-13847.
cal specifications. J Clin Sleep Med 2007;3:109-120. 46. Klein BA, Olzsowy KM, Klein A, et al. Caste-dependent sleep of
21. Tononi G, Cirelli C. Sleep function and synaptic homeostasis. Sleep worker honey bees. J Exp Biol 2008;211:3028-3040.
Med Rev 2006;10:49-62. 47. Edgar DM, Dement WC, Fuller CA. Effect of SCN lesions on sleep
22. Massimini M, Huber R, Ferrarelli F, et al. The sleep slow oscillation in squirrel monkeys: evidence for opponent processes in sleep-wake
as a traveling wave. J Neurosci 2004;24:6862-6870. regulation. J Neurosci 1993;13:1065-1079.
23. Crowley K, Sullivan EV, Adalsteinsson E, et al. Differentiating 48. Kripke DF, Simons RN, Garfinkel L, et al. Short and long sleep
pathologic delta from healthy physiologic delta in patients with and sleeping pills: is increased mortality associated? Arch Gen Psy-
Alzheimer disease. Sleep 2005;28:865-870. chiatry 1979;36:103-116.
24. Darchia N, Campbell IG, Tan X, et al. Kinetics of NREM delta 49. Zhdanova IV, Yu L, Lopez-Patino M, et al. Aging of the circadian
EEG power density across NREM periods depend on age and delta- system in zebrafish and the effects of melatonin on sleep and cogni-
band designation. Sleep 2007;30:71-79. tive performance. Brain Res Bull 2008;75:433-441.
25. Terry JR, Anderson C, Horne JA. Nonlinear analysis of EEG during 50. Hida A, Kusanagi H, Satoh K, et al. Expression profiles of
NREM sleep reveals changes in functional connectivity due to PERIOD1, 2 and 3 in peripheral blood mononuclear cells from
natural aging. Hum Brain Mapp 2004;23:73-84. older subjects. Life Sci 2009;84:33-37.
26. Bruce EN, Bruce MC, Vennelaganti S. Sample entropy tracks 51. Kolker DE, Vitaterna MH, Fruechte EM, et al. Effects of age on
changes in EEG power spectrum with sleep state and aging. circadian rhythms are similar in wild-type and heterozygous Clock
J Clin Neurophysiol 2009;26:257-266. mutant mice. Neurobiol Aging 2004;25:517-523.
27. Richardson GS, Carskadon MA, Orav EJ. Circadian variation of 52. Foley DJ, Monjan AA, Brown SL, et al. Sleep complaints among
sleep tendency in elderly and young adult subjects. Sleep elderly persons: an epidemiologic study of three communities. Sleep
1982;5(Suppl. 2):S82-S94. 1995;18:425-432.
28. Bliwise DL. Normal aging. In: Kryger MH, Roth T, Dement WC, 53. Ganguli M, Reynolds CF, Gilby JE. Prevalence and persistence of
editors. Principles and practice of sleep medicine. 3rd ed. Philadel- sleep complaints in a rural older community sample: the MoVIES
phia: Saunders; 2000. pp. 26-42. project. J Am Geriatr Soc 1996;44:778-784.
29. Monk TH, Buysse DJ, Reynolds CF III, et al. Circadian 54. Lichstein KL, Durrence HH, Riedel BW, et al. Epidemiology of
temperature rhythms of older people. Exp Gerontol 1995;30: sleep. Mahwah, NJ: Erlbaum; 2004.
455-474. 55. Hale L, Do DP. Racial differences in self-reports of sleep duration
30. Ferrari E, Cravello L, Falvo F, et al. Neuroendocrine features in in a population-based study. Sleep 2007;30:1096-1103.
extreme longevity. Exp Gerontol 2008;43:88-94. 56. Unruh ML, Redline S, An M-W, et al. Subjective and objective
31. Bliwise DL, Ansari FP, Straight LM, et al. Age changes in timing sleep quality and aging in the Sleep Heart Health Study. J Am
and 24-hour distribution of self-reported sleep. Am J Geriat Psy- Geriatr Soc 2008;56:1218-1227.
chiatry 2005;13:1077-1082. 57. Paudel ML, Taylor BC, Diem SJ, et al. Associations between
32. Duffy JF, Zeitzer JM, Rimmer DW. Peak of circadian melatonin depressive symptoms and sleep disturbances in community-dwelling
rhythm occurs later within the sleep of older subjects. Am J Physiol older men. J Am Geriatr Soc 2008;56:1228-1235.
Endocrinol Metab 2002;282:E297-E303. 58. Kochar J, Fredman L, Stone KL, et al. Sleep problems in elderly
33. Czeisler CA, Duffy JF, Shanahn TL, et al. Stability, precision and women caregivers depend on the level of depressive symptoms:
near-24-hour period of the human circadian pacemaker. Science results of the CaregiverStudy of Osteoporotic Fractures. J Am
1999;284:2177-2181. Geriatr Soc 2007;55:2003-2009.
34. Zeitzer JM, Duffy JF, Lockley SW. Plasma melatonin rhythms in 59. Spira A, Stone K, Beaudreau SA, et al. Anxiety symptoms and
young and older humans during sleep, sleep deprivation, and wake. objectively measured sleep quality in older women. Am J Geriatr
Sleep 2007;30:1437-1443. Psychiatry 2009;17:136-143.
35. Dijk DJ, Duffy JF, Czeisler CA. Age-related increase in awakenings: 60. Perlis ML, Smith LJ, Lyness JM, et al. Insomnia as a risk factor
impaired consolidation of nonREM sleep at all circadian phases. for onset of depression in the elderly. Beh Sleep Med 2006;4:
Sleep 2001;24:565-577. 104-113.
CHAPTER 3 Normal Aging 39

61. Carrier J, Bliwise DL. Sleep and circadian rhythms in normal aging. 86. Ohayon MM, Vecchierini M-F. Normative sleep data, cognitive
In: Billiard M, editor. Le sommeil normal et pathologique. function and daily living activities in older adults in the community.
Hingham, Mass: Kluwer; 2003. pp. 297-332. Sleep 2005;28:981-989.
62. Vitiello MV, Moe KE, Prinz PN. Sleep complaints cosegregate with 87. Gangwisch JE, Heymsfield SB, Boden-Albala B, et al. Sleep dura-
illness in older adults: clinical research informed by and informing tion associated with mortality in elderly, but not middle-aged, adults
epidemiologic studies of sleep. J Psychosom Res 2002; in a large US sample. Sleep 2008;31:1087-1096.
53:555-559. 88. Patel SR, Blackwell T, Redline S, et al. The association between
63. Bliwise DL, Foley DJ, Vitiello MV, et al. Nocturia and disturbed sleep duration and obesity in older adults. Int J Obesity 2008;
sleep in the elderly. Sleep Med 2009;10:540-548. 32:1825-1834.
64. Gooneratne NS, Gehrman PR, Nkwuo JE, et al. Consequences of 89. Patel SR, Malhotra A, White DP, et al. Association between reduced
comorbid insomnia symptoms and sleep-related breathing disorder sleep and weight gain in women. Am J Epidemiol 2006;164:
in elderly subjects. Arch Inter Med 2006;166:1732-1738. 947-954.
65. Klink ME, Quan SF, Kaltenborn WT, et al. Risk factors associated 90. Gangwisch JE, Malaspina D, Boden-Albala B, et al. Inadequate
with complaints of insomnia in a general adult population. Arch Int sleep as a risk factor for obesity: analyses of the NHANES I. Sleep
Med 1992;152:1634-1637. 2005;28:1289-1296.
66. Dodge R, Cline MG, Quan SF. The natural history of insomnia and 91. Gottlieb DJ, Redline S, Nieto FJ, et al. Association of usual sleep
its relationship to respiratory symptoms. Arch Intern Med duration with hypertension: the Sleep Heart Health Study. Sleep
1995;155:1797-1800. 2006;29:1009-1014.
67. Foley DJ, Monjan A, Simonsick EM, et al. Incidence and remission 92. Van den Berg JF, Tulen JHM, Neven AK, et al. Sleep duration and
among elderly adults: an epidemiologic study of 6,800 persons over hypertension are not associated in the elderly. Hypertension
three years. Sleep 1999;22(Suppl. 2):S366-S372. 2007;50:585-589.
68. Quan SF, Katz R, Olson J, et al. Factors associated with incidence 93. Phillips B, Buzkova P, Enright P, et al. Insomnia did not predict
and persistence of symptoms of disturbed sleep in an elderly cohort: incident hypertension in older adults in the Cardiovascular Health
the Cardiovascular Health Study. Am J Med Sci 2005;329: Study. Sleep 2009;32:65-72.
163-172. 94. Gottlieb DJ, Punjabi NM, Newman AB, et al. Association of sleep
69. Morgan K. Daytime activity and risk factors for late-life insomnia. time with diabetes mellitus and impaired glucose tolerance. Arch
J Sleep Res 2003;12:231-238. Intern Med 2005;165:863-868.
70. Reid KJ, Martinovich Z, Finkel S, et al. Sleep: a marker of physical 95. Gangwisch JE, Heymsfield SB, Boden-Albala B, et al. Sleep dura-
and mental health in the elderly. Am J Geriatr Psychiatry tion as a risk factor for diabetes incidence in a large US sample.
2006;14:860-866. Sleep 2007;30:1667-1673.
71. Morin CM, LeBlanc M, Daley M, et al. Epidemiology of insomnia: 96. Hogl B, Kiechl S, Willeit J, et al. Restless legs syndrome: a com-
prevalence, self-help treatments, consultations, and determinants of munity-based study of prevalence, severity and risk factors. Neurol-
help-seeking behaviors. Sleep Med 2006;7:123-130. ogy 2005;64:1920-1924.
72. Soldatos CR, Allaert FA, Ohta T, et al. How do individuals sleep 97. Mizuno S, Miyaoka T, Inagaki T, et al. Prevalence of restless legs
around the world? Results from a single-day survey in ten countries. syndrome in non-institutionalized Japanese elderly. Psychiatry Clin
Sleep Med 2005;6:5-13. Neurosci 2005;59:461-465.
73. Yokoyama E, Saito Y, Kaneita Y, et al. Association between subjec- 98. Wenning GK, Kiechl S, Seppi K, et al. Prevalence of movement
tive well-being and sleep among the elderly in Japan. Sleep Med disorders in men and women aged 50-89 years (Bruneck
2008;9:157-164. Study cohort): a population-based study. Lancet Neurol 2005;4:
74. Zilli I, Ficca G, Salzarulo P. Factors involved in sleep satisfaction in 815-820.
the elderly. Sleep Med 2009;10:233-239. 99. Rijsman R, Neven AK, Graffelman W, et al. Epidemiology
75. Strine TW, Chapman DP. Associations of frequent sleep insuffi- of restless legs in the Netherlands. Eur J Neurol 2004;11:607-
ciency with health-related quality of life and health behaviors. Sleep 611.
Med 2005;6:23-27. 100. Ferri R, Manconi M, Lanuzza B, et al. Age-related changes in
76. Krystal AD. Treating the health, quality of life, and functional periodic leg movements during sleep in patients with restless legs
impairments in insomnia. J Clin Sleep Med 2007;3:63-72. syndrome. Sleep Med 2008;9:790-798.
77. Ancoli-Israel S, Krystal AD, McCall WV, et al. A 12-week, random- 101. Ancoli-Israel S, Kripke DF, Klauber MR, et al. Periodic limb move-
ized, double-blind, placebo-controlled study evaluating the effect of ments in sleep in community-dwelling elderly. Sleep 1991;14:
eszopiclone 2 mg on sleep/wake function in older adults with 496-500.
primary and co-morbid insomnia. Sleep 2009; in press. 102. Gehrman P, Stepnowsky C, Cohen-Zion M, et al. Long-term fol-
78. Latimer Hill E, Cumming RG, Lewis R, et al. Sleep disturbances low-up of periodic limb movements in sleep in older adults. Sleep
and falls in older people. J Gerontol A Biol Sci Med Sci 2007; 2002;25:340-343.
62:62-66. 103. Claman DM, Redline S, Blackwell T, et al. Prevalence and corre-
79. Stone KL, Ancoli-Israel S, Blackwell T, et al. Actigraphy-measured lates of periodic limb movements in older women. J Clin Sleep Med
sleep characteristics and risk of falls in older women. Arch Inter 2006;2:438-445.
Med 2008;168:1768-1775. 104. Mehra R, Stone KL, Ancoli-Israel S, et al. Interpreting wrist
80. Dam T-TL, Ewing S, Ancoli-Israel S, et al. Association between actigraphic indices of sleep in epidemiologic studies of the
sleep and physical function in older men: the Osteoporotic Frac- elderly: the Study of Osteoporotic Fractures. Sleep 2008;31:
tures in Men Sleep Study. J Am Geriatr Soc 2008;56:1665-1673. 1569-1576.
81. Goldman SE, Stone KL, Ancoli-Israel S, et al. Poor sleep is associ- 105. Carrier J, Frenette S, Montplaisir J, et al. Effects of periodic leg
ated with poorer physical performance and greater functional limi- movements during sleep in middle-aged subjects without sleep com-
tations in older women. Sleep 2007;30:1317-1324. plaints. Mov Disord 2005;20:1127-1132.
82. Stone KL, Ensrud KE, Ancoli-Israel S. Sleep, insomnia and falls in 106. Bliwise DL. Restless legs syndrome: manifestations in aging and
elderly patients. Sleep Med 2008;9(Suppl. 1):S18-S22. dementia. In: Avidan AY, Alessi CA, editors. Geriatric sleep medi-
83. Blackwell T, Yaffe K, Ancoli-Israel S, et al. Poor sleep is associated cine. New York: Informa Healthcare; 2008. pp. 197-208.
with impaired cognitive function in older women: the study of 107. Trotti LM, Bliwise DL, Greer SA, et al. Correlates of PLMs
osteoporotic fractures. J Gerontol Med Sci 2006;61A:405-410. variability over multiple nights and impact upon RLS diagnosis.
84. Yaffe K, Blackwell T, Barnes DE, et al. Preclinical cognitive decline Sleep Med 2009;10:668-671.
and subsequent sleep disturbance in older women. Neurology 108. OKeeffe ST, Gavin K, Lavan JN. Iron status and restless legs
2007;69:237-242. syndrome in the elderly. Age Ageing 1994;23:200-203.
85. Tworoger SS, Lee S, Schernhammer ES, et al. The association of 109. OKeeffe ST. Secondary causes of restless legs syndrome in older
self-reported sleep duration, difficulty sleeping, and snoring with people. Age Ageing 2005;34:349-352.
cognitive function in older women. Alzheimer Dis Assoc Disord 110. Berger K, von Eckardstein A, Trenkwalder C, et al. Iron
2006;20:41-81. metabolism and the risk of restless legs syndrome in an elderly
40 PARTI / Section 1 Normal Sleep and Its Variations

general populationthe MEMO study. J Neurol 2002;249:1195- 135. Endeshaw YW, Unruh ML, Kutner M, et al. Sleep disordered
1199. breathing and frailty in the Cardiovascular Health Study Cohort.
111. Earley CJ, Connor JR, Beard JL, et al. Ferritin levels in the cere- Am J Epidemiol 2009;170:193-202.
brospinal fluid and restless legs syndrome: effects of different clini- 136. Fried LP, Tangen CM, Walston J, et al. Frailty in older adults:
cal phenotypes. Sleep 2005;28:1069-1075. evidence for a phenotype. J Gerontol A Biol Sci Med Sci 2001;
112. Bliwise DL. Chronologic age, physiologic age, and mortality in 56:M146-M156.
sleep apnea. Sleep 1996;19:275-276. 137. Newman AB, Nieto FJ, Guidry U, et al. Relation of sleep-disor-
113. Young T, Shahar E, Nieto FJ, et al. Predictors of sleep-disordered dered breathing to cardiovascular disease risk factors. Am J Epide-
breathing in community-dwelling adults. Arch Intern Med miol 2001;15:50-59.
2002;162:893-900. 138. Haas DC, Foster GL, Nieto FJ, et al. Age-dependent associations
114. Bixler EO, Vgontzas AN, Lin H-M, et al. Excessive daytime sleepi- between sleep-disordered breathing and hypertension: importance
ness in a general population sample: the role of sleep apnea, age, of discriminating between systolic/diastolic hypertension and iso-
obesity, diabetes and depression. J Clin Endocrinol Metab lated systolic hypertension in the Sleep Heart Health Study. Circu-
2005;90:4510-4515. lation 2005;111:614-621.
115. Tishler PV, Larkin EK, Schluchter MD, et al. Incidence of sleep- 139. Shahar E, Whitney CW, Redline S, et al. Sleep-disordered
disordered breathing in an urban adult population: the relative breathing and cardiovascular disease: cross-sectional results of the
importance of risk factors in the development of sleep disordered Sleep Heart Health Study. Am J Respir Crit Care Med 2001;
breathing. JAMA 2003;289:2230-2237. 163:19-25.
116. Hoban TF, Bliwise DL. Ontogeny. In: Kushida C, editor. Obstruc- 140. Martinez-Garcia MA, Soler-Cataluna JJ, Roman-Sanchez P, et al.
tive sleep apnea: pathophysiology, comorbidities and consequences. Obstructive sleep apnea has little impact on quality of life in the
New York: Informa Healthcare; 2007. pp. 39-59. elderly. Sleep Med 2009;10:104-111.
117. Thurnheer R, Wraith PK, Douglas NJ. Influence of age and gender 141. Lavie L, Lavie P. Ischemic preconditioning as a possible explanation
on upper airway resistance in NREM and REM sleep. J Appl for the age decline relative mortality in sleep apnea. Med Hypoth-
Physiol 2001;90:981-988. eses 2006;66:1069-1073.
118. Eikermann M, Jordan AS, Chamberlin NL, et al. The influence of 142. Goff EA, ODriscoll DM, Simonds AK, et al. The cardiovascular
aging on pharyngeal collapsibility during sleep. Chest 2007;131: response to arousal from sleep decreases with age in healthy adults.
1702-1709. Sleep 2008;31:1009-1017.
119. Malhotra A, Huang Y, Fogel R, et al. Aging influences on pharyn- 143. Pack AI, Dinges DF, Gehrman PR, et al. Risk factors for exces-
geal anatomy and physiology: the predisposition to pharyngeal col- sive sleepiness in older adults. Ann Neurol 2006;59:893-
lapse. Am J Med 2006;119:e9-e14. 904.
120. Bliwise DL. Epidemiology of age-dependence in sleep-disordered 144. Stone KL, Ewing SK, Lui L-Y, et al. Self-reported sleep and
breathing in old age: the Bay Area Sleep Cohort. Sleep Med Clinics nap habits and risk of falls and fractures in older women: the
2009;4:57-64. Study of Osteoporotic Fractures. J Am Geriatr Soc 2006;54:
121. Van Lunteren E, Vafaie H, Salomone RJ. Comparative effects of 1177-1183.
aging on pharyngeal and diaphragm muscles. Respir Physiol 145. Bixler EO, Vgontzas AN, Lin HM, et al. Prevalence of sleep-dis-
1995;99:113-125. ordered breathing in women: effects of gender. Am J Respir Crit
122. Oliven A, Carmi N, Coleman R, et al. Age-related changes in upper Care Med 2001;163:608-613.
airway muscles: morphological and oxidative properties. Exp 146. Foley DJ, Vitiello MV, Bliwise DL, et al. Frequent napping is associ-
Gerontol 2001;36:1673-1686. ated with excessive daytime sleepiness, depression, pain and noctu-
123. Carlson BW, Neelon VJ, Carlson JR, et al. Respiratory periodicity ria in adults: findings from the National Sleep Foundation 2003
and electroencephalogram arousals during sleep in older adults. Biol Sleep in America poll. Am J Geriatr Psychiatry 2007;
Res Nurs 2007;8:249-260. 15:344-350.
124. Wellman A, Malhotra A, Jordan AS, et al. Chemical control stability 147. Picarsic JL, Glynn NW, Taylor CA, et al. Self-reported napping and
in the elderly. J Physiol 2007;581(Pt. 1):291-298. duration and quality of sleep in the lifestyle interventions and inde-
125. Endeshaw YW, Bloom HL, Bliwise DL. Sleep-disordered breathing pendence for elders pilot study. J Am Geriatr Soc 2008;
and cardiovascular disease in the Bay Area Sleep Cohort. Sleep 56:1674-1680.
2008;31:563-568. 148. Gooneratne NS, Weaver TE, Cater JR, et al. Functional outcomes
126. Endeshaw YE, White WB, Kutner M, et al. Sleep-disordered of excessive daytime sleepiness in older adults. J Am Geriatr Soc
breathing and 24-hour blood pressure pattern among older adults. 2003;51:642-649.
J Gerontol A Biol Sci Med Sci 2009;64A:280-285. 149. Naska A, Oikonomou E, Trichopoulou A, et al. Siesta in healthy
127. Mehra R, Stone KL, Blackwell T, et al. Prevalence and correlates adults and coronary mortality in the general population. Arch Intern
of sleep-disordered breathing in older men: osteoporotic fractures Med 2007;167:296-301.
in men sleep study. J Am Geriatr Soc 2007;55:1356-1364. 150. Campbell SS, Murphy PJ, Stauble TN. Effects of a nap on night-
128. Munoz R, Duran-Cantolla J, Martinez-Vila E, et al. Severe sleep time sleep and waking function in older subjects. J Am Geriatr Soc
apnea as a risk of ischemic stroke in the elderly. Stroke 2006; 2005;53:48-53.
37:2317-2321. 151. Dautovich ND, McCrae CS, Rowe M. Subjective and objective
129. Canales MT, Taylor BC, Ishani A, et al. Reduced renal function and napping and sleep in older adults: are evening naps bad for night-
sleep-disordered breathing in community-dwelling older men. time sleep? J Am Geriatr Soc 2008;56:1681-1686.
Sleep Med 2008;9:637-645. 152. Goldman SE, Hall M, Boudreau R, et al. Association between
130. Endeshaw YW, Johnson TM, Kutner MH, et al. Sleep-disordered nighttime sleep and napping in older adults. Sleep 2008;31:
breathing and nocturia in older adults. J Am Geriatr Soc 733-740.
2004;52:957-960. 153. Bursztyn M, Stessman J. The siesta and mortality: twelve years
131. Kemmer H, Mathes AM, Dilk O, et al. Obstructive sleep apnea of prospective observations in 70-year-olds. Sleep 2005;28:345-
syndrome is associated with overactive bladder and urgency incon- 347.
tinence in men. Sleep 2009;32:271-275. 154. Elwood P, Hack M, Pickering J, et al. Sleep disturbance, stroke and
132. Spira AP, Blackwell T, Stone KL, et al. Sleep-disordered breathing heart disease events: evidence from the Caerphilly cohort. J Epide-
and cognition in older women. J Am Geriatr Soc 2008;56:45- miol Community Health 2006;60:69-73.
50. 155. Cheng H, Gurland BJ, Maurer MS. Self-reported lack of energy
133. Cohen-Zion M, Stepnowsky C, Marler M, et al. Changes in cog- (anergia) among elders in a multiethnic community. J Gerontol A
nitive function associated with sleep disordered breathing in older Biol Sci Med Sci 2008;63:707-714.
people. J Am Geriatr Soc 2001;49:1622-1627. 156. Hardy SE, Studenski SA. Fatigue predicts mortality in older adults.
134. Tarasiuk A, Greenberg-Dotan S, Simon-Tuval T, et al. The effect J Am Geriatr Soc 2008;56:1910-1914.
of obstructive sleep apnea on morbidity and health care utilization 157. Lan T-Y, Lan T-H, Wen C-P, et al. Nighttime sleep, Chinese
of middle-aged and older adults. J Am Geriatr Soc 2008;56: afternoon nap and mortality in the elderly. Sleep 2007;30:1105-
247-254. 1110.
CHAPTER 3 Normal Aging 41

158. Vitiello MV. We have much more to learn about the relationship 161. Naidoo N, Giang W, Galante RJ, et al. Sleep deprivation induces
between napping and health in older adults. J Am Geriatr Soc the unfolded protein response in mouse cerebral cortex. J Neuro-
2008;56:1753-1755. chem 2005;92:1150-1157.
158a. Bushey D, Hughes KA, Tononi G, Cirelli C. Sleep, aging, and 162. Naidoo N, Ferber M, Master M, et al. Aging impairs the unfolded
lifespan in Drosophila. BMC Neurosci 2010;11:56. protein response to sleep deprivation and leads to proapoptotic
159. Cirelli C, Bushey D, Hill S, et al. Reduced sleep in Drosophila Shaker signaling. J Neurosci 2008;28:6539-6548.
mutants. Nature 2005;434:1087-1092. 163. Bloom HG, Ahmed I, Alessi CA, et al. Evidence-based recommen-
160. Koh K, Joiner WJ, Wu MN, et al. Identification of SLEEPLESS, a dations for the assessment and management of sleep disorders in
sleep-promoting factor. Science 2008;321:372-376. older persons. J Am Geriatr Soc 2009 May;57(5):761-789.
42 PARTI / Section 1 Normal Sleep and Its Variations
Daytime Sleepiness and Alertness
Timothy Roehrs, Mary A. Carskadon, William C. Dement,
and Thomas Roth 4
Abstract significance. Sleepiness is caused by reduced sleep time as
Sleepiness is a problem reported by 10% to 25% of the popu- often seen in otherwise healthy adults, by fragmented and
lation, depending on the definition of sleepiness used and disrupted sleep as found in patients with primary sleep dis-
the population sampled. It is more frequent in young adults orders, by administration of sedating drugs and discontinu-
and elderly people. Sleepiness is a physiological need state ation of alerting drugs, and by various neurological disorders.
with its intensity evident by how rapidly sleep onset occurs, Sleepiness has a normal circadian rhythm that is increased
how easily sleep is disrupted, and how long sleep endures. in circadian rhythm misalignments such as those occurring
Validated self-rated scales and physiological measures are in shift work or jet lag. Excessive and persistent sleepiness
available to assess the presence and degree of sleepiness. is life-threatening, but when its presence is recognized and
Relative to sleepy, healthy adults, the chronicity and irrevers- its etiology identified, it can be successfully treated or at
ibility of sleepiness is indicative of its clinical and pathological least minimized.

INTRODUCTION Sleepiness in Limited Populations or

Scientific and clinical attention to sleepiness arose from Populations of Convenience
the recognition of excessive daytime sleepiness (EDS) as a In surveys of relatively small, selected populations, 0.3%
symptom associated with serious life-threatening medical to 36% of respondents reported excessive sleepiness.
conditions. In the late 1960s, this symptomwhich earlier Surveys with reported excessive sleepiness rates of less than
had been ignored, attributed to lifestyle excesses, viewed 3% generally are from earlier studies that focused on
as a sign of laziness and malingering, or at best, seen as a hypersomnia.1 In later studies, in which 4% to 9% rates
sign of narcolepsybegan to be seriously studied by scien- were reported, more specific questions about excessive
tistclinicians. Methods to detect and quantify sleepiness sleepiness during the day or relative to ones peers were
were developed. The result has been a growing scientific asked.2 In some surveys, postprandial, or midday, sleepi-
literature on the nature of sleepiness and its determinants ness was distinguished from sleepiness at other times of
in clinical populations, selected populations of healthy vol- the day, a distinction discussed later in regard to the cir-
unteers, and the general population. cadian correlates of sleepiness. Somewhat higher rates are
This chapter will review information regarding the reported in other selected populations using sleepiness
prevalence of sleepiness in the population. The various scales. For example, the Epworth Sleepiness Scale (ESS),
methods used to measure sleepiness in the population and a scale that requires one to estimate the likelihood of
laboratory will be described and guidelines regarding clini- falling asleep in different situations, was completed by 740
cal assessment of sleepiness will be offered. The nature day workers in 8 industrial plants in Israel and 23% of
and neurobiological substrates of sleepiness will be dis- respondents had ESS scores indicative of excessive sleepi-
cussed and the known determinants of sleepiness will be ness (i.e., scores >10).3
described. Finally, the clinical and public health signifi- Prevalence rates for sleepiness of 15% and greater
cance of persistent complaints of sleepiness will be also have been found for specific age groups that are
discussed. consistent with smaller laboratory studies using the physi-
ological measure of sleepiness, the Multiple Sleep Latency
Test (MSLT), which is described later. Young adults
EPIDEMIOLOGY OF SLEEPINESS were sleepier, on average, than a comparison group of
Prevalence estimates of sleepiness in the population vary middle-aged adults, and about 20% of the young adults
widely depending on the definition of sleepiness used and had mean daily sleep latencies of less than 5 minutes, a
the type of population sampled. Surveys and question- level of sleepiness considered pathological.4 Healthy
naires have queried regarding the experience of a mood or elderly also were found to be physiologically sleepier
feeling state of sleepiness, fatigue, or tiredness; about than middle-aged adults.5 In surveys of the work force
falling asleep unintentionally; or about struggling to stay engaged in shift or night work, complaints of excessive
awake and fighting sleep onset. New developments in the sleepiness during waking hours are more frequent than
epidemiology of sleepiness have included the use of stan- among day workers, and continuous ambulatory electro-
dardized sleepiness scales and physiological assessments of encephalographic (EEG) field monitoring has confirmed
sleepiness that measure the behavior of falling asleep, its the sleepiness.6
estimated likely occurrence or the speed of its actual occur-
rence. While another important focus has been on sleepi- Sleepiness in Representative Populations
ness in children and adolescents, this chapter only will Representative survey studies of national populations
address sleepiness in adults. have been done. In a study representative of the Finnish

CHAPTER 4 Daytime Sleepiness and Alertness 43

population, 11% of women and 7% of men reported at home, on average 7.1hr vs. standard laboratory 8.5hr).
daytime sleepiness almost every day.7 In another survey, In Figure 4-1 the distribution of sleepiness, defined as
representative of a large geographical area in Sweden, 12% average sleep latency on the MSLT, is illustrated for the
of respondents thought their sleep was insufficient.8 In that Michigan population representative sample. The average
survey, insufficient sleep, and not its consequent daytime sleep latency of various clinical samples and experimental
sleepiness, was the focus of the questions. Two studies sleep time manipulations is provided for comparisons.
representative of the United States population used the
MSLT to assess sleepiness. Given the necessary time com- Risk Factors for Sleepiness
mitment required of participants in MSLT studies, the The risk factors for sleepiness identified in the various
representative integrity of study results is critically depen- surveys includes hours of daily sleep, employment status,
dent on the recruitment response rate. From a large south- marital status, snoring, and depression. Among 26- to
eastern Michigan random sample (n=1648) representative 35-year-old members of a large health maintenance orga-
of the U.S. population, a subsample (n=259) with a 68% nization in Michigan, respondents reported 6.7 hours of
response rate was recruited to undergo a nocturnal poly- sleep on weekdays and 7.4 hours on weekend days, on
somnogram (NPSG) and MSLT the following day. The average.11 The hours of sleep were inversely related to
prevalence of excessive sleepiness, defined as a MSLT daytime sleepiness scores on the SleepWake Activity
average sleep latency of less than 6 minutes, was 13%.9 In Inventory (SWAI). Both these variables were related to
another probability sample of 6,947 Wisconsin state employment and marital status, with full employment and
employees, a subsample (n=632), collected with a 52% being single predictive of less sleep time and more sleepi-
response rate, slept at home and then completed a MSLT ness. Self-reported snoring and depression, as measured by
in the laboratory the next day. Twenty-five percent had an a structured diagnostic interview, were also associated with
average sleep latency of less than 5 minutes.10 These two increased sleepiness. In the Finnish study cited earlier,
studies also used the ESS to assess sleepiness; in the Michi- sleepiness was associated with moderate to severe depres-
gan study 20% had ESS scores greater than 10, and in the sion and with snoring more than three times per week.7
Wisconsin study 25% had scores greater than 11. The
higher prevalence in the Wisconsin study, despite the NATURE OF SLEEPINESS
more stringent definition of sleepiness (MSLT of 5 vs. 6
minutes and ESS of 11 vs. 10), could be attributed to an Physiological Need State
age difference in the samples (51 vs. 42 yr on average) or Sleepiness, according to a consensus among sleep research-
the previous nights sleep time and circumstances (habitual ers and clinicians, is a basic physiological need state.12

7.7% 29.0%

Excessive daytime sleepiness Moderate sleepiness

Anesthesia Residents
Mean = 11.4 8 hr TIB x 5 nights
Sleep apnea
6 hr TIB x 4 nights
0 hr TIB x 1 night
10 hr TIB x 14 nights
(full alertness)




0 5 10 15 20
MSLT mean sleep latency (min)

Figure 4-1 The distribution of mean daily sleep latency (min) on the multiple sleep latency test in a subsample (n = 259) recruited
(68% response rate) from a large Southeastern Michigan random sample (N = 1648) representative of the U.S. population. The
population mean is 11.4 minutes and this is compared to means reported for various patient groups60,70,71 and the means found
in healthy normals after various bedtime manipulations.23,62 TIB, time in bed.
44 PARTI / Section 1 Normal Sleep and Its Variations

It may be likened to hunger or thirst, which are physiologi- theta activity shown during a simulated driving task.15
cal need states basic to the survival of the individual organ- Studies have also shown that compensation occurs to the
ism. The presence and intensity of this state can be inferred cognitive and behavioral effects of experimental sleep
by how readily sleep onset occurs, how easily sleep is dis- restriction and increased sleepiness, particularly when the
rupted, and how long sleep endures. Deprivation or restric- sleep loss is mild and accumulates at a slow rate.16 The
tion of sleep increases sleepiness, and as hunger or thirst absence of a readily apparent behavioral deficiency prob-
is reversible by eating or drinking, sleep reverses sleepi- ably also contributes to the subjective-objective disparity
ness. In the organisms daily homeostatic economy, severe seen in chronically sleepy individuals. Finally, findings
deprivation states do not normally occur and hence are not from a general population study suggest that subjective
routinely responsible for regulating eating or drinking; sleepiness has multiple dimensions, beyond an increased
other factors (i.e., taste, smell, time-of-day, social factors, tendency to fall asleep.17 Consequently, patients often
biological variables) modulate these behaviors before mistake chronic debilitating fatigue for sleepiness.18
severe deprivation states develop. Similarly, routine con- The specific nature of this physiological need state is
sumption of sleep is not purely homeostatic, but is greatly unclear. Whether sleepiness is one-dimensional, varying
influenced by social (i.e., job, family, and friends) and envi- only in severity, or multidimensional, varying as to etiol-
ronmental (i.e., noise, light, and bed) factors. ogy or chronicity, has been discussed.19 If it is one-dimen-
The subjective experience of sleepiness and its behav- sional, whether or not sleepiness and alertness are at
ioral indicators (yawning, eye rubbing, nodding) can be opposite poles of the dimension is also an issue. Earlier, it
reduced under conditions of high motivation, excitement, was noted that sleepiness and alertness are being used as
exercise, and competing needs (e.g., hunger, thirst); that antonyms, which suggests a unipolar state. However, it is
is, physiological sleepiness may not necessarily be mani- possible that sleepiness varies from presence to absence
fest. The expression of mild to moderate sleepiness can be and is distinct from alertness. It was noted that sleepiness
masked by any number of factors that are alerting, includ- may be multidimensional, and among the different types
ing motivation, environment, posture, activity, light, or of sleepiness cited are rapid eye movement (REM) versus
food intake. Studies have shown that average sleep latency nonrapid eye movement (NREM) and core versus
on the MSLT is increased by 6 minutes when sitting versus optional sleepiness.19 A complete discussion of the heuris-
lying in bed and also by 6 minutes when immediately pre- tic value and evidence to support these distinctions is
ceded by a 5-minute walk.13 However, when physiological beyond the scope of this chapter. Nonetheless, the point
sleepiness is most severe and persistent, the ability to must be made that these theoretical perspectives may be
reduce its impact on overt behavior wanes. The likelihood colored by different measures, experimental demands,
of sleep onset increases and the intrusion of microsleeps populations studied, and subject or patient motivations
into ongoing behavior occurs. On the other hand, a physi- (i.e., sensitivity to and capacity to counteract sleepiness).
ologically alert (sleepiness and alertness are used here as
antonyms) person does not experience sleepiness or appear Neural Substrates of Sleepiness
sleepy even in the most soporific situations. Heavy meals, The substrates of sleepiness have yet to be determined. It
warm rooms, boring lectures, and the monotony of long- is assumed that sleepiness is a central nervous system
distance automobile driving unmask physiological sleepi- (CNS) phenomenon with identifiable neural mechanisms
ness when it is present, but they do not cause it. and neurochemical correlates. Various electrophysiologi-
Within a conventional 24-hour sleep and wake schedule, cal events suggestive of incipient sleep processes appear in
maximum sleepiness ordinarily occurs in the middle of the behaviorally awake organisms undergoing sleep depriva-
night when the individual is sleeping, and consequently tion. In sleep-deprived animals, ventral hippocampal spike
this sleepiness typically is not experienced or remembered. activity, which normally is a characteristic of NREM sleep,
When forced to be awake in the middle of the night, one increases during behavioral wakefulness and in the absence
experiences loss of energy, fatigue, weariness, difficulty of the usual changes in cortical EEG indicative of sleep.20
concentrating, and memory lapses. When significant phys- Human beings deprived, or restricted, of sleep show iden-
iological sleepiness (as a result of reduced sleep quantity tifiable microsleep episodes (brief intrusions of EEG indi-
or quality) intrudes on ones usual waking activities during cations of sleep) and increased amounts of alpha and theta
the day, similar symptoms are experienced. activity while behaviorally awake.21 The evidence suggests
Adaptation to the chronic experience of sleepiness most that these electrophysiological events are indicants of
probably occurs. Clinicians have reported anecdotally that sleepiness.
successfully treated patients will frequently comment that An emerging literature of neuroimaging studies, both
they had forgotten the experience of complete alertness. structural and functional, have suggested specific brain
Reduced sensitivity to chronic sleepiness is a likely expla- systems that may be involved in sleepiness. Sleep depriva-
nation for the disparities between subjective assessments, tion in young healthy volunteers reduced regional cerebral
even when done with validated scales and the MSLT.5,14 glucose metabolism, as assessed by positron emission
Typically, it is the most sleepy individuals that show the tomography, in thalamic, basal ganglia, and limbic regions
greatest disparity in subjective versus objective assess- of the brain.22 Functional magnetic resonance imaging
ments.5,14 Such individuals deny sleepiness despite signifi- (fMRI) after chlorpheniramine (a sedating antihistamine)
cant objective indicators of sleepiness. On the other hand, compared with placebo showed increased frontal and tem-
basally alert individuals (ESS mean 5.6, SEM 0.3) after a poral activation.23 Because fMRI was conducted while the
one night acute sleep restriction were quite accurate in subject was performing cognitive tasks, the authors inter-
estimating their sleepiness relative to the increases in EEG preted the observed brain activation to have resulted from
CHAPTER 4 Daytime Sleepiness and Alertness 45

the increased mental effort, due to sleepiness, required to dence limits making definitive conclusions. The space here
perform the task. Two groups of patients with severe or is too limited to discuss all the evidence in detail. In
slight hypersomnia associated with paramedian thalamic conclusion, although it is widely held that sleepiness is a
stroke on an MRI showed lesions involving dorso- and physiological state, its physiological substrates are as yet
centromedial thalamic nuclei, bilateral lesions in the severe not fully defined.
group and unilateral in the slight group.24 As yet, these
imaging data are not conclusive. They do suggest it may ASSESSMENT OF SLEEPINESS
be possible to identify brain regions and functions that vary
with sleepiness. But the nature of the alteration may Quantifying Sleepiness
depend on the behavioral load imposed on the sleepy The behavioral signs of sleepiness include yawning, ptosis,
subject as well as the cause of the sleepiness. reduced activity, lapses in attention, and head nodding. An
The neurochemistry of sleepiness and alertness involves individuals subjective report of his or her level of sleepi-
critical and complex issues that have not yet been fully ness also can be elicited. As noted earlier, a number of
untangled (see Chapters 18, 37, 42, and 44). First, a basic factors such as motivation, stimulation, and competing
issue concerns whether sleepiness and alertness have a neu- needs can reduce the behavioral manifestation of sleepi-
rochemistry specific and unique from that associated with ness. Thus, behavioral and subjective indicators often
the sleep process, per se. Second, it is not clear whether underestimate physiological sleepiness.
sleepiness and alertness are controlled by separate neuro- Assessment problems were evident early in research on
chemicals or by a single substance or system. Third, the the daytime consequences of sleep loss. Sleep loss compro-
relation of the neurochemistry of sleepiness and alertness mises daytime functions; virtually everyone experiences
to circadian mechanisms has not yet been determined. dysphoria and reduced performance efficiency when not
Given the number of questions, it should be of no surprise sleeping adequately. But a majority of the tasks used to
that these are areas of active research. assess the effects of sleep loss are insensitive.31 In general,
Neurophysiological studies of sleep and wake mecha- only long and monotonous tasks are reliably sensitive to
nisms have implicated histamine, serotonin, the catechol- changes in the quantity and quality of nocturnal sleep. An
amines, and acetylcholine in control of wake and exception is a 10-minute visual vigilance task, completed
gamma-aminobutyric acid (GABA) for sleep.25 Evidence repeatedly across the day, during which lapses (response
from animal studies is emerging that suggests extracellular times greater than or equal to 500 milliseconds) and
adenosine is the homeostatic sleep factor, with brain levels declines in the best response times are increasingly observed
accumulating during prolonged wakefulness and declining as sleep is lost, either during total deprivation or cumula-
during sleep.26 The peptide hypocretin/orexin has received tively over nights of restricted bedtimes.32
much attention for its role in the pathophysiology of nar- In various measures of mood, including factor analytic
colepsy.27 It is considered to be a major wake-promoting scales, visual analogue scales, and scales for specific aspects
hypothalamic neuropeptide and a hypocretin/orexin defi- of mood, subjects have shown increased fatigue or sleepi-
ciency has been found in human narcolepsy. Its interactive ness with sleep loss. Among the various subjective mea-
role in the homeostatic control of sleep and sleepiness has sures of sleepiness, the Stanford Sleepiness Scale (SSS) is
yet to be determined. It is discussed in greater detail in the best validated.33 Yet clinicians have found that chroni-
Chapters 18, 37, 42, and 44. cally sleepy patients may rate themselves alert on the SSS
Pharmacological studies provide other interesting even while they are falling asleep behaviorally.34 Such
hypotheses regarding the neurochemistry of sleepiness- scales are state measures that query individuals about how
alertness. For example, the benzodiazepines induce sleepi- they feel at the present moment. Another perspective is
ness and facilitate GABA function at the GABAA receptor to view sleepiness behaviorally, as in the likelihood of
complex, thus implicating this important and diffuse falling asleep, and thus ask individuals to rate that likeli-
inhibitory neurotransmitter.25 Another example involves hood in different social circumstances and over longer
histamine, which is now considered to be a CNS neu- periods. The Epworth Sleepiness Scale (ESS) has been
rotransmitter and is thought to have CNS-arousing activ- validated in clinical populations showing a 74% sensitivity
ity.25 Antihistamines that penetrate the CNS produce and 50% specificity relative to the MSLT in a study of
sleepiness.28 A functional neuroimaging study of histamine sleep disorders patients.35 It asks about falling asleep in
H1 receptors in human brain found that the degree of settings in which patients typically report falling asleep
sleepiness associated with cetirizine (20mg) was correlated (e.g., while driving, at church, in social conversation). The
to the degree of H1 receptor occupancy.29 time frame over which ratings are to be made is 2 to 4
Stimulant drugs suggest several other transmitters and weeks.
neuromodulators. The mechanism of action of one class The standard physiological measure of sleepiness, the
of drugs producing psychomotor stimulation and arousal, MSLT, similarly conceptualizes sleepiness as the tendency
the amphetamines, is blockade of catecholamine uptake.30 to fall asleep by measuring the speed of falling asleep. The
Another class of stimulants, the methylxanthines, which MSLT has gained wide acceptance within the field of sleep
include caffeine and theophylline, are adenosine receptor and sleep disorders as the standard method of quantifying
antagonists. Adenosine, considered the key neurochemical sleepiness.36 Using standard polysomnographic techniques,
in the homeostatic regulation of sleep, has inhibitory activ- this test measures, on repeated opportunities at 2-hour
ity on the two major excitatory neurotransmitters acetyl- intervals throughout the day, the latency to fall asleep
choline and glutamate. It may be a biomarker of while lying in a quiet, dark bedroom. The MSLT is based
sleepiness.26 On the other hand, some contradictory evi- on the assumption, as outlined earlier, that sleepiness is a
46 PARTI / Section 1 Normal Sleep and Its Variations

physiological need state that leads to an increased tendency high. For example, MSLT scores after sleep deprivation,44
to fall asleep. The metric typically used to express sleepi- after administration of sedating antihistamines,45 and after
ness has been average daily sleep latency (i.e., mean of the benzodiazepine administration46 correlate with measures
four or five tests conducted), but survival analyses have also of performance and even prove to be the most sensitive
been successfully used.37 The reliability and validity of this measure. Studies also have compared levels of sleepiness
measure have been documented in a variety of experimen- to the known performance-impairing effects of alcohol.47
tal and clinical situations.38 In contrast to tests of perfor- A study relating performance lapses on a vigilance task to
mance, motivation does not seem to reduce the impact of the cumulative effects of sleep restriction found a function
sleep loss as measured by the MSLT. After total sleep comparable to that of the MSLT under a similar cumula-
deprivation, subjects can compensate for impaired perfor- tive sleep restriction (Fig. 4-2).48 The reason many studies
mance, but they cannot stay awake long while in bed in a have found weak correlations between performance and
darkened room, even if they are instructed to do so.39 MSLT at normal or moderate levels of sleepiness is that
An alternative to the MSLT, suggested by some clinical laboratory performance and MSLT are differentially
investigators, is the Maintenance of Wakefulness Test affected by variables such as age, education, and
(MWT). This test requires that subjects lie in bed or sit in motivation.
a chair in a darkened room and try to remain awake.40 Like For the most part, the literature relating sleepiness and
the MSLT, the measure of ability to remain awake is the behavioral functioning has focused on psychomotor and
latency to sleep onset. The test has not been standardized: attention behaviors with the major outcomes being
there are 20-minute and 40-minute versions, and the response slowing and attentional lapses. These impair-
subject is variously sitting upright in a chair, lying in bed, ments can be attributed to slowed processing of informa-
or semirecumbent in bed. The reliability of the MWT has tion and microsleeps, that is intrusion of sleep preparation
not been established either. One study reported sensitivity and sleep onset behaviors. Research has focused on other
to the therapeutic effects of continuous positive airway behavioral domains not as clearly associated with sleep-
pressure (CPAP) in patients with sleep apnea,41 and several mediated behaviors, including decision-making and pain
studies reported sensitivity to the therapeutic effects of sensation. Several studies have shown that increased sleepi-
stimulants in narcolepsy.42 A study attempted to tease apart ness is associated with poor risk-taking decisions.49 Sleep
the critical factors being measured by the MWT and con- loss and its associated sleepiness have also been shown to
cluded that, unlike the MSLT which measures level of increase pain sensitivity.50
sleepiness, the MWT measures the combined effects of
level of sleepiness and the degree of arousal as defined by
heart rate.43
The rationale for the MWT is that clinically the critical
issue for patients is how long wakefulness can be main-
tained. A basic assumption underlying this rationale,
however, may not be valid: it assumes that a set of circum- 16
stances can be evaluated in the laboratory that will reflect
an individuals probability of staying awake in the real 14
world. Such a circumstance is not likely because environ-
ment, motivation, circadian phase, and any competing 12
drive states all affect an individuals tendency to remain
awake. Stated simply, an individual crossing a congested 10
intersection at midday is more likely to stay awake than an
individual driving on an isolated highway in the middle of 8
the night. The MSLT, on the other hand, addresses the
question of the individuals risk of falling asleep by estab- 6
lishing a setting to maximize the likelihood of sleep onset:
all factors competing with falling asleep are removed from 4
the test situation. Thus, the MSLT identifies sleep ten- PVT lapses
dency or clinically identifies maximum risk for the patient. 2 Sleep latency (min)
Obviously, the actual risk will vary from individual to indi-
vidual, from hour to hour, and from environment to 0
environment. B P1 P2 P3 P4 P5 P6 P7
Relation of Sleepiness to Day
Behavioral Functioning
Figure 4-2 Similar functions relating mean daily sleep latency
Given that the MSLT is a valid and reliable measure of on the multiple sleep latency test (MSLT) and mean daily lapses
sleepiness, the question arises as to how this measure on the visual psychomotor vigilance test (PVT) to the cumula-
relates to an individuals capacity to function. Direct cor- tive effects of sleep restriction (about 5 hours of bedtime
relations of the MSLT with other measures of perfor- nightly) across 7 consecutive nights (P1 to P7). (Redrawn from
Dinges DF, Pack F, Williams K, et al. Cumulative sleepiness,
mance under normal conditions have not been too robust. mood disturbance, and psychomotor vigilance performance
Several studies have found, however, that when sleepiness decrements during a week of sleep restricted to 4-5 hours per
is at maximum levels, correlations with performance are night. Sleep 1997;20:275.)
CHAPTER 4 Daytime Sleepiness and Alertness 47

and a score of more than 10 minutes is considered to be

Clinical Assessment of Sleepiness in the normal range (see Fig. 4-1 for MSLT results in the
Assessing the clinical significance of a patients complaint general population). The MSLT is also useful in identify-
of excessive sleepiness can be complex for an inexperienced ing sleep-onset REM periods (SOREMPs), which are
clinician. The assessment depends on two important common in patients with narcolepsy.52 The American
factors: chronicity and reversibility. Chronicity can be Academy of Sleep Medicine Standards of Practice Com-
explained simply. Although a healthy normal individual mittee has concluded that the MSLT is indicated in the
may be acutely sleepy, the patients sleepiness is persistent evaluation of patients with suspected narcolepsy.52 MSLT
and unremitting. As to reversibility, unlike the healthy results, however, must also be evaluated with respect to
normal person, increased sleep time may not completely the conditions under which the testing was conducted.
or consistently ameliorate a patients sleepiness. Patients Standards have been published for administering the
with excessive sleepiness may not complain of sleepiness MSLT, which must be followed to obtain a valid, inter-
per se, but rather its consequences: loss of energy, fatigue, pretable result.36
lethargy, weariness, lack of initiative, memory lapses, or
difficulty concentrating. DETERMINANTS OF SLEEPINESS
To clarify the patient complaint, it is important to focus
on soporific situations in which physiological sleepiness is Quantity of Sleep
more likely to be manifest, as was discussed earlier. Such The degree of daytime sleepiness is directly related to the
situations might include watching television, reading, amount of nocturnal sleep. The performance effects of
riding in a car, listening to a lecture, or sitting in a warm acute and chronic sleep deprivation are discussed in Chap-
room. Table 4-1 presents the commonly reported sleep- ters 5 and 6. As to sleepiness, partial or total sleep depriva-
inducing situations for a large sample of patients with tion in healthy normal subjects is followed by increased
sleep apnea syndrome. After clarifying the complaint, one daytime sleepiness the following day.38 Therefore, modest
should ask the patient about the entire day: morning, nightly sleep restriction accumulates over nights to pro-
midday, and evening. In the next section, it will become gressively increase daytime sleepiness and performance
clear that most adults experience sleepiness over the lapses (see Fig. 4-1).53 However, the speed at which sleep
midday. However, patients experience sleepiness at other loss is accumulated is critical, as studies have shown adap-
times of the day as well, and often throughout the day. tation to a slow accumulation of 1 to 2 hours nightly
Whenever possible, objective documentation of sleepiness occurs, which then increases the duration of the subse-
and its severity should be sought. As indicated earlier, the quent recovery process.23 Increased sleep time in healthy,
standard and accepted method to document sleepiness but sleepy, young adults by extending bedtime beyond the
objectively is the MSLT. usual 7 to 8 hours per night produces an increase in alert-
Guidelines for interpreting the results of the MSLT are ness (i.e., reduction in sleepiness).54 Further, the pharma-
available (see Fig. 4-1).36 A number of case series of patients cological extension of sleep time by an average of 1 hour
with disorders of excessive sleepiness have been published in elderly people produces an increase in mean sleep
with accompanying MSLT data for each diagnostic clas- latency on the MSLT (i.e., increased alertness).55
sification.51 These data provide the clinician with guide- Reduced sleep time explains the excessive sleepiness of
lines for evaluating the clinical significance of a given several patient and nonpatient groups. For example, a sub-
patients MSLT results. Although these data cannot be group of sleep clinic patients has been identified whose
considered norms, a scheme for ranking MSLT scores to excessive daytime sleepiness can be attributed to chronic
indicate degree of pathology has been suggested.41 An insufficient sleep.56 These patients show objectively docu-
average daily MSLT score of 5 minutes or fewer suggests mented excessive sleepiness, normal nocturnal sleep with
pathological sleepiness, a score of more than 5 minutes but unusually high sleep efficiency (time asleeptime in bed),
fewer than 10 minutes is considered a diagnostic gray area, and report about 2 hours more sleep on each weekend day
than each weekday. Regularizing bedtime and increasing
time in bed produces a resolution of their symptoms and
normalized MSLT results.57 The increased sleepiness of
Table 4-1 Sleep-Inducing Situations for Patients healthy young adults also can be attributed to insufficient
with Apnea*
nocturnal sleep. When the sleepiest 25% of a sample of
SITUATION PERCENTAGE OF PATIENTS young adults is given extended time in bed (10 hours) for
as long as 5 to 14 consecutive nights, their sleepiness is
Watching television 91 reduced to a level resembling the general population
Reading 85 mean.54
Riding in a car 71 Individual differences in tolerability to sleep loss have
Attending church 57 been reported.58 These differences can be attributed to a
Visiting friends and 54 number of possible factors. A difference in the basal level
relatives of sleepiness at the start of a sleep time manipulation is
Driving 50 quite possible given the range of sleepiness in the general
Working 43 population (see Fig. 4-1). The basal differences may reflect
Waiting for a red light 32
insufficient nightly sleep relative to ones sleep need.54
There also may be differences in the sensitivity and respon-
*N = 384 patients. sivity of the sleep homeostat to sleep loss, that is how large
48 PARTI / Section 1 Normal Sleep and Its Variations

a sleep deficit the system can tolerate and how robustly the surgery (i.e., apneas remain) show no decrease in arousals
sleep homeostat produces sleep when detecting deficiency. or sleepiness, despite improved sleeping oxygenation.62
Finally, genetic differences in sleep need, the set point Similarly, CPAP, by providing a pneumatic airway splint,
around which the sleep homeostat regulates daily sleep reduces breathing disturbances and consequent arousals
time, have long been hypothesized and one study has sug- from sleep and reverses EDS.63 The reversal of daytime
gested a gene polymorphism may mediate vulnerability to sleepiness following CPAP treatment of sleep apnea syn-
sleep loss.59 These all are fertile areas for research. drome is presented in Figure 4-4. The hours of nightly
CPAP use predicts both subjective and objective measures
Quality of Sleep of sleepiness.64
Daytime sleepiness also relates to the quality and the con- Experimental fragmentation of the sleep of healthy
tinuity of a previous nights sleep. Sleep in patients with a normal subjects has been produced by inducing arousals
number of sleep disorders is punctuated by frequent, brief with an auditory stimulus. Several studies have shown that
arousals of 3 to 15 second durations. These arousals are subjects awakened at various intervals during the night
characterized by bursts of EEG speeding or alpha activity demonstrate performance decrements and increased sleep-
and, occasionally, transient increases in skeletal muscle iness on the following day.65 Studies have also fragmented
tone. Standard scoring rules for transient EEG arousals sleep without awakening subjects by terminating the stim-
have been developed.60 A transient arousal is illustrated in ulus on EEG signs of arousal rather than on behavioral
Figure 4-3. These arousals typically do not result in awak- response. Increased daytime sleepiness (shortened laten-
ening by either Rechtschaffen and Kales sleep staging cri- cies on the MSLT) resulted from nocturnal sleep fragmen-
teria or behavioral indicators, and the arousals recur in tation in one study,66 and in a second study, the recuperative
some conditions as often as 1 to 4 times per minute. The effects (measured as increased latencies on the MSLT) of
arousing stimulus differs in the various disorders and can a nap following sleep deprivation were compromised by
be identified in some cases (apneas, leg movements, pain) fragmenting the sleep on the nap.67
but not in others. Regardless of etiology, the arousals gen- One nonclinical population in which sleep fragmenta-
erally do not result in shortened sleep but rather in frag- tion is an important determinant of excessive sleepiness is
mented or discontinuous sleep, and this fragmentation the elderly. Many studies have shown that even elderly
produces daytime sleepiness.61 people without sleep complaints show an increased number
Correlational evidence suggests a relation between sleep of apneas and periodic leg movements during sleep.68 As
fragmentation and daytime sleepiness. Fragmentation, as noted earlier, the elderly as a group are sleepier than other
indexed by number of brief EEG arousals, number of shifts groups.5 Furthermore, it has been demonstrated that
from other sleep stages to stage 1 sleep or wake, and the elderly people with the highest frequency of arousal during
percentage of stage 1 sleep correlates with EDS in various sleep have the greatest daytime sleepiness.69
patient groups.61 Treatment studies also link sleep frag-
mentation and excessive sleepiness. Patients with sleep Circadian Rhythms
apnea syndrome who are successfully treated by surgery A biphasic pattern of objective sleep tendency was observed
(i.e., number of apneas are reduced) show a reduced fre- when healthy, normal young adult and elderly subjects
quency of arousals from sleep as well as a reduced level of were tested every 2 hours over a complete 24-hour day.70
sleepiness, whereas those who do not benefit from the During the sleep period (11:30 pm to 8:00 am) the latency

1 night
12 *,**
Latency to sleep (min)

14 nights
EMG *,**
(submental) 9 42 nights
C4 A1

Oz A1 6 *

50 V
1 sec
Figure 4-3 A transient arousal (on right side of figure) frag- Pre Post
menting sleep. The preexistence of sleep is evident by the
K-complex at second 9 of the epoch preceding the arousal. CPAP treatment
LE-A1, Left electrooculogram referenced to A1; RE-A1, right
electrooculogram referenced to A1; EMG, electromyogram from Figure 4-4 Mean daily sleep latency on the multiple sleep
submental muscle; C4-A1, electroencephalogram referenced to latency test (MSLT) in patients with obstructive sleep apnea
A1 from C4 placement; Oz-A1, electroencephalogram refer- syndrome before (pre) and after (post) 1, 14, and 42 nights of
enced to A1 from Oz placement; V5, electrocardiogram from continuous positive airway pressure (CPAP) treatment.
V5 placement. (Redrawn from American Sleep Disorders Asso- *, P < .05; **, P < .01. (Redrawn from Lamphere J, Roehrs T,
ciation. EEG arousals: scoring rules and examples. Sleep 1992; Wittig R, et al. Recovery of alertness after CPAP in apnea. Chest
15:173-184.) 1989;96:1364-1367.)
CHAPTER 4 Daytime Sleepiness and Alertness 49

Earlier, it was noted that shift workers are unusually

Mean latency to sleep onset (min) sleepy, and jet travelers experience sleepiness acutely in a
new time zone. The sleepiness in these two conditions
20 results from the placement of sleep and wakefulness at
times that are out of phase with the existing circadian
15 rhythms. Thus, not only is daytime sleep shortened and
fragmented but also wakefulness occurs at the peak of
sleepiness or trough of alertness. Several studies have
5 shown that pharmacological extension and consolidation
Elderly Young adult of out-of-phase sleep can improve daytime sleepiness (see
0 Chapters 42, 73, and 81 for more detail).76 Yet, the basal
0930 1330 1730 2130 0130 0530 0930 circadian rhythm of sleepiness remains, although the
Time of day overall level of sleepiness has been reduced. In other words,
the synchronization of circadian rhythms to the new sleep
Figure 4-5 Latency to sleep at 4-hour intervals across the wake schedule is not hastened.
24-hour day. Testing during the daytime followed standard
multiple sleep latency test (MSLT) procedures. During the night, CNS Drugs
from 11:30 PM to 8:00 AM (shaded area), subjects were awak-
ened every 2 hours for 15 minutes, and latency of return to Sedating Drug Effects
sleep was measured. Elderly subjects (n = 10) were 60 to 83 Central nervous system (CNS) depressant drugs, as
years old; young subjects (n = 8) were 19 to 23 years old. expected, increase sleepiness. Most of these drugs act as
(Redrawn from Carskadon MA, Dement WC: Daytime sleepiness:
Quantification of a behavioral state. Neurosci Biobehav Rev agonists at the GABAA receptor complex. The benzodiaz-
1987;111:307-317. Copyright 1987, Elsevier Science.) epine hypnotics hasten sleep onset at bedtime and shorten
the latency to return to sleep after an awakening during
the night (which is their therapeutic purpose), as demon-
strated by a number of objective studies.77 Long-acting
benzodiazepines continue to shorten sleep latency on the
testing was accomplished by awakening subjects for 15 MSLT the day following bedtime administration.77 Finally,
minutes and then allowing them to return to sleep. Two ethanol administered during the daytime (9:00 am) reduces
troughs of alertnessone during the nocturnal hours sleep latency in a dose-related manner as measured by the
(about 2:00 to 6:00 am) and another during the daytime MSLT.78
hours (about 2:00 to 6:00 pm)were observed. Figure 4-5 Second generation antiepileptic drugs, including gaba-
shows the biphasic pattern of sleepinessalertness. pentin, tiagabine, vigabatrin, pregabalin, and others,
Other research protocols have yielded similar results. In enhance GABA activity through various mechanisms that
constant routine studies, where external environmental directly or indirectly involve the GABAA receptor.79 The
stimulation is minimized and subjects remain awake, sedating effects of these various drugs have not been thor-
superimposed on the expected increase in self-rated fatigue oughly documented, but some evidence indicates they do
resulting from the deprivation of sleep is a biphasic circa- have sedative activity. GABAB receptor agonists are being
dian rhythmicity of self-rated fatigue similar to that seen investigated as treatments for drug addictions and the pre-
for sleep latency.71 In another constant routine study in clinical animal research suggests these drugs may have
which EEG was continuously monitored, a biphasic pattern sedative activity as well.80
of unintentional sleep was observed.72 In studies with Antagonists acting at the histamine H1 receptor also
sleep scheduled at unusual times, the duration of sleep have sedating effects. One of the most commonly reported
periods has been used as an index of the level of sleepiness. side effects associated with the use of H1 antihistamines is
A pronounced circadian variation in sleep duration is daytime sleepiness. Several double-blind, placebo-con-
found with the termination of sleep periods closely related trolled studies have shown that certain H1 antihistamines,
to the biphasic sleep latency function in the studies cited such as diphenhydramine, increase sleepiness using sleep
earlier.73 If individuals are permitted to nap when they are latency as the objective measure of sleepiness, whereas
placed in time-free environments, this biphasic pattern others, such as terfenadine or loratadine do not.81 The
becomes quite apparent in the form of a midcycle nap.74 difference among these compounds relates to their dif-
This circadian rhythm in sleepiness is part of a circadian ferential CNS penetration and binding. Others of the H1
system in which many biological processes vary rhythmi- antihistamines (e.g., tazifylline) are thought to have a
cally over 24 hours. The sleepiness rhythm parallels the greater peripheral compared with central H1 affinity, and,
circadian variation in body temperature, with shortened consequently, effects on daytime sleep latency are found
latencies occurring in conjunction with temperature only at relatively high doses.81
troughs.70 But these two functions, sleep latency and body Antihypertensives, particularly beta adrenoreceptor
temperature, are not mirror images of each other; the blockers, are also reported to produce sedation during the
midday body temperature decline is relatively small com- daytime.82 These CNS effects are thought to be related to
pared with that of sleep latency. Furthermore, under free- the differential liposolubility of the various compounds.
running conditions, the two functions become dissociated.75 However, we are unaware of any studies that directly
However, no other biological rhythm is as closely associ- measure the daytime sleepiness produced by beta-blockers;
ated with the circadian rhythm of sleepiness as is body the information is derived from reports of side effects. As
temperature. noted earlier, it is important to differentiate sleepiness
50 PARTI / Section 1 Normal Sleep and Its Variations

from tiredness or fatigue. Patients may be describing tired- effects of ethanol. In contrast, caffeine and methylpheni-
ness or fatigue resulting from the drugs peripheral effects date produced a similar increase in alertness, regardless of
(i.e., lowered cardiac output and blood pressure), not the basal level of sleepiness. Clinically, these findings
sleepiness, a presumed central effect. imply, for example, that a sleepy driver with minimal blood
Sedative effects of dopaminergic agonists used in treat- ethanol levels may be as dangerous as an alert driver who
ing Parkinsons disease have been reported as adverse is legally intoxicated.89
events in clinical trials and in case reports as sleep attacks The basal state of sleepiness also influences drug-seek-
while driving.83 It is now clear these sleep attacks are not ing behavior. The likelihood that a healthy normal person
attacks per se, but are the expression of excessive sleepi- without a drug abuse history will self-administer methyl-
ness. Whereas the dose-related sedative effect of these phenidate is greatly enhanced after 4 hours of sleep the
drugs has been established, the mechanism by which the previous night compared to 8 hours of sleep (see Fig. 4-4).
sedative effects occur is unknown. The dopaminergic ago- Though not experimentally demonstrated as yet, self
nists are also known to disrupt and fragment sleep.84 Thus, administration of caffeine also is probably influenced by
the excessive sleepiness may be secondary to disturbed basal state of sleepiness. The high volume of caffeine use
sleep, or to a combination of disturbed sleep and direct in the population probably relates to the high rate of self
sedative effects. medication for sleepiness due to chronic insufficient sleep
in the population.
Alerting Drug Effects
Stimulant drugs reduce sleepiness and increase alertness. CNS Pathologies
The drugs in this group differ in their mechanisms of Pathology of the CNS is another determinant of daytime
action. Amphetamine, methylphenidate, and pemoline sleepiness. The previously noted hypocretin/orexin defi-
block dopamine reuptake and to a lesser extent enhance ciency is thought to cause excessive sleepiness in patients
the release of norepinephrine, dopamine, and serotonin. with narcolepsy.90 Another sleep disorder associated with
The mechanism of modafinil is not established; some evi- excessive sleepiness due to an unknown pathology of the
dence suggests that modafinil has a mechanism distinct CNS is idiopathic CNS hypersomnolence. A report of a
from the classical stimulants. Amphetamine, methylpheni- series of rigorously diagnosed cases (n=77) found moder-
date, pemoline, and modafinil are used to treat the EDS ate MSLT scores (8.5 minutes mean latency) relative to
associated with narcolepsy and some have been studied as narcoleptics (4.1 minutes mean latency).91 As yet hypocre-
medications to maintain alertness and vigilance in normal tin/orexin deficiency has not been shown in this disorder.
subjects under conditions of sustained sleep loss (e.g., mili- These two conditions are described in detail in Chapters
tary operations). Studies in patients with narcolepsy using 84 and 86.
MSLT or MWT have shown improved alertness with Excessive sleepiness is reported in other neurological
amphetamine, methylphenidate, modafinil, and pemo- diseases. A study in patients with myotonic dystrophy,
line.85 There is dispute as to the extent to which the exces- type 1 reported excessive sleepiness on the MSLT and
sive sleepiness of narcoleptics is reversed and the reduced cerebrospinal levels of hypocretin/orexin.92 Sleep
comparative efficacy of the various drugs. In healthy attacks have been reported in Parkinsons disease and
normal persons restricted or deprived of sleep, both assessment with the MSLT suggests these attacks are the
amphetamine and methylphenidate increase alertness on expression of excessive daytime sleepiness.93 What remains
the MSLT and improve psychomotor performance.86,87 unresolved in the excessive sleepiness of Parkinsons
Caffeine is an adenosine receptor antagonist. Caffeine, in disease is the relative contribution of the disease itself, the
doses equivalent to 1 to 3 cups of coffee, reduced daytime fragmentation of sleep due to periodic leg movement or
sleepiness on the MSLT in normal subjects after 5 hours apnea , and the dopaminergic drugs used in treating Par-
of sleep the previous night.88 kinsons disease.94 The previously cited study found no
differences in sleepiness as a function of prescribed drug
Influence of Basal Sleepiness or sleep fragmentation, although further assessment in
The preexisting level of sleepinessalertness interacts with larger samples is necessary to confirm this finding. Sleepi-
a drug to influence the drugs behavioral effect. In other ness may be prominent after traumatic brain injury.94a,94b
words, a drugs effect differs when sleepiness is at its
maximum compared to its minimum. As noted previously,
the basal level of daytime sleepiness can be altered by CLINICAL AND PUBLIC HEALTH
restricting or extending time in bed58; this in turn alters SIGNIFICANCE OF SLEEPINESS
the usual effects of a stimulating versus a sedating drug. A Although the patients at sleep disorders centers are not
study showed comparable levels of sleepinessalertness representative of the general population, they do provide
during the day following 5 hours in bed and morning (9:00 some indications regarding the clinical significance of
am) caffeine consumption compared to 11 hours in bed and sleepiness. Their sleepwake histories directly indicate the
morning (9:00 am) ethanol ingestion. Follow-up studies serious impact excessive sleepiness has on their lives.95
explored the dose relations of ethanols interaction with Nearly half the patients with excessive sleepiness report
basal sleepiness.89 Dose-related differences in daytime automobile accidents; half report occupational accidents,
sleepiness following ethanol and 8 hours of sleep were some life threatening; and many have lost jobs because of
diminished after even 1 night of 5 hour sleep, although the their sleepiness. In addition, sleepiness is considerably dis-
measured levels of ethanol in breath were consistent day ruptive of family life.96 An elevated automobile accident
to day. In other words, sleepiness enhanced the sedative rate (i.e., sevenfold) among patients with excessive sleepi-
CHAPTER 4 Daytime Sleepiness and Alertness 51

ness has been verified through driving records obtained than 10 hours of sleep a day were about 1.8 times more
from motor vehicle agencies.97 likely to die prematurely than those sleeping between 7
Population-based information regarding traffic and and 8 hours daily.111 This survey, however, associated
industrial accidents also suggests a link between sleepiness hypersomnia and increased mortality and not necessarily
and life-threatening events. Verified automobile accidents EDS, for which the relation is currently unknown.
occurred more frequently in a representative sample of
people with MSLT scores of 5 minutes or less.98 The
highest rate of automobile accidents occurs in the early Clinical Pearl
morning hours, which is notable because the fewest auto- Sleepiness, when most excessive and persistent, is a
mobiles are on the road during these hours. Also during signal to the individual to stop operating because it
these early morning hours, the greatest degree of sleepi- is dangerous and life-threatening to continue without
ness is experienced.99 Long-haul truck drivers have acci- sleep. It is also a signal to the clinician that there may
dents most frequently (even corrected for hours driving be some underlying pathology that can be success-
before the accident) during the early morning hours, again fully treated, or in the very least minimized, as to its
when sleepiness reaches its zenith.100 vital, life-threatening impact.
Workers on the graveyard shift were identified as a
particularly sleepy subpopulation. In 24-hour ambulatory
EEG recordings of sleep and wakefulness, workers (20% REFERENCES
in one study) were found to actually fall asleep during the 1. Bixler ED, Kales JD, Soldatos CR, et al. Prevalence of sleep disor-
night shift.8 Not surprisingly, the poorest job performance ders in the Los Angeles metropolitan area. Am J Psychiatry
consistently occurs on the night shift, and the highest rate 1979;136:1257-1262.
of industrial accidents is usually found among workers on 2. Partinen M. Sleeping habits and sleep disorders of Finnish men
this shift.101 Medical residents are another particularly before, during, and after military service. Ann Med Milit Fenn
sleepy subpopulation. In surveys those reporting five or 3. Melamed S, Oksenberg A. Excessive sleepiness and risk of occupa-
fewer hours of sleep per night were more likely to make tional injuries in non-shift daytime workers. Sleep 2001;25:315-322.
medical errors, report serious accidents, and were two 4. Levine B, Roehrs T, Zorick F, et al. Daytime sleepiness in young
times more likely to be named in medical malpractice adults. Sleep 1988;11:39-46.
5. Dement WC, Carskadon MA. An essay on sleepiness. In: Baldy-
suits.102,103 In a survey of medical house-staff, 49% reported Mouliner M, editor. Actualits en medecine experimentale. Mont-
falling asleep while driving and 90% of the episodes pellier, France: Euromed; 1981. pp. 47-71.
occurred postcall compared to 13% fall-asleep episodes 6. Torsvall L, Akerstedt T, Gillander K, et al. Sleep on the night shift:
reported by the medical faculty and 20 of the 70 house- 24h EEG monitoring of spontaneous sleep/wake behavior. Psycho-
staff were involved in automobile accidents compared to physiology 1989;26:352-358.
7. Hublin C, Kaprio J, Partinen M, et al. Daytime sleepiness in an
11 of the 85 faculty.104 adult Finnish population. J Intern Med 1996;239:417-423.
Cognitive function is also impaired by sleepiness. Adults 8. Broman JE, Lundh LG, Hetta J. Insufficient sleep in the general
with various disorders of excessive sleepiness have cogni- population. Neurophysiol Clin 1996;26:30-39.
tive and memory problems.105 The memory deficiencies 9. Drake CL, Roehrs T, Richardson G, et al. Epidemiology and mor-
bidity of excessive daytime sleepiness. Sleep 2002;25:A91.
are not specific to a certain sleep disorder but rather spe- 10. Punjabi, NM, Bandeen-Roche K, Young T. Predictors of objective
cific to the sleepiness associated with the disorder. When sleep tendency in the general population. Sleep 2003;26:678-683.
treated adequately, sleepiness is rectified and the memory 11. Breslau N, Roth T, Rosenthal L, et al. Daytime sleepiness: an epi-
and cognitive deficits similarly improve.106 Results of sleep demiological study of young adults. Am J Public Health
deprivation studies in healthy normal patients support the 1997;87:1649-1653.
12. Carskadon MA, Dement WC. The multiple sleep latency test: what
relation between sleepiness and memory deficiency. Even does it measure? Sleep 1982;5:S67-S72.
modest reductions of sleep time are associated with cogni- 13. Bonnet MH, Arand DL. Sleepiness as measured by the MSLT varies
tive deficiencies.107 as a function of preceding activity. Sleep 1998;21:477-484.
Sleepiness also depresses arousability to physiological 14. Richardson G, Drake CL, Roehrs T, et al. Habitual sleep time
predicts accuracy of self-reported alertness. Sleep 2002;25:A145.
challenges: 24-hour sleep deprivation decreases upper 15. Horne JA, Baulk SD. Awareness of sleepiness when driving. Psy-
airway dilator muscle activity108 and decreases ventilatory chophysiology 2003;41:161-165.
responses to hypercapnia and hypoxia.109 In a canine model 16. Drake C, Roehrs T, Burduvali E, et al. Effects of rapid versus slow
of sleep apnea, periodic disruption of sleep with acoustic accumulation of eight hours of sleep loss. Psychophysiology
stimuli (i.e., sleep fragmentation, in contrast to sleep depri- 2001;38:979-987.
17. Kim H, Young T. Subjective sleepiness dimensions and correlates
vation) resulted in lengthened response times to airway in the general population. Sleep 2005;28:625-634.
occlusion, greater oxygen desaturation, increases in inspi- 18. Watson NF, Jacobsen C, Goldberg J, et al. Subjective and objective
ratory pressures, and surges in blood pressure.110 Depressed sleepiness in monozygotic twins discordant for chronic fatigue syn-
physiological responsivity due to sleepiness is clinically drome. Sleep 2004;27:973-977.
19. Pivik RT. The several qualities of sleepiness: psychophysiological
significant for patients with sleep apnea and other breath- considerations. In: Monk T, editor. Sleep, sleepiness and perfor-
ing disorders as they are all exacerbated by sleepiness. The mance. New York: John Wiley & Sons; 1991. pp. 3-37.
emerging data on sleepiness and pain threshold, cited 20. Friedman L, Bergmann BM, Rechtschaffen A. Effects of sleep
earlier, is also clinically significant in the management of deprivation on sleepiness, sleep intensity, and subsequent sleep in
both acute and chronic pain conditions. the rat. Sleep 1979;1:369-391.
21. Strijkstra AM, Beersma DGM, Drayer B, et al. Subjective sleepiness
Finally, life expectancy data directly link excessive sleep correlates negatively with global alpha (8-12Hz) and positively with
(not specifically sleepiness) and mortality. A 1976 study central frontal theta (4-8 Hz) frequencies in the human resting
found that men and women who reported sleeping more awake electroencephalogram. Neurosci Lett 2003;340:17-20.
52 PARTI / Section 1 Normal Sleep and Its Variations

22. Wu JC, Gillin JC, Buchsbaum MS, et al. The effect of sleep depriva- 51. Van den Hoed J, Kraemer H, Guilleminault C, et al. Disorders of
tion on cerebral glucose metabolic rate in normal humans assessed excessive somnolence: polygraphic and clinical data for 100 patients.
with positron emission tomography. Sleep 1991;14:155-162. Sleep 1981;4:23-37.
23. Starbuck VN, Kay GG, Platenberg RC. Functional magnetic reso- 52. Littner MR, Kushida C, Wise M, et al. Practice parameters for
nance imaging shows evidence of daytime sleepiness following clinical use of the Multiple Sleep Latency Test and the Maintenance
evening dosing with chlorpheniramine. J Allergy Clin Immunol of Wakefulness Test: an American Academy of Sleep Medicine
1998;101:408. report. Sleep 2005;28:113-121.
24. Lovblad KO, Bassetti C, Mathis J, et al. MRI of paramedian tha- 53. Axelsson J, Kecklund G, Akerstedt T, et al. Sleepiness and perfor-
lamic stroke with sleep disturbance. Neuroradiology 1997;39:693- mance in response to repeated sleep restriction and subsequent
698. recovery during semi-laboratory conditions. Chronobiology Inter-
25. Saper CB, Scammell TE. Hypothalamic regulation of sleep and nat 2008;25:297-308.
circadian rhythms. Nature 2005;437:1257-1263. 54. Roehrs T, Shore E, Papineau K, et al. A two-week sleep extension
26. Porkka-Heiskanen T, Strecker RE, McCarley RW. Brain site-spec- in sleepy normals. Sleep 1996;19:576-582.
ificity of extracellular adensosine concentration changes during 55. Roehrs T, Zorick F, Wittig R, et al. Efficacy of a reduced triazolam
sleep deprivation and spontaneous sleep: An in vivo microdialysis dose in elderly insomniacs. Neurobiol Aging 1985;6:293-296.
study. Neuroscience 2000;99:507-517. 56. Roehrs T, Zorick F, Sicklesteel J, et al. Excessive daytime sleepiness
27. Mignot E. A commentary on the neurobiology of the hypocretin/ associated with insufficient sleep. Sleep 1983;6:319-325.
orexin system. Neuropsychopharmacology 2001;25:S5-S13. 57. Manber R, Bootzin RR, Acebo C, et al. The effects of regularizing
28. Roehrs T, Zwyghuizen-Doorenbos A, Roth T. Sedative effects and sleep-wake schedules on daytime sleepiness. Sleep 1996;19:432-
plasma concentrations following single doses of triazolam, diphen- 441.
hydramne, ethanol and placebo. Sleep 1993;16:301-305. 58. Klerman EB, Dijk D. Interindividual vaiation in sleep duration and
29. Tashiro M, Mochizuki H, Iwabuchi K, et al. Roles of histamine in its associating with sleep debt in young adults. Sleep 2005;28:
regulation of arousal and cognition: functional neuroimaging of 1252-1259.
histamine H1 receptors in human brain. Life Sci 2002;72:409- 59. Viola AU, Archer SM, James LM. PER3 polymorphism predicts
414. sleep structure and waking performance. Curr Biol 2007;17:
30. Chiarello RJ, Cole JO. The use of psychostimulants in general 613-618.
psychiatry. Arch Gen Psychiatry 1987;44:286-295. 60. Iber C, Ancoli-Israel S, Chesson AL, Quan SF. The AASM Manual
31. Balkin TJ, Rupp T, Picchioni D, Wesensten NJ. Sleep loss and for the Scoring of Sleep and Associated Events: Rules, Terminol-
sleepiness: current issues. Chest 2008;134:653-660. ogy, and Technical Specifications. Westchester, Ill: American
32. Dinges DF, Orne MT, Whithouse WG, et al. Temporal placement Academy of Sleep Medicine; 2007.
of a nap for alertness: contributions of circadian phase and prior 61. Stepanski E. The effect of sleep fragmentation on daytime function.
wakefulness. Sleep 1987;10:313-329. Sleep 2002;25:268-276.
33. Hoddes E, Zarcone VP, Smythe H. Quantification of sleepiness: a 62. Zorick F, Roehrs T, Conway W, et al. Effects of uvulopalatopha-
new approach. Psychophysiology 1973;10:431-436. ryngoplasty on the daytime sleepiness associated with sleep apnea
34. Dement WC, Carskadon MA, Richardson G. Excessive daytime syndrome. Bull Eur Physiopathol Respir 1983;19:600-603.
sleepiness in the sleep apnea syndrome. In: Guilleminault C, 63. Lamphere J, Roehrs T, Wittig R, et al. Recovery of alertness after
Dement WC, editors. Sleep apnea syndromes. New York: Alan CPAP in apnea. Chest 1989;96:1364-1367.
R Liss; 1978. pp. 23-46. 64. Weaver TE, Maislin G, Dinges Bloxham T, et al. Relationship
35. Johns MW. Sleepiness in different situations measured by the between hours of CPAP use and achieving normal levels of sleepi-
Epworth Sleepiness Scale. Sleep 1994;17:703-710. ness and daily functioning. Sleep 2007;30:711-719.
36. Carskadon MA, Dement WC, Mitler MM, et al. Guidelines for 65. Bonnet MH. Performance and sleepiness as a function of the fre-
the multiple sleep latency test (MSLT): a standard measure of quency and placement of sleep disruption. Psychophysiology
sleepiness. Sleep 1986;9:519-524. 1986;23:263-271.
37. Punjabi NM, Bandeen-Roche K, Young T. Predictors of objective 66. Stepanski E, Lamphere J, Roehrs T, et al. Experimental sleep frag-
sleep tendency in the general population. Sleep 2002:26: mentation in normal subjects. Int J Neurosci 1987;33:207-214.
678-683. 67. Levine B, Roehrs T, Stepanski E, et al. Fragmenting sleep dimin-
38. Carskadon MA, Dement WC. Nocturnal determinants of daytime ishes its recuperative value. Sleep 1987;10:590-599.
sleepiness. Sleep 1982;5:S73-S81. 68. Ancoli-Israel S, Kripke D, Mason W, et al. Sleep apnea and noc-
39. Hartse KM, Roth T, Zorick FJ. Daytime sleepiness and daytime turnal myoclonus in a senior population. Sleep 1981;4:349-358.
wakefulness: the effect of instruction. Sleep 1982;5:S107-S118. 69. Carskadon MA, Brown E, Dement WC. Sleep fragmentation in the
40. Sullivan SS, Kushida CA. Multiple sleep latency test and mainte- elderly: relationship to daytime sleep tendency. Neurobiol Aging
nance of wakefulness test. Chest 2008;134:854-861. 1982;3:321-327.
41. Sangal RB, Thomas L, Mitler MM. Disorders of excessive sleepi- 70. Richardson GS, Carskadon MA, Orav EJ, et al. Circadian variation
ness: treatment improves ability to stay awake, but does not reduce of sleep tendency in elderly and young adult subjects. Sleep
sleepiness. Chest 1992;102:699-703. 1982;5:S82-S94.
42. Mitler MM, Hajdukovic R. Relative efficacy of drugs for the treat- 71. Monk TH. Circadian aspects of subjective sleepiness: a behavioral
ment of sleepiness in narcolepsy. Sleep 1991;14:218-220. messenger? In: Monk TH, editor. Sleep, sleepiness and perfor-
43. Bonnet MH, Arand DL. Arousal components which differentiate mance. New York: John Wiley & Sons; 1991. pp. 39-63.
the MWT from the MSLT. Sleep 2001;24:441-447. 72. Carskadon MA, Dement WC. Multiple sleep latency tests during
44. Carskadon MA, Dement WC. Effects of total sleep loss on sleep the constant routine. Sleep 1992;15:393-399.
tendency. Percept Mot Skills 1977;48:495-506. 73. Strogatz SH, Kronauer RE, Czeisler CA. Circadian pacemaker
45. Nicholson AN, Stone BM. Impaired performance and the tendency interferes with sleep onset at specific times each day: role in insom-
to sleep. Eur J Clin Pharmacol 1986;30:27-32. nia. Am J Physiol 1987;253:R172-R178.
46. Roehrs T, Kribbs N, Zorick F, et al. Hypnotic residual effects 74. Zulley J, Campbell SS. Napping behavior during spontaneous
of benzodiazepines with repeated administration. Sleep 1986;9: internal desynchronization: sleep remains in synchrony with body
309-316. temperature. Hum Neurobiol 1985;4:123-126.
47. Roehrs T, Burduvali E, Bonahoom A, et al. Ethanol and sleep loss: 75. Jacklet JW. The neurobiology of circadian rhythm generators.
A dose comparison of impairing effects. Sleep 2003;26:981-985. Trends Neurosci 1985;8:69-73.
48. Dinges DF, Pack F, Williams K, et al. Cumulative sleepiness, mood 76. Seidel WF, Roth T, Roehrs T, et al. Treatment of a 12-hour shift of
disturbance, and psychomotor vigilance performance decrements sleep schedule with benzodiazepines. Science 1984;22:1262-1264.
during a week of sleep restricted to 4-5 hours per night. Sleep 77. National Institutes of Health State-of-the Science Conference
1997;20:267-277. Statement on Manifestations and Management of Chronic Insom-
49. Roehrs T, Greenwald M, Roth T. Risk-taking behavior: effects of nia in Adults, June 13-15, 2005. Sleep 2005;28:1049-1057.
ethanol, caffeine, and basal sleepiness. Sleep 2004;27:887-893. 78. Zwyghuizen-Doorenbos A, Roehrs T, Lamphere J, et al. Increased
50. Roehrs TA, Hyde M, Blaisdell B, et al. Sleep loss and REM sleep daytime sleepiness enhances ethanols sedative effects. Neuropsy-
loss are hyperalgesic. Sleep 2006;29:145-151. chopharmacology 1988;1:279-286.
CHAPTER 4 Daytime Sleepiness and Alertness 53

79. Ashton H, Young AH. GABA-ergic drugs: exit stage left, enter stage editors. Sleep 1976. Basel, Switzerland: Karger; 1977. pp. 95-
right. J Psychopharm 2003;17:174-178. 100.
80. Cousins MS, Roberts DCS, de Wit H. GABAB receptor agonists 96. Broughton R, Ghanem Q, Hishikawa Y, et al. Life effects of
for the treatment of drug addiction: a review of recent findings. narcolepsy in 180 patients from North America, Asia and
Drug Alcohol Depend 2002;65:209-220. Europe compared to matched controls. J Can Sci Neurol 1981;8:
81. Nicholson AN, Stone BM. Antihistamines: impaired performance 299-304.
and the tendency to sleep. Eur J Clin Pharmacol 1986;30:27-32. 97. Findley LJ, Unverzagt ME, Suratt PM. Automobile accidents
82. Conway J, Greenwood DT, Middlemiss DN. Central nervous involving patients with obstructive apnea. Am Rev Respir Dis
actions of beta-adrenoreceptor antagonists. Clin Sci Mol Med 1988;138:337-340.
1978;54:119-124. 98. Drake C, Scofield H, Jefferson C, et al. MSLT defined sleepiness
83. Olanow CW, Schapira AHV, Roth T. Waking up to sleep episodes predicts verified automotive crashes in the general population. Sleep
in Parkinsons disease. Mov Disord 2000;15:212-215. 2007;30:A131 [abstract].
84. Clarenbach P. Parkinsons disease and sleep. J Neurol 2000;247: 99. Mitler MM, Carskadon MA, Czeisler CA, et al. Catastrophes, sleep,
IV20-IV23. and public policy: consensus report. Sleep 1988;11:100-109.
85. Mitler MM, Shafor R, Hajdukovich R, et al. Treatment of narco- 100. Mackie RR, Miller JC. Effects of hours of service, regularity of
lepsy: objective studies on methylphenidate, pemoline and protrip- schedules, and cargo loading on truck and bus driver fatigue. Wash-
tyline. Sleep 1986;9:260-264. ington, DC: US Government Printing Office; 1978. Technical
86. Newhouse PA, Belenky G, Thomas M, et al. The effects of Report 1765-F DOT-HS-5-01142.
d-amphetamine on arousal, cognition, and mood after prolonged 101. Dorrian J, Tolley C, Lamond N, et al. Sleep and errors in a group
total sleep deprivation. Neuropsychopharmacology 1989;2:153- of Australian hospital nurses at work and during the commute. App
163. Ergon 2008;39:605-613.
87. Bishop C, Roehrs T, Rosenthal L, et al. Alerting effects of methyl- 102. Baldwin DC, Daugherty SR. Sleep deprivation and fatigue in resi-
phenidate under basal and sleep-deprived conditions. Exp Clin Psy- dency training: results of a national survey of first- and second-year
chopharmacol 1997;4:344-352. residents. Sleep 2004;27:217-223.
88. Lumley M, Roehrs T, Asker D, et al. Ethanol and caffeine effects 103. Marcus CL, Loughlin GM. Effect of sleep deprivation on driving
on daytime sleepiness/alertness. Sleep 1987;10:306-312. safety in housestaff. Sleep 1996;19:763-766.
89. Roehrs T, Beare D, Zorick F, et al. Sleepiness and ethanol effects 104. Roehrs TA, Merrion M, Pedrosi B, et al. Neuropsychological func-
on simulated driving. Alcohol Clin Exp Res 1994;18:154-158. tion in obstructive sleep apnea syndrome (OSAS) compared to
90. Kilduff TS, Bowersox SS, Kaitin KI, et al. Muscarinic cholinergic chronic obstructive pulmonary disease (COPD). Sleep 1995;18:
receptors and the canine model of narcolepsy. Sleep 1986;9: 382-388.
102-106. 105. Rao V, Spiro J, Samus QM, et al. Insomnia and anytime sleepiness
91. Anderson KN, Pilsworth S, Sharples LD, et al. Idiopathic hyper- in people with dementia residing in assisted living: finding from the
somnia: a study of 77 cases. Sleep 2007:30:1274-1281. Maryland Assisted Living Study. Int J Ger Psychiat 2008;23:199-
92. Martinez-Rodriguez JE, Lin L, Iranzo A, et al. Decreased hypocre- 206.
tin-1 (Orexin-A) levels in the cerebrospinal fluid of patients with 106. Aguirre M, Broughton RJ, Stuss D. Does memory impairment
myotonic dystrophy and excessive daytime sleepiness. Sleep 2003; exist in narcolepsy-cataplexy? J Clin Exp Neuropsychol 1985;7:
26:287-290. 14-24.
93. Roth T, Rye DB, Borchert LD, et al. Assessment of sleepiness and 107. Blagrove M, Alexander C, Horne JA. The effects of chronic sleep
unintended sleep in Parkinsons disease patients taking dopamine reduction on the performance of cognitive tasks sensitive to sleep
agonists. Sleep Med 2003;4:275-280. deprivation. Appl Cogn Psychol 1994;9:21-40.
94. Arnulf I, Leu S, Oudiette D. Abnormal sleep and sleepiness in 108. Leiter JC, Knuth SL, Barlett D. The effect of sleep deprivation on
Parkinsons disearse. Curr Opin Neur 2008;21:472-477. activity of the genioglossus muscle. Am Rev Respir Dis 1985;
94a. Castriotta RJ, Atanasov S, Wilde MC, et al. Treatment of sleep 132:1242-1245.
disorders after traumatic brain injury. J Clin Sleep Med 2009; 109. White DP, Douglas NJ, Pickett CK, et al. Sleep deprivation and
5:137-144. control of ventilation. Am Rev Respir Dis 1983;128:984-986.
94b. Kaplan GB, Vasterling JJ, Vedak PC. Brain-derived neurotrophic 110. Brooks D, Horner RL, Kimoff RJ, et al. Effect of obstructive sleep
factor in traumatic brain injury, post-traumatic stress disorder, and apnea versus sleep fragmentation on responses to airway occlusion.
their comorbid conditions: role in pathogenesis and treatment. Am J Respir Crit Care Med 1997;155:1609-1617.
Behav Pharmacol 2010;21:427-437. 111. Kripke DF, Simons NR, Garfinkel L, et al. Short and long sleep
95. Guilleminault C, Carskadon M. Relationship between sleep and sleeping pills. Is increased mortality associated? Arch Gen Psy-
disorders and daytime complaints. In: Koeller WP, Oevin PW, chiatry 1979;36:103-116.
Acute Sleep Deprivation
Michael H. Bonnet
Abstract man. However, several new means of assessing sleep depriva-
tion effects, including brain scans, genetic assessment, and
Sleep deprivation is extremely common in modern society. animal models, show promise for a better understanding of
Sleep loss is accompanied by significant and increasingly sleep and sleep loss. Studies have also shown that high-
apparent alterations in mood, alertness, and performance. frequency periodic sleep fragmentation produces nonrestor-
This chapter reviews the behavioral effects of sleep depriva- ative sleep that results in a state of sleepiness and decreased
tion, including both sleep/circadian influences and arousal performance that is similar in many dimensions to sleepiness
system influences such as activity, light, noise, posture, moti- after total sleep deprivation. Recovery sleep after sleep loss
vation, and drugs. The effects of sleep loss are broad, and or sleep fragmentation shows a characteristic pattern of ele-
numerous systems are affected. Work showing similar changes vated slow-wave sleep (SWS) with elevated sensory thresholds
after alcohol ingestion provides one means of comparatively followed by elevated rapid eye movement (REM) sleep.
describing the effects of sleep loss. A number of relatively mild
physiologic changes accompany total sleep deprivation in

Sleep deprivation is both extremely common and critically Over a thousand studies of sleep deprivation have been
relevant in our society. As a clinical entity, sleep depriva- published during the past 10 years, and the resulting
tion is recognized by the diagnosis of insufficient sleep knowledge database has been remarkably consistent.
syndrome (International Classification of Diseases [ICD]- However, new techniques and increasingly sensitive tests
2, #307.44). As an experimental methodology, sleep depri- continue to add both theoretical and practical understand-
vation serves as a major tool in understanding the function ing of the impact of sleep loss. This review includes sec-
of sleep. A broad range of physiologic responses and tions on total sleep deprivation, sleep fragmentation, and
behavioral abilities have been examined after varying recovery from sleep deprivation.
periods without sleep, and lawful relationships have been
described. These relationships and the theory they repre-
sent are important in their own right, but the findings also TOTAL SLEEP DEPRIVATION
serve as an extensive guide to symptoms associated with The first published studies of total sleep loss date to 1894
insufficient sleep. Furthermore, methods developed to for puppies1 and 1896 for humans.2 The puppy study indi-
lessen the impact of sleep deprivation also serve as possible cated that prolonged sleep loss in animals could be fatal,
clinical treatments for disorders related to insufficient an idea reinforced by numerous, more recent animal
sleep or excessive sleepiness. studies. The human study included a range of physiologic
This chapter will review behavioral, physiologic, and and behavioral measurements and remains a model study.
theoretical implications of acute sleep deprivation. Chronic
partial sleep deprivation is examined in Chapter 6. Studies Behavioral Effects
of sleep deprivation can suffer from some common meth- The most striking effect of sleep loss is sleepiness, and this
odological problems that require consideration. The most can be inferred from subjective reports, the multiple sleep
important control issue is that one cannot perform a latency test (MSLT), electroencephalographic (EEG)
blinded sleep deprivation study. Both experimenter and change, or simply looking at the face of the participant.
subject motivation can have an impact on results, particu- The variables that determine the impact of sleep loss can
larly in the behavioral and subjective domains. Motivation be divided into four categories: sleep/circadian influences,
effects are frequently apparent near the end of sleep depri- arousal system influences, individual characteristics, and
vation studies (where performance improvement is some- test characteristics (Box 5-1).
times found) and may account for the difficulty in showing
early decrements. Animal studies are less susceptible to Sleep/Circadian Influences
subject expectation effects, but they may contain additional Sleep deprivation, like nutritional status, is a relative
elements of stress that may interact with sleep loss, so these concept. How an individual responds to sleep loss depends
studies may not be directly comparable with stress control on the prior sleep amount and distribution. Performance
conditions. In addition, almost all sleep loss experiments during a period of sleep loss is also directly dependent
involve more than simple loss of sleep. Maintenance of on the length of time awake and the circadian time, and
wakefulness usually includes upright posture, light, move- predictive models support these factors. Experiments
ment, cognition, and all the underlying physiologic pro- usually try to control prior wakefulness and sleep amount
cesses implied by these activities. The experimental setting by requiring normal nights of sleep before initiation of
itself is usually far from routine. Some studies have a sleep loss episode. Data from multiple regression analy-
attempted to control some of these factors during sleep ses of behavioral and EEG data during 64 hours of sleep
loss, but trying to control all variables in a single experi- loss3 suggest that time awake accounts for 25% to 30%
ment is daunting. of the variance in alertness, and that circadian time

CHAPTER 5 Acute Sleep Deprivation 55

Box 5-1 Determinants of the Impact of 1.3

Sleep Loss MSLT
1.1 DSST
Sleep/Circadian Influences
Prior sleep amount and distribution

Proportion of baseline
Length of time awake 0.9
Circadian time

Arousal Influences 0.7

Bright light 0.5
Posture 0.3
Motivation or interest
Drugs 0.1
Group effects

1: 0
4: 0
7: 0
10 0

19 0
22 0
Repeated exposure to sleep loss

4: 0
7: 0
10 0
13 0
16 00

1: 0

13 0
16 0
19 0




Individual Characteristics Time of day
Individual sensitivity Figure 5-1 Latency to stage 2 sleep (boxes) and number of
Personality and psychopathology correctly completed symbol substitutions in repeated 5-minute
test sessions (dots) over a period of 64 hours of sleep depriva-
Test Characteristics and Types tion, both expressed as a proportion of baseline values. (Data
Length of test from Bonnet MH, Gomez S, Wirth O, etal. The use of caffeine
Knowledge of results versus prophylactic naps in sustained performance. Sleep
1995;18:97-104.) DSST, digit symbol substitution task; MSLT
Test pacing multiple sleep latency test.
Proficiency level
Difficulty or complexity of test
Memory requirement
Executive function measures performing tasks provided transient reversal of some psy-
Subjective (versus objective) measures chomotor decrements resulting from sleep loss. However,
Electroencephalographic measuresmultiple sleep more ambitious studies comparing high activity and low
latency test (MSLT) activity that continued over 40 to 64 hours of sleep depri-
vation have shown no beneficial effects of exercise on
overall performance.5 Perhaps, arousing stimuli act for
only a discrete period of time that is decreased by increas-
accounts for about 6% of the variance. When prophylactic ing sleep loss.6 There may also be a trade-off between
naps of varying length were interjected early in a period production of arousal and production of physical fatigue.
of sleep loss, it was found that the prophylactic nap sleep
accounted for about 5% of the variance in alertness during B RIGHT LIGHT
the sleep deprivation period. In terms of reducing the Bright light can shift circadian rhythms. Some controversy
effect of sleep loss, the overall effect of increasing the exists concerning whether bright light can also act as a
prophylactic nap period was linear for additional sleep source of stimulation during a sleep-loss state to help to
amounts ranging up to 8 hours in length. Figure 5-1 maintain alertness. Two of five studies found that periods
displays the effects of time awake and the circadian rhythm of bright light immediately before sleep onset significantly
on objective alertness as measured by the MSLT and the increased sleep latencies. Other studies found improved
ability to complete correct symbol substitutions during nocturnal performance during bright light conditions,
64 hours of sleep loss. with elevated heart rate as a probable correlate.7

Arousal Influences N OISE

Environmental and emotional surroundings have a large Noise, despite complex and occasionally negative effects
impact on the course of a period of sleep loss. In early on the performance of well-rested individuals, may produce
stages of sleep deprivation, many intervening variables can small beneficial effects during sleep deprivation.8 It is gen-
easily reverse all measurable sleep loss decrements. These erally assumed that noise increases arousal level, and this
influences include activity, bright light, noise, tempera- may provide maximal benefit during sleep loss.
ture, posture, stress, and drugs.
A CTIVITY Although temperature variation is commonly used as an
A 5-minute walk immediately preceding MSLT evalua- acute stimulus to maintain alertness, little research sup-
tions had a large impact (about 6 minutes) on MSLT ports ambient temperature as a large modulator of alert-
values, which masked the impact of a 50% reduction of ness. One study has shown that heat (92F) was effective
nocturnal sleep (about a 2-minute reduction on MSLT) in improving performance during the initial minutes of a
and continued for at least 90 minutes.4 Exercise before vigilance task during sleep deprivation.9 However, another
56 PARTI / Section 1 Normal Sleep and Its Variations

study10 showed only a small decrease in subjectively rated sure to 2500-lux bright light had no impact beyond the
sleepiness for about 15 minutes after a car air-conditioner effect of caffeine alone on the maintenance of wakefulness
was turned on during simulated driving. test but did provide significant benefit above caffeine alone
on a vigilance task.17
P OSTURE Recent work has shown that armodafinil (the levorota-
One study has shown a significant increase in sleep latency, tory R-enantiomer of modafinil), which has a 10- to
as a measure of alertness, when subjects were asked to fall 14-hour half-life, is effective in maintaining alertness
asleep in the sitting position (60-degree angle) as opposed and performance throughout 1 night of sleep depriva-
to lying down.11 Such a difference could be accounted for tion.18 Results with armodafinil (200 mg) appeared to
by increasing sympathetic nervous system activity, which be similar to those with modafinil (200 mg) but may
occurs as one moves to the upright posture. have provided some additional benefit 11 to 13 hours
after administration.18
D RUGS Nicotine, infused intravenously at dosages of 0.25, 0.37
Many drugs have been studied in conjunction with sleep and 0.5mg after 48 hours of wakefulness, had no signifi-
loss, and an extensive review by the American Academy of cant impact on MSLT or psychomotor performance.19
Sleep Medicine has been published.12 Most studies have Cocaine (96mg), like amphetamine, did not improve per-
examined stimulants, including amphetamine, caffeine, formance in subjects before sleep loss. However, reaction
methylphenidate, modafinil, armodafinil, nicotine, and time and alertness, as measured by the Profile of Mood
cocaine. States, was improved after 24 and 48 hours of sleep loss.20
Numerous studies have shown that caffeine (dosages of Alcohol use has been found to consistently reduce alert-
200 to 600mg), modafinil (100 to 400mg), and amphet- ness.21 Subjects were tested with the MSLT and simulated
amine (5 to 20mg) can improve objective alertness and driving after 0.6 g/kg alcohol or placebo following normal
psychomotor performance for periods of time related to sleep or 4 hours of sleep. There was a significant main
dosage, half-life, and hours of total sleep deprivation. effect for condition, and MSLT latencies were 10.7, 6.3,
However, head-to-head studies often provide the clearest 6.1, and 4.7 minutes after placebo (normal sleep), placebo
comparison of compounds. One study13 examined alertness (4 hours sleep), ethanol (normal sleep), and ethanol (4
and response speed hourly for 12 hours during the first hours sleep), respectively. Similar results were found in
night of sleep deprivation after administration of modafinil the driving simulator, where there were three crashes
(100, 200, and 400mg) in comparison with caffeine all in the reduced-sleep-with-ethanol condition. One dif-
600mg. In that study, modafinil dosages of 200 and ficulty in assessing the magnitude of performance effects
400mg were shown to be equivalent to caffeine (600mg) associated with sleep loss is the lack of a clear standard
in maintaining response speed consistently above placebo of pathology for most measures. The fact that society has
levels for 12 hours. The same group compared modafinil established very specific rules for blood alcohol content
(400mg), caffeine (600mg), and dextroamphetamine with respect to driving has led to the use of impairment
(20mg) given just before midnight of the second night of associated with blood alcohol level as a standard reference
total sleep loss.14 In this study, caffeine and dextroamphet- for sleep deprivation as well. Several studies of alcohol
amine significantly improved response speed only at mid- use in direct comparison with sleep deprivation have
night, 2:00 am, and 4:00 am, whereas modafinil improved shown decrements on different tasks. Response speed on
response speed through 10 am. The decreased sensitivity the Mackworth task was reduced by approximately half a
to caffeine probably reflects both the half-life of caffeine second by 3:45 am (i.e., with sleep loss) and to a similar
and the increased sleep pressure from the second night of extent by a blood alcohol content (BAC) of 0.1%. Also,
deprivation. Side effects were fewest after modafinil at handeye coordination (in a visual tracking task) declined
400mg. However, the authors concluded that (1) these in a linear fashion during sleep loss and with increasing
stimulants all provided some benefits and had some associ- BAC, such that performance was equivalent at 3:00 am
ated costs, (2) caffeine, with efficacy, availability, and low to a blood alcohol level of 0.05% and equivalent at
cost, can be a first choice for alertness, and (3) modafinil, 8:00 am (after a full night of sleep loss) to a blood alcohol
with good efficacy and few side effects, would be a good level of 0.1%. In a third study,22 performance was
substitute if caffeine were ineffective (related to time measured in a driving simulator after alcohol use or sleep
course of available administration, tolerance, side effects, deprivation. After a night of sleep deprivation (at
or degree of sleep loss). 7:30 am), subjects averaged one off-road (i.e., vehicle
Effects of stimulants can be enhanced by the use of naps driving off the road) incident every 5 minutes. This same
or other sources of arousal. The beneficial effect of caffeine level of off-road driving was reached with a BAC of 0.08%.
(300mg) during periods of sleep loss was approximately These studies suggest that the changes in response speed,
equivalent to that seen after a 3- to 4-hour prophylactic visual tracking, and driving commonly found during the
nap before the sleep loss period.15 The combination of a first night of total sleep deprivation are equivalent to
4-hour prophylactic nap followed by 200mg of caffeine at changes associated with legal intoxication. Such metrics
1:30 am and 7:30 am resulted in significantly improved provide useful understanding of the consequences associ-
performance (remaining at baseline levels) compared with ated with short periods of sleep loss.
the nap alone, for 24 hours.16 The combination of naps and
caffeine was additive16 and superior to the provision of 4 M OTIVATION OR INTEREST
hours of nocturnal naps. The combination of 200mg of Motivation is relatively easy to vary by paying subjects and
caffeine administered at 10:00 pm and 2:00 am and expo- has therefore received attention. In one study, monetary
CHAPTER 5 Acute Sleep Deprivation 57

rewards for hits on a vigilance task and fines for false that the reliable changes in subjective alertness, objective
alarms23 resulted in performance being maintained at base- alertness, and performance are not related to one another.
line levels for the first 36 hours of sleep loss in the high- Performance consistently declined in the same subjects
incentive group. Performance began to decline during the when sleep loss was repeated, but decrements did not gen-
following 24 hours but remained significantly better than eralize across performance tasks. This has led some to
in the no incentive group. However, the incentive was speculate that different brain areas could be responsible for
ineffective in maintaining performance at a higher level different tasks. Some studies have begun to look for central
during the third day of sleep loss. Knowledge of results predictors of performance loss. Early functional magnetic
for example, the publication of daily test resultswas suf- resonance imaging (fMRI) studies showed that subjects
ficient to remove the effects of 1 night of sleep loss. In with higher levels of global brain activation (consistent at
another variation, simple knowledge that a prolonged both baseline and during sleep loss) maintained better per-
episode of sleep deprivation was going to end in a few formance on a working memory task,30 and a more recent
hours was sufficient for performance to improve by 30% study linked working memory during sleep loss with left
in a group of soldiers.24 frontal and parietal brain areas.31 Other studies have shown
that extroverts and caffeine-sensitive individuals were
G ROUP EFFECTS more sensitive to sleep loss.32 Such findings imply indi-
There are few studies, but interest is growing in how vidual trait ability to maintain higher levels of arousal in
groups perform during sleep deprivation. Such studies are specific brain areas or systems to help to maintain perfor-
difficult because groups of individuals interact in many mance during sleep loss.
ways. One early study suggested that, if all group members In another approach, over 400 potential subjects were
were working at a similar task, greater deficits were seen screened genetically, and groups were formed on the basis
as sleep deprivation progressed, compared with individual of the PER3 polymorphism (PER3[5/5] versus PER3[4/4])
work. However, a more recent study that distributed work before a 40-hour constant routine. The PER3(5/5) group
so that each individual added a unique component to task appeared sleepier by all measures (significantly shorter
completion found that the deficits that accumulated during sleep latency, more slow-wave sleep (SWS), greater slow
sleep deprivation were less pronounced in the group task.25 eye movements during sleep loss, and significantly worse
performance during sleep loss, particularly on executive
R EPEATED PERIODS OF SLEEP LOSS tasks done early in the morning [0600 to 0800]).33,34
Studies of repeated episodes of sleep loss have agreed that
the magnitude of performance loss increases as a function P ERSONALITY AND PSYCHOPATHOLOGY
of the number of exposures to sleep loss.26 Increasingly Mood changes, including increased sleepiness, fatigue,
poor performance may be secondary to decreased motiva- irritability, difficulty in concentrating, and disorientation,
tion or to familiarity with the sleep deprivation paradigm are commonly reported during periods of sleep loss. Per-
(resulting in decreased arousal). ceptual distortions and hallucinations, primarily of a visual
nature, occur in up to 80% of normal individuals, depend-
Individual Characteristics ing on work load, visual demands, and length of depriva-
The impact of sleep loss on a given individual depends on tion.35 Such misperceptions are normally quite easy to
characteristics that each participant brings to the sleep loss differentiate from the primarily auditory hallucinations of
situation. For example, age and personality represent dif- a schizophrenic patient, but normal individuals undergo-
ferences in physiologic or psychological function that may ing sleep loss may express paranoid thoughts. Two percent
interact with the sleep loss event. of 350 individuals sleep deprived for 112 hours experi-
enced temporary states resembling acute paranoid schizo-
A GE phrenia. Some predisposition toward psychotic behavior
Tests of performance and alertness in older subjects under- existed in individuals who experienced significant paranoia
going sleep loss reveal a decrease in performance and alert- during sleep loss, and the paranoid behavior tended to
ness similar to that seen in younger individuals. If anything, become more pronounced during the night, with partial
older men had a smaller decrease in psychomotor perfor- recovery during the day and disappearance after recovery
mance ability at nocturnal times during sleep loss27,28 than sleep. In a review of the area, Johnson concluded, Each
younger men. Older individuals perform more poorly than subjects response to sleep loss will depend on his age,
young adults on a broad range of tasks, but, because of physical condition, the stability of his mental health,
decreased amplitude of the circadian body temperature expectations of those around him, and the support he
rhythm, this relationship may not be maintained across the receives.36, p. 208 At a more general level, normal adults
night or during sleep loss.27 The same flattened curve asso- undergoing sleep deprivation typically express some
ciated with lower temperatures and decreased performance increase in somatic complaints, anxiety, depression, and
during the day also produces relatively elevated tempera- paranoia that do not reach clinical levels.37
tures that could be related to improved performance at In view of the commonly reported effects of sleep depri-
night. vation, it seems quite unusual that one would seek to treat
depression by sleep deprivation. However, sleep depriva-
S ENSITIVITY TO SLEEP LOSS tion has been used as a successful treatment for depression
A number of studies have now demonstrated consistent in 40% to 60% of cases for over 30 years. Imaging studies
individual differences in both alertness and performance have shown that depressed patients have elevated metabo-
during sleep deprivation.29 The studies have also shown lism in the prefrontal cortex and ventral anterior cingulate
58 PARTI / Section 1 Normal Sleep and Its Variations

cortex (possibly related to reduced transmission of dopa- paced, errors occur if items are presented during lapses
mine and serotonin) that is normalized by sleep depriva- in attention.
tion.38 One theory proposes that sleep deprivation is more
effective in depressed patients with high levels of activation P ROFICIENCY LEVEL
or high central noradrenergic activity because it limits the Sleep loss is likely to affect newly learned skills more than
effects of chronic hyperarousal. As a result, patients felt well-known activities, as long as arousal level remains con-
tired but also had improved mood and energy. stant. For example, in a study of the effects of sleep loss
on doctors in training, significant performance decrements
Test Characteristics and Types were found in postgraduate year (PGY)-1 surgical resi-
Two meta-analyses of subtopics of sleep deprivation have dents but not in PGY-2 to -5 surgical residents.42
been published.39 Both analyses indicated that sleep depri-
vation has a significant impact on psychomotor perfor- D IFFICULTY OR COMPLEXITY
mance. In general, longer periods of sleep loss had greater Performance on simple tasks such as monitoring a light on
impact on performance, and decrements in speed of per- a control panel (on/off) declines less than performance on
formance were greater than decreases in accuracy. Also, more complex tasks such as mental subtraction43 during
mood measures were more sensitive than cognitive tasks, sleep loss. Task difficulty can also be adjusted by increasing
which were more sensitive than motor tasks,39 during sleep the speed at which the work must be performed. When 2
loss. Therefore, the measured response to sleep depriva- seconds was allowed to complete mental arithmetic prob-
tion is critically dependent on the characteristics of the test lems, no significant performance decline was found after 2
used. To some extent, the type of test has also been used nights of sleep loss, but when the rate of presentation was
to infer specific brain area dysfunction. Sleep-deprived increased to 1.25 seconds, significant performance decline
individuals appear most sensitive to the following was found.
L ENGTH OF TEST Impairment of immediate recall for elements placed in
Individuals undergoing sleep loss can usually rally momen- short-term memory is a classic finding in sleep depriva-
tarily to perform at their nonsleep-deprived levels, but tion studies. Because subjects are usually required to
their ability to maintain that performance decreases as write down each item as presented, the observed decre-
the length of the task increases. For example, subjects ments, which can usually be seen after 1 night of sleep
attempted significantly fewer addition problems than loss, do not result from impaired sensory registration
baseline after 10 minutes of testing following 1 night of of items. Observed decrements may result from decreased
sleep loss but reached the same criterion after 6 minutes ability to encode,44 from an increasing inability to rehearse
of testing following the second night of sleep loss. It old items (due to lapses) while the items are being pre-
took 50 minutes of testing to show a significant decrease sented, or from a combination of memory effects with
in percentage of correct problems after 1 night of sleep reduced ability to respond. Deficits have been shown in
loss, and 10 minutes of testing to reach that criterion both explicit/declarative memory and procedural/implicit
after the second night.40 It is frequently difficult to show memory.44
reliable differences during short-term sleep loss from
almost any test that is shorter than 10 minutes in dura- E XECUTIVE FUNCTION
tion. Momentary arousal, even as minor as an indication Increasing behavioral and physiologic data implicate loss
that 5 minutes remained on a task, was sufficient to reverse of function in prefrontal brain areas during sleep loss.
75% of the decrement accumulated over 30 minutes of Prefrontal areas are heavily involved in divergent thinking,
testing. temporal memory (planning, prioritization, organization),
and novelty. Numerous studies have shown deficits on
K NOWLEDGE OF RESULTS these tasks during sleep loss. One study45 has shown
Immediate performance feedback, possibly acting through decreased ability to make emotional judgments after total
motivation, has been shown to improve performance sleep loss. Another aspect of prefrontal behavior is related
during sleep deprivation.41 Simply not giving knowledge to risk taking. Studies have shown a shift toward accepting
of results to subjects with normal sleep doubled their short-term rewards even when long-term consequences
number of very long responses (gaps) on a serial-reaction were more severe after sleep loss.46
time test. One night of total sleep loss increased the
number of gaps by 9.3 times the baseline level, but provi- S UBJECTIVE (VERSUS OBJECTIVE) MEASURES
sion of immediate knowledge of results decreased the Measures of mood such as sleepiness, fatigue, and ability
number of gaps back to baseline levels.41 to think or concentrate are inversely correlated with per-
formance and body temperature during sleep loss. Mood
T EST PACING changes occur early, are easy to measure, and are
Self-paced tasks are usually more resistant to the effects prominent.
of sleep loss than tasks that are timed or in which items
are presented by the experimenter. In a self-paced task, E EG MEASURES
the subject can concentrate long enough to complete Clear EEG changes are seen during sleep loss (see Neu-
items correctly and not be penalized for lapses in attention rologic Changes, later). The MSLT, a standard test devel-
that occur between items. When tasks are externally oped as an objective measure of sleepiness, was validated,
CHAPTER 5 Acute Sleep Deprivation 59

in part, by being shown to be sensitive to several types of loss were usually accompanied by a slowing of the EEG54
partial and total sleep loss.47 That the MSLT is more sensi- that was labeled a microsleep.
tive than psychomotor tasks can be seen in Figure 5-1,
which displays performance changes in terms of the Imaging Studies
number of symbol substitutions correctly completed in Global decreases in brain activation correlated with
5-minute test periods and MSLT data.15 increasing sleep loss have been found using positron emis-
sion tomography (PET). Larger decreases were found in
Summary prefrontal, parietal, and thalamic areas.55 Several fMRI
Tasks most affected by sleep loss are long, monotonous, studies have examined the activity in prefrontal cortex and
without feedback, externally paced, newly learned, and parietal lobes after sleep loss56 that was measured while
have a memory component. One example of a task con- subjects performed various tasks. In some studies of verbal
taining many of these elements is driving, which was dis- tasks, activity in these areas increases with task difficulty
cussed earlier in reference to the effects of alcohol. Since and after sleep loss as long as performance level is main-
1994, more than 20 studies have examined the impact of tained, which has been interpreted as representing
reduced sleep on various measures of driving ability or increased effort after sleep loss.56 However, studies that
safety. One study,48 for example, found that 49% of medical have found declines in performance or examined subjects
residents who worked on call and averaged 2.7 hours of with poor performance after sleep loss have reported
sleep reported falling asleep at the wheel (90% of the epi- decreases in parietal activation during the performance.31,57
sodes were after being on call). The residents also had 67% These fMRI results suggest that imaging patterns could
more citations for moving violations and 82% more car predict performance deficits during sleep loss.
accidents than the control group.48
Clinical EEG
Although the neurologic changes associated with signifi-
PHYSIOLOGIC EFFECTS OF SLEEP cant sleep loss are relatively minor in normal young adults,
DEPRIVATION sleep loss has repeatedly been shown to be a highly activat-
Physiologic changes that occur during sleep loss can be ing stress in individuals suffering seizure disorders, perhaps
categorized into neurologic (including EEG), autonomic, by reduction of central motor inhibition.58 Using a period
genetic, biochemical, and clinical changes. Physiologic and of sleep loss as a challenge to elicit abnormal EEG events
biochemical effects of sleep deprivation were extensively is currently a standard neurologic test.58
reviewed by Horne.49
Autonomic Changes
Neurologic Changes In humans, autonomic changes, even during prolonged
Although it is easy to identify a sleep-deprived individual periods of sleep loss, are relatively minor. Individual studies
by appearance and to demonstrate obvious behavioral have reported either increases or decreases in systolic
changes, measurable neurologic changes during sleep loss blood pressure, diastolic blood pressure, finger pulse
are relatively minor and quickly reversible. In extended volume, heart rate, respiration rate, and tonic and phasic
sleep loss studies (205 or more hours), mild nystagmus, skin conductance. However, the majority of 10 to 15
hand tremor, intermittent slurring of speech, and ptosis studies have reported no change in these variables during
have been noted.50 Sluggish corneal reflexes, hyperactive sleep loss in humans.49 It has been suggested that these
gag reflex, hyperactive deep tendon reflexes, and increased variable findings could be explained in part by measure-
sensitivity to pain were reported after deprivation that is ment circumstances. Those studies that have had more
more extensive. All of these changes reversed immediately strict activity controls and have made measurements from
after recovery sleep. recumbent subjects have been more likely to find evidence
Sleep loss is consistently accompanied by characteristic for decreased or no change in activation, whereas studies
EEG changes.51 In careful studies, subjects have been of sitting or more active participants have tended to find
required to stand or to be involved in tasks in an attempt increases in these parameters.59
to stabilize arousal level. Several studies have reported a There may be about a 20% reduction in response to
generally linear decrease in alpha activity during sleep loss. hypoxia and hypercapnia60 during sleep deprivation.
Subjects were unable to sustain alpha activity for longer However, this reduction was suggestive of a transient set-
than 10 seconds after 24 hours of sleep loss, and this ability point change rather than system failure. Sleep deprivation
continued to decline to 4 to 6 seconds after 72 hours, and has been associated with small decreases in forced expira-
1 to 3 seconds after 120 hours of sleep loss.52 After 115 tory volume in 1 second (FEV1) and forced vital capacity
hours of sleep loss, eye closure failed to produce alpha (FVC) in patients with pulmonary disease.61 Both a study
activity. In another study, in which individuals were of infants61a and one of adults62 have shown more apneic
recorded standing with their eyes closed, the percentage events61a and longer apneic events after sleep loss. Brooks
of time spent with an alpha pattern in the EEG decreased and colleagues63 have shown that apneas become longer as
from 65% in the early deprivation period to about 30% a function of the sleep fragmentation produced by the
after 100 hours of sleep loss.53 Delta and theta activity in apneas (as opposed to the respiratory pathology).
the waking EEG were increased from 17% and 12% of the Several studies in humans have found a small overall
time to 38% and 26% of the time, respectively.53 The decrease (0.3 to 0.4C) in body temperature during
increase in delta activity was most pronounced in frontal sleep loss.2,64 Changes in thermoregulation have been
areas in younger subjects. Performance errors during sleep described as heat retention deficits. Much larger changes
60 PARTI / Section 1 Normal Sleep and Its Variations

in thermoregulation producing huge increases in energy deprivation progresses, one gene, for the enzyme aryl-
expenditure have been found in rats after longer periods sulfotransferase (AST), showed stronger induction as a
of sleep deprivation (see Chapter 28).65 function of length of sleep deprivation. AST induction
No sleep deprivationrelated changes in whole-body could reflect a homeostatic response to continuing central
metabolism were found at normal temperatures and in a noradrenergic activity during sleep loss,76 and this might
cold-stress situation.66 This finding in humans is of par- imply a role for sleep in reversing activity of brain cat-
ticular interest because a series of elegant studies in rats echolaminergic systems.77 In another approach, a gene
has shown that after a week of sleep loss, metabolic levels named Sleepless was identified as required for sleep in
are greatly increased, increased food consumption is Drosophila. Flies with significant reduction in Sleepless
accompanied by significant weight loss, and significant dif- protein had reduced sleep and sleepiness before and after
ficulty with thermoregulation is apparent.65 Several studies sleep deprivation.78
have examined aspects of brain metabolism in animals
during short periods of sleep deprivation. Direct measures Clinical Changes
of brain metabolic rate were not different after a short Immune Function
period of sleep deprivation,67 although several related A number of studies in humans have examined various
enzymes did differ. However, some of the noted differ- aspects of immune function after varying periods of partial
ences could have been related to stress rather than sleep or total sleep loss (see Chapter 25). Several studies have
deprivation. found decreases in natural killer (NK) cell numbers after
short periods of sleep deprivation,79 but increases after
Biochemical Changes longer periods of sleep loss. Some studies have shown
Several studies (10 or more for some variables) have exam- increases in interleukin (IL)-179 and IL-680 during total
ined various biochemical changes in humans during sleep sleep loss. In general, immune function studies are difficult
loss. There is generally no significant change in cortisol,68 to compare because parameters measured, time and
adrenaline and related compounds, catecholamine number of blood draws, and degree of sleep deprivation
output,66,69 hematocrit,70 plasma glucose,70 creatinine,66 or vary across studies.
magnesium66 during sleep loss. At a more macroscopic level, one study reported the
Results from analyses of blood components largely par- development of respiratory illness or asthma in three sub-
allel the results found in urine components. None of the jects after a 64-hour sleep deprivation protocol,81 whereas
adrenal or sex hormones (including cortisol, adrenaline, another reported no incidence of illness after a similar
noradrenaline, luteinizing hormone, follicle-stimulating protocol.79 In longer studies involving strenuous exercise
hormone, variants of testosterone, and progesterone) rises and other factors along with sleep loss, increased infection
during sleep deprivation in humans.68 Some of these hor- rates are reported about 50% of the time.82
mones actually decreased somewhat during sleep loss, One animal study has suggested that mice immunized
perhaps as a result of sleepiness and decreased physiologic against a respiratory influenza virus responded to that virus
activation. Thyroid activity, as indexed by thyrotropin, as if they had never been immunized, only when exposed
thyroxine, and triiodothyronine, was increased, probably after sleep loss.83 However, in another study,84 total sleep
as a result of the increased energy requirements of con- loss actually slowed the progression of a viral infection in
tinuous wakefulness.71 Studies appear about equally mice. An extensive study of sleep loss in rats (7 to 49 days)
divided between those showing an increase in melatonin was unable to show significant changes in spleen cell
and no change in melatonin during sleep deprivation.72 numbers, mitogen responses, or in vivo or in vitro splenic
A finding of decreased melatonin in young adults after antibody-secreting cell responses.85
sleep deprivation suggested that earlier findings of
increased melatonin may have been related to lack of Pain
control for posture, activity, and light.73 Most studies have Increased sensitivity to pain has been an incidental finding
concluded that there is no significant change in hematocrit in sleep deprivation research for many years, but 10 studies
levels,66 erythrocyte count, or plasma glucose during total have now made specific pain measurements in several
sleep deprivation in humans.70 As would be expected, conditions of partial, sleep stage, or total sleep deprivation.
hormones such as noradrenaline, prolactin, ghrelin, and Initial studies linked increased pain to SWS but not rapid
growth hormone, which are dependent on sleep for their eye movement (REM) deprivation. Later studies showed
circadian rhythmicity or appearance, lose their periodic that total sleep deprivation decreased pressure pain or
pattern of excretion during sleep loss (see reference 74 heat pain tolerance. However, the most recent studies86,87
for review). Rebounds in growth hormone and adreno- found effects for REM-stage deprivation not found earlier,
corticotropic hormone (ACTH) during recovery sleep are and they split on the effectiveness of total sleep depriva-
seen after sleep loss or SWS deprivation.75 tion. One study,87 which found increased pain sensitivity
after disturbed sleep compared with reduced sleep, sug-
Gene Studies gested that all of the pain findings associated with sleep-
A recent review of gene expression has described a stage deprivation may actually have been caused by the
number of changes that occur during wakefulness and sleep fragmentation necessary to produce sleep-stage
extended wakefulness76 (see Chapter 15). A number of deprivation. Another study has reported increased pain
genes expressed during wakefulness that regulate mito- sensitivity to esophageal acid perfusion, specifically in
chondrial activity and glucose transport probably reflect patients with gastroesophageal reflux disease (but not con-
increased energy use while awake. However, as sleep trols) after sleep reduced to 3 hours or less for 1 night.88
CHAPTER 5 Acute Sleep Deprivation 61

This suggests that individual differences could also play

a role in the modulation of pain. Animal studies have SLEEP FRAGMENTATION
replicated findings of increasing pain sensitivity after REM Sleep is a time-based cumulative process that can be
deprivation and showed that pain sensitivity remained impeded both by deprivation and by systematic distur-
elevated even after low dosages of morphine and 24 hours bance. A number of studies have shown that very brief
of recovery sleep.89 periodic arousals from sleep reduce the restorative power
of sleep and leave deficits similar to those seen after total
Weight Control and Insulin sleep deprivation.94
There are numerous reports of increased sympathetic
activity, impaired glucose tolerance, and weight gain asso- Experimental Sleep Fragmentation
ciated with chronic partial sleep deprivation (see Chapter Many studies have examined the relationship between
6), but, remarkably, these effects have not been replicated various empiric schedules of sleep fragmentation and
following total sleep deprivation. One study, in a small residual sleepiness on the following day. Data from eight
group of subjects, before and after 1 night of total sleep studies are plotted in Figure 5-2. There is a strong rela-
deprivation found no increase in cortisol, blood glucose, tionship (r = .775, P < .01) between rate of fragmentation
insulin, ACTH, catecholamines, or lactate.90 This study (plotted as minutes of sleep allowed between disturbances)
did report a decrease in basal glucagon, possibly linked and decrease in sleep latency on the following day as mea-
to pancreatic islet secretion and increased hunger. From sured by MSLT.94 As expected, increased sleepiness after
such reports, it is unclear whether the results reported sleep fragmentation was also associated with decreased
from chronic partial sleep deprivation are related to psychomotor performance on a broad range of tasks,95 and
increased or chronic stress or whether time of sampling with degraded mood.94
may play a role. Studies have been carefully designed to produce brief
EEG arousals or even nonvisible EEG sleep disturbance,
Exercise with the result that there are few96 in standard sleep EEG
Effects of sleep loss on the ability to perform exercise are parameters despite the periodic sleep fragmentation. With
subtle. Animal studies have consistently shown that sleep preservation of normal-EEG sleep amounts, participants
deprivation decreases spontaneous activity by up to 40%,91 were still significantly sleepier on the day after sleep
but most human studies have focused on maximal exercise fragmentation. In another approach, fragmentation rates,
ability, where large differences as a function of sleep consolidated sleep periods, and SWS amounts were experi-
loss are more difficult to demonstrate. For example, one mentally varied in participants in an attempt to tease
study92 reported a 7% decrease in maximum oxygen out sleep stage versus fragmentation effects, with similar
uptake ( V O 2max ) during 64 hours of sleep loss. This
change was not associated with heart rate, respiratory
exchange ratio, or blood lactate, which remained
unchanged. Recovery from exercise may be slowed by 1.1
sleep loss. Studies are evenly divided between claims that
the amplitude of the circadian rhythm of temperature is 1.0
increased, decreased, or unchanged during sleep loss.49
Proportion of baseline MSLT

A large number of studies have reported autonomic, bio- 0.8
chemical, and immune function variables during sleep loss.
Many of the older studies were based on single observa- 0.7
tion points before and after sleep loss. Many studies suffer
from poor activity controls (use of activity to maintain 0.6
wakefulness may produce or mask changes in underlying
variables of interest). More recent studies have been able
to make use of sampling as often as once per hour and
have begun to consider circadian and activity effects. For
example, NK cell numbers were increased93 during the
night when subjects remained awake (compared with a
sleep control) but were then decreased on the following
0 2 4 6 8 10 12 14 16 18 20
afternoon, with the result that numbers averaged across
the entire study were the same in sleep loss and baseline Length of sleep allowed (min)
conditions. This means that a study could find an increase,
Figure 5-2 Proportion of baseline multiple sleep latency test
a decrease, or the same number of NK cells based on (MSLT) value (i.e., sleep latency after fragmentation nights
the time of sampling. In a similar manner, it has been divided by sleep latency after baseline) in eight sleep fragmen-
found that IL-6 was decreased during a night of sleep tation studies (two separate fragmentation conditions were
deprivation compared with the sleep control but increased identified in three studies), plotted as a function the amount of
sleep time allowed until disturbance (e.g., 2 means that sub-
compared with control during the next day, so that jects were aroused briefly after each 2 minutes of sleep). (From
numbers averaged across the entire study were, again, Bonnet MH, Arand DL. Clinical effects of sleep fragmentation
about the same. versus sleep deprivation. Sleep Med Rev 2003;7:297-310.)
62 PARTI / Section 1 Normal Sleep and Its Variations

conclusionsresidual sleepiness was more related to the primarily the result of the associated sleep fragmenta-
sleep fragmentation than to sleep-stage parameters.94 tion.63, p. 1609
Other studies have directly compared the impact of rela- Other clinical studies have documented that the number
tively high rates of sleep fragmentation (usually distur- of brief arousals is significantly correlated with the magni-
bance every 1 to 2 minutes) with the effect of total sleep tude of daytime sleepiness in groups of patients.94 Tradi-
deprivation in the same study. In one study, profiles of tional sleep-stage rebounds (see Recovery Sleep, next) are
cortisol and ACTH were similar during total sleep depri- seen when the pathology is corrected. After effective treat-
vation and sleep fragmentation.97 In other studies,94 MSLT ment of sleep apnea and the corresponding decrease in
was decreased to similar low values after both total sleep frequency of arousals during sleep, alertness was improved
deprivation and high-frequency sleep fragmentation. A as measured by either MSLT or reduction in traffic acci-
significant increase in apnea-hypopnea index was found dents.101 There are many other instances of sleep fragmen-
after both sleep fragmentation and sleep deprivation, and tation as a component in both medical illnesses (such as
this increase was actually greater after sleep fragmenta- fibrositis, intensive-care-unit syndrome, chronic move-
tion.94 These findings of similar impacts on hormones, ment disorders, and chronic pain disorders) and life
respiratory parameters, psychomotor performance, and requirements (infant care, medical residents). Some of
objective sleepiness after similar periods of sleep depriva- these impositions may not produce the critical number of
tion and sleep fragmentation indicate that there is much arousals required for significant decrements in the apnea
in common between the high-frequency sleep fragmenta- patients and in sleep fragmentation studies. However,
tion and total sleep deprivation. Clearly, the restorative most of these situations are a combination of chronic
function of sleep is impaired by high rates of sleep frag- partial sleep loss and chronic sleep fragmentation.
mentation. However, as indicated by Figure 5-2, the impact
of periodic sleep fragmentation decreases rapidly as the
intervals between arousals increase, and this may imply RECOVERY SLEEP
that normal restoration during sleep requires periods of Sleep is all that is required to reverse the effects of sleep
consolidated sleep of 10 to 20 minutes.94 Recovery sleep deprivation in almost all circumstances. The EEG charac-
following high rates of sleep fragmentation is characterized teristics of recovery sleep depend on the amount of prior
by rebounds of SWS and REM sleep like that seen after wakefulness and the circadian time. These effects have
total sleep deprivation. In addition, recovery sleep after been successfully modeled (see Chapter 37).
sleep fragmentation and sleep deprivation is notable for
decreased arousals.98 Performance Effects
A broad range of physiologic indices have been exam- Several efforts have been made to assess recovery of per-
ined as possible measures of sleep fragmentation.94 A formance after sleep deprivation. It is commonly reported
number of respiratory, cardiac, and alternative EEG mea- that recovery from periods of sleep loss of up to 10 days
sures have been examined, with the general conclusion and nights is rapid and can occur within 1 to 3 nights.
that most of these physiologic measures, including tradi- Several studies have reported recovery of performance
tional arousals, are moderately correlated with daytime after a single night (usually 8 hours) of sleep following
sleepiness. However, much empirical work needs to be anywhere from 40 to 110 hours of continuous wakeful-
done to determine the extent to which these other physi- ness.5 Such experiments suggest that an equal amount of
ologic measures are simply correlates of EEG arousals sleep is not required to recover from sleep lost. However,
rather than new measures of sleep continuity. A physio- sleep deprivation itself was typically the main concern of
logic event more specific than the EEG arousal remains these studies and, therefore, recovery was given minimal
to be identified. attention.
New animal models of sleep fragmentation in rats have One study specifically examined the rate of performance
revealed decreases in hippocampal neurogenesis and recovery during sleep in young adults, normal older sub-
increases in basal forebrain adenosine to levels found after jects, and insomniacs after 40- and 64-hour sleep loss
similar amounts of total sleep deprivation.99 periods.27 Participants were awakened from stage 2 sleep
for 20-minute test batteries approximately every 2 hours
Sleep Disorders and Fragmentation during baseline and recovery nights. Therefore, it was pos-
The earliest study of sleep fragmentation in dogs docu- sible to follow the time course of return to baseline per-
mented impaired arousal responses to hypercapnia and formance during recovery sleep in the three groups. In
hypoxia during sleep.100 Further studies following dogs normal young adults, reaction time returned to levels not
with experimentally produced apnea or sleep fragmenta- significantly less than baseline after 4 hours of sleep during
tion for periods of more than 4 months showed that recovery sleep following 40 hours of sleep loss. However,
both the sleep fragmentation procedure and the experi- reaction time remained significantly slower than baseline
mentally produced apnea produced increased time to in young adults throughout the first night of recovery sleep
arousal as well as greater oxygen desaturation, greater (including the postsleep morning test) following 64 hours
peak inspiratory pressure, and greater surges in blood of sleep loss. In contrast, reaction time in both older
pressure in response to airway occlusion. It was concluded normal sleeper and insomniac groups was significantly
that the sleep fragmentation alone was responsible for slower than baseline at 5:30 am but had returned to base-
the sleepiness symptoms associated with sleep apnea, line levels by 8:00 am after the first recovery night after 64
and the changes in the acute responses to airway occlu- hours of sleep loss. The young adults not only recovered
sion resulting from OSA (obstructive sleep apnea) are more slowly from sleep loss on the initial recovery night
CHAPTER 5 Acute Sleep Deprivation 63

Table 5-1 Effects of Sleep Loss on Stages of Recovery Sleep




Sleep latency 0.38 (0.09) 0.22 0.14

Wake time 0.44 (0.19) 0.51 (0.11) 0.13 0.48
Stage 1 0.42 (0.12) 0.59 (0.14) 0.61 0.68 0.98 0.60 0.56 0.52
Stage 2 0.87 (0.10) 0.95 (0.07) 1.06 1.08 1.38 0.99 1.07 1.20
Stage 3 0.98 1.32 (0.25) 1.14 1.16 2.36 3.00
Stage 4 2.40 2.06 (0.45) 1.35 1.12 7.00 5.25
Slow-wave 1.53 (0.11) 1.56 (0.23) 1.23 1.15 1.37 1.52 2.56 3.30
sleep stage
REM sleep 0.89 (0.13) 1.04 (0.09) 0.84 0.78 1.26 1.03 0.26 0.92
REM sleep 1.01 (0.20) 0.77 (0.20) 2.2 2.0 1.13 1.26 0.35 0.96

The values presented in this table are the mean percentages of baseline levels of the indicated sleep stages for the indicated groups
during the first sleep recovery night. Where sufficient studies were available, the standard deviation around the mean percentage is also
given (SD). For example, on their recovery night after 1 night of sleep loss, young adults have a sleep latency that is 38% 9% of their
baseline sleep latency. (See text for references.)

but also had some decrease in their reaction times that secondarily in some less robust differences found in small
extended into the second recovery night. This result is groups of long and short sleepers. These latter findings
consistent with other data showing that older subjects had might be related to differential sleep-stage distributions
daytime MSLT values at baseline levels following sleep secondary to long or short sleep times.
loss and a single night of recovery sleep, whereas shorter- On the first recovery night after total sleep loss, there
than-normal latencies continued in young adults.102 is a large increase in SWS over baseline amounts.70,111 As
would be expected, wake time and stage 1 sleep are usually
EEG Effects reduced. Stage 2 and REM sleep may both be decreased
A large number of studies have reported consistent effects on the first recovery night after 64 hours of sleep loss,27,47
on sleep EEG when totally sleep deprived individuals are at least in young adults, as a function of increased SWS.
finally allowed to sleep. If undisturbed, young adults typi- In older normal sleepers and insomniacs, there is less abso-
cally sleep only 12 to 15 hours, even after 264 hours of lute increase in SWS than in young adults on the first
sleep loss.103 If sleep times are held to 8 hours on recovery recovery night, although the percentage increase in SWS
nights, effects on sleep stages may be seen for 2 or more may be as great. Because there is less SWS rebound, there
nights. may be no change (geriatric normals) or even an increase
The effects of 40 and 64 hours of sleep loss on recovery in stage 2.27,113 Normal older individuals had a decrease
sleep stages during the initial recovery night are summa- in REM latency during recovery sleep102,109,113 rather than
rized in Table 5-1 for normal young adults,47,104-107 young the increased REM latency common in young adults. It
adult short sleepers,108 young adult long sleepers,108 60- to was found that REM latency in the older population was
80-year-old normal sleepers,102,109,110 60- to 70-year-old positively correlated with baseline SWS amounts109,113 and
chronic insomniacs,109,111 and 60- to 80-year-old depressed that sleep-onset REM periods occurred in about 20% of
and demented patients.112,113 The table presents percentage those carefully screened normal subjects.109,113 These REM
change from baseline data, with an indication of study- changes were interpreted to be the result of decreased
to-study variability where the number of studies allowed pressure for SWS in older humans. The REM latency
computation. The table is presented as a summary device findings did not apply to older depressed or demented
so that EEG effects of sleep deprivation can be predicted individuals. REM rebound effects appear to be related to
(roughly by multiplying population baseline values by the amount of lost SWS, so that REM rebound is more
figures presented in the table), and so that the potential likely on an early recovery night when there is less SWS
differential effects of sleep deprivation on EEG recovery loss as a function of either a shorter period of sleep loss
sleep as a function of group can be more clearly seen. The or age.
results of these several studies indicate that recovery sleep On the second recovery night after total sleep loss, SWS
EEG changes that occur as a function of sleep deprivation amounts approached normal values, and an increase in
are remarkably consistent across studies and across several REM sleep was found in young adults.47 Total sleep time
experimental groups including men, women, older sub- was still elevated. By the third recovery night, all sleep
jects, and older insomniacs. Significant deviations from EEG values approached baseline. In situations where
population recovery values are seen primarily in REM REM rebounds on the first sleep-recovery night, sleep
latency changes in depressed and demented patients, and EEG values may normalize by the second recovery night.
64 PARTI / Section 1 Normal Sleep and Its Variations

Exceptions to these general rules may include older insom-

niacs, who have increased total sleep for at least 3 nights Acknowledgements
after 64 hours of sleep loss27 and individuals who have had Supported by the Medical Research Service of the Dayton
significant selective REM deprivation. Department of Veterans Affairs Medical Center and
Wright State University, Dayton, Ohio. Literature searches
Relationship between EEG and were supported by the Sleep-Wake Disorders Research
Psychomotor Performance Institute, Dayton, Ohio.
Recovery Effects
The increase in SWS during recovery from sleep loss leads REFERENCES
to the speculation that SWS is implicated in the sleep 1. Manaceine M. Quelques observations experimentales sur linfluence
de linsomnie absolue. Arch Ital Biol 1894;21:322-325.
recovery process. Unfortunately, human studies designed 2. Patrick GTW, Gilbert JA. On the effect of loss of sleep. Psychol
to test this hypothesis directly by experimentally varying Rev 1896;3:469-483.
the amount of SWS during the recovery sleep period or 3. Mikulincer M, Babkoff H, Caspy T, Sing H. The effects of 72 hours
during a sleep fragmentation period have not implicated of sleep loss on psychological variables. Br J Psychol 1989;80:
any sleep stages as central in the recovery process.114 145-162.
4. Bonnet MH, Arand DL. Sleep latency testing as a time course
However, these studies were not designed to look for more measure of state arousal. J Sleep Res 2005;14:387-392.
subtle effects that might have occurred in the initial recov- 5. Lubin A, Hord DJ, Tracy ML, et al. Effects of exercise, bedrest and
ery night. napping on performance decrement during 40 hours. Psychophysi-
Studies have also examined recovery of alertness and ology 1976;13:334-339.
6. Bonnet MH, Arand DL. Level of arousal and the ability to maintain
performance after total sleep deprivation for 1 or 2 nights. wakefulness. J Sleep Res 1999;8:247-254.
An early study that examined performance recovery in the 7. Yokoi M, Aoki K, Shimomura Y, et al. Exposure to bright light
sleep period found recovery of response speed to baseline modifies HRV responses to mental tasks during nocturnal sleep
levels after 1 night of recovery sleep following 40 hours of deprivation. J Physiol Anthropol 2006;25:153-161.
sleep loss and recovery to baseline levels during the second 8. Wilkinson RT. Interaction of noise with knowledge of results and
sleep deprivation. J Exp Psychol 1963;66:332-337.
night of recovery sleep following 64 hours of sleep loss.115 9. Poulton EC, Edwards RS, Colquhoun WP. The interaction of the
More recent studies with larger groups of subjects have loss of a nights sleep with mild heat: task variables. Ergonomics
reported that simple response speed had recovered to base- 1974;17:59-73.
line levels after 1 night of recovery sleep following 64 10. Reyner LA, Horne JA. Evaluation of in-car countermeasures to
sleepiness: cold air and radio. Sleep 1998;21:46-50.
hours of sleep loss in one study116 but did not recover to 11. Bonnet MH, Arand DL. Arousal components which differentiate
baseline levels even after 5 recovery nights in a second the MWT from the MSLT. Sleep 2001;24:441-450.
study.117 MSLT was still significantly shorter after 1 night 12. Bonnet MH, Balkin TJ, Dinges DF, et al. The use of stimulants to
of recovery sleep following both 1 and 2 nights of total modify performance during sleep loss: a review by the Sleep Depri-
sleep loss.116,117 One study reported recovery for the MSLT vation and Stimulant Task Force of the American Academy of Sleep
Medicine. Sleep 2005;28:1163-1187.
after a second night of recovery sleep following 64 hours 13. Wesensten J, Belenky G, Kautz MA, et al. Maintaining alertness
of sleep loss,116 but the second study did not.117 and performance during sleep deprivation: modafinil versus caf-
feine. Psychopharmacology 2002;159:238-247.
14. Wesensten NJ, Killgore WD, Balkin TJ. Performance and alertness
CONCLUSIONS effects of caffeine, dextroamphetamine, and modafinil during sleep
deprivation. J Sleep Res 2005;14:255-266.
The physiologic and behavioral effects of sleep loss in 15. Bonnet MH, Gomez S, Wirth O, et al. The use of caffeine versus
humans are consistent and well defined. There is a physi- prophylactic naps in sustained performance. Sleep 1995;18:97-104.
ologic imperative to sleep in man and other mammals, and 16. Bonnet MH, Arand DL. The use of prophylactic naps and caffeine
the drive to sleep can be as strong as the drive to breathe. to maintain performance during a continuous operation. Ergonom-
ics 1994;37:1009-1020.
Future work should (1) examine in more detail the physi- 17. Wright KP, Badia P, Myers BL, et al. Combination of bright
ologic microstructure of the sleep process and its relation- light and caffeine as a countermeasure for impaired alertness and
ship to sleep restoration, (2) reconcile response differences performance during extended sleep deprivation. J Sleep Res 1997;
among species, (3) examine differences in response to sleep 6:26-35.
18. Dinges DF, Arora S, Darwish M, et al. Pharmacodynamic effects
deprivation in normal and depressed humans, and (4) on alertness of single doses of armodafinil in healthy subjects during
further explore the interaction of the sleep and the arousal a nocturnal period of acute sleep loss. Curr Med Res Opin
systems, (5) examine impact in specific occupations.118 2006;22:159-167.
19. Newhouse PA, Penetar DM, Fertig JB, et al. Stimulant drug effects
on performance and behavior after prolonged sleep deprivation: a
comparison of amphetamine, nicotine, and deprenyl. Mil Psychol-
Clinical Pearl ogy 1992;4:207-233.
20. Fischman MW, Schuster CR. Cocaine effects in sleep-deprived
The impact of sleep deprivation on performance and humans. Psychopharmacology 1980;72:1-8.
physiology has been examined in thousands of 21. Roehrs T, Beare D, Zorick F, et al. Sleepiness and ethanol effects
studies for over a hundred years. These findings on simulated driving. Alcohol Clin Exp Res 1994;18:154-158.
might not seem directly relevant in a clinical sense, 22. Arnedt JT, Wilde GJ, Munt PW, et al. How do prolonged wakeful-
ness and alcohol compare in the decrements they produce on a
but it should be remembered that the clinical diag- simulated driving task? Accid Anal Prev 2001;33:337-344.
nosis of insufficient sleep syndrome is based on sleep 23. Horne JA, Pettitt AN. High incentive effects on vigilance perfor-
deprivation. Sleepiness symptoms secondary to sleep mance during 72 hours of total sleep deprivation. Acta Psychologica
apnea and periodic limb movements (sleep fragmen- 1985;58:123-139.
tation) also evolve from sleep deprivation. 24. Haslam DR. The incentive effect and sleep deprivation. Sleep
CHAPTER 5 Acute Sleep Deprivation 65

25. Baranski JV, Thompson MM, Lichacz FM, et al. Effects of sleep 52. Rodin EA, Luby ED, Gottleib JS. The EEG during prolonged
loss on team decision making: motivational loss or motivational experimental sleep deprivation. Electroencephalogr Clin Neuro-
gain? Hum Factors 2007;49:646-660. physiol 1962;14:544-551.
26. Webb WB, Levy CM. Effects of spaced and repeated total sleep 53. Naitoh P, Pasnau RO, Kollar EJ. Psychophysiological changes after
deprivation. Ergonomics 1984;27:45-58. prolonged deprivation of sleep. Biol Psychiatry 1971;3:309-320.
27. Bonnet MH, Rosa RR. Sleep and performance in young adults and 54. Williams HL, Granda AM, Jones RC, et al. EEG frequency and
older insomniacs and normals during acute sleep loss and recovery. finger pulse volume as predictors of reaction time during sleep loss.
Biol Psychol 1987;25:153-172. Electroencephalogr Clin Neurophysiol 1962;14:64-70.
28. Adam M, Retey JV, Khatami R, et al. Age-related changes in the 55. Thomas M, Sing H, Belenky G, et al. Neural basis of alertness and
time course of vigilant attention during 40 hours without sleep in cognitive performance impairments during sleepiness. I. Effects of
men. Sleep 2006;29:55-57. 24h of sleep deprivation on waking human regional brain activity.
29. Leproult R, Colecchia EF, Berardi AM, et al. Individual differences J Sleep Res 2000;9:335-352.
in subjective and objective alertness during sleep deprivation are 56. Drummond SP, Brown GG, Salamat JS, et al. Increasing task dif-
stable and unrelated. Am J Physiol Regul Integr Comp Physiol ficulty facilitates the cerebral compensatory response to total sleep
2003;284:R280-R290. deprivation. Sleep 2004;27:445-451.
30. Mu Q, Mishory A, Johnson KA, et al. Decreased brain activation 57. Lim J, Choo WC, Chee MW. Reproducibility of changes in behav-
during a working memory task at rested baseline is associated with iour and fMRI activation associated with sleep deprivation in a
vulnerability to sleep deprivation. Sleep 2005;28:433-446. working memory task. Sleep 2007;30:61-70.
31. Chee MW, Chuah LY, Venkatraman V, et al. Functional imaging of 58. Scalise A, Desiato MT, Gigli GL, et al. Increasing cortical excit-
working memory following normal sleep and after 24 and 35 h of ability: a possible explanation for the proconvulsant role of sleep
sleep deprivation: correlations of fronto-parietal activation with deprivation. Sleep 2006;29:1595-1598.
performance. Neuroimage 2006;31:419-428. 59. Zhong X, Hilton HJ, Gates GJ, et al. Increased sympathetic and
32. Retey JV, Adam M, Gottselig JM, et al. Adenosinergic mechanisms decreased parasympathetic cardiovascular modulation in normal
contribute to individual differences in sleep deprivation-induced humans with acute sleep deprivation. J Appl Physiol 2005;98:
changes in neurobehavioral function and brain rhythmic activity. 2024-2032.
J Neurosci 2006;26:10472-10479. 60. White DP, Douglas NJ, Pickett CK, et al. Sleep deprivation and the
33. Viola AU, Archer SN, James LM, et al. PER3 Polymorphism pre- control of ventilation. Am Rev Respir Dis 1983;128:984-986.
dicts sleep structure and waking performance. Curr Biol 2007; 61. Phillips BA, Cooper KR, Burke TV. The effect of sleep loss on
17:613-618. breathing in chronic obstructive pulmonary disease. Chest 1987;
34. Groeger JA, Viola AU, Lo JCY, et al. Early morning executive 91:29-32.
functioning during sleep deprivation is compromised by a PERIOD3 61a. Canet E, Gaultier C, DAllest AM, et al. Effects of sleep depriva-
polymorphism. Sleep 2008;31:1159-1167. tion on respiratory events during sleep in healthy infants. J Appl
35. Mullaney DJ, Kripke DF, Fleck PA, et al. Sleep loss and nap effects Physiol 1989;66:1158-1163.
on sustained continuous performance. Psychophysiol 1983;20: 62. Persson HE, Svanborg E. Sleep deprivation worsens obstructive
643-651. sleep apnea. Comparison between diurnal and nocturnal polysom-
36. Johnson LC. Physiological and psychological changes following nography. Chest 1996;109:645-650.
total sleep deprivation. In: Kales A, editor. Sleep physiology and 63. Brooks D, Horner RL, Kimoff RJ, et al. Effect of obstructive sleep
pathology. Philadelphia: JB Lippincott; 1969. p. 206-220. apnea versus sleep fragmentation on responses to airway occlusion.
37. Kahn-Greene ET, Killgore DB, Kamimori GH, et al. The effects Am J Respir Crit Care Med 1997;155:1609-1617.
of sleep deprivation on symptoms of psychopathology in healthy 64. Minors D, Waterhouse J, Akerstedt T, et al. Effect of sleep loss on
adults. Sleep Med 2007;8:215-221. core temperature when movement is controlled. Ergonomics
38. Wu JC, Buchsbaum M, Bunney W, et al. Antidepressant effects. In: 1999;42:647-656.
Kushida CA, editor. Sleep deprivation: basic science, physiology, 65. Shaw PJ. Thermoregulaory changes. In: Kushida CA, editor. Sleep
and behavior. New York: Marcel Dekker; 2005. p. 421-430. deprivation: basic science, physiology, and behavior. New York:
39. Pilcher JJ, Huffcutt AI. Effects of sleep deprivation on performance: Marcel Dekker; 2005. p. 319-338.
a meta-analysis. Sleep 1996;19:318-326. 66. Fiorica V, Higgins EA, Iampietro PF, et al. Physiological responses
40. Donnell JM. Performance decrement as a function of total sleep of men during sleep deprivation. J Appl Physiol 1968;24:167-176.
loss and task duration. Percept Mot Skills 1969;29:711-714. 67. Van Den Noort S, Brine K. Effect of sleep on brain labile phos-
41. Wilkinson RT. Interaction of lack of sleep with knowledge of phates and metabolic rate. Am J Physiol 1970;218:1434-1439.
results, repeated testing, and individual differences. J Exp Psychol 68. Akerstedt T, Palmblad J, de la Torre B, et al. Adrenocortical and
1961;62:263-271. gonadal steroids during sleep deprivation. Sleep 1980;3:23-30.
42. Light AI, Sun JH, McCool C, et al. The effects of acute sleep depri- 69. Froberg JE. Twenty-four-hour patterns in human performance,
vation on level of resident training. Curr Surg 1989;46:29-30. subjective and physiological variables and differences between
43. Alluisi EA, Coates GD, Morgan BBJ. Effects of temporal stressors morning and evening active subjects. Biol Psychol 1977;5:119-134.
on vigilance and information processing. In: Mackie RR, editor. 70. Kollar EJ, Slater GG, Palmer JO, et al. Stress in subjects undergo-
Vigilance: theory, operational performance, and physiological cor- ing sleep deprivation. Psychosom Med 1966;28:101-113.
relates. New York: Plenum Press; 1977. p. 361-421. 71. Gary KA, Winokur A, Douglas SD, et al. Total sleep deprivation
44. Forest G, Godbout R. Attention and memory changes. In: Kushida and the thyroid axis: effects of sleep and waking activity. Aviat Space
CA, editor. Sleep deprivation: basic science, physiology, and behav- Environ Med 1996;67:513-519.
ior. New York: Marcel Dekker; 2005. p. 199-222. 72. Goh VH, Tong TY, Lim C, et al. Effects of one night of sleep
45. Killgore WD, Killgore DB, Day LM, et al. The effects of 53 hours deprivation on hormone profiles and performance efficiency. Mil
of sleep deprivation on moral judgment. Sleep 2007;30:345-352. Med 2001;166:427-431.
46. McKenna BS, Dicjinson DL, Orff HJ, Drummond SP. The effects 73. Zeitzer JM, Duffy JF, Lockley SW, et al. Plasma melatonin rhythms
of one night of sleep deprivation on known-risk and ambiguous-risk in young and older humans during sleep, sleep deprivation, and
decisions. J Sleep Res 2007;16:245-252. wake. Sleep 2007;30:1437-1443.
47. Carskadon MA, Dement WC. Effects of total sleep loss on sleep 74. Spiegel K, Leproult R, Van Cauter E. Metabolic and endocrine
tendency. Percept Mot Skill 1979;48:495-506. changes. In: Kushida CA, editor. Sleep deprivation: basic science,
48. Marcus CL, Loughlin GM. Effect of sleep deprivation on driving physiology, and behavior. New York: Marcel Dekker; 2005. p.
safety in housestaff. Sleep 1996;19:763-766. 293-318.
49. Horne JA. A review of the biological effects of total sleep depriva- 75. Schussler P, Uhr M, Ising M, et al. Nocturnal ghrelin, ACTH, GH
tion in man. Biol Psychol 1978;7:55-102. and cortisol secretion after sleep deprivation in humans. Psycho-
50. Kollar EJ, Namerow N, Pasnau RO, et al. Neurological findings neuroendocrinology 2006;31:915-923.
during prolonged sleep deprivation. Neurology 1968;18:836-840. 76. Cirelli C. Functional genomic of sleep and circadian rhythm
51. Finelli LA. Cortical and electroencephalographic changes. In: Invited review: how sleep deprivation affects gene expression in the
Kushida CA, editor. Sleep deprivation: basic science, physiology, brain: a review of recent findings. J Appl Physiol 2002;92:
and behavior. New York: Marcel Dekker; 2005. p. 223-264. 394-400.
66 PARTI / Section 1 Normal Sleep and Its Variations

77. Cirelli C. Changes in gene expression. In: Kushida CA, editor. Sleep 97. Spath-Schwalbe E, Gofferje M, Kern W, et al. Sleep disruption
deprivation: basic science, physiology, and behavior. New York: alters nocturnal ACTH and cortisol secretory patterns. Biol Psy-
Marcel Dekker; 2005. p. 387-397. chiatry 1991;29:575-584.
78. Koh K, Joiner WJ, Wu MN, et al. Identification of SLEEPLESS, 98. Sforza E, Chapotot F, Pigeau R, et al. Effects of sleep deprivation
a sleep-promoting factor. Science 2008;321:372-376. on spontaneous arousals in humans. Sleep 2004;27:1068-1075.
79. Dinges DF, Douglas SD, Zaugg L, et al. Leukocytosis and natural 99. Guzman-Marin R, Bashir T, Suntsova N, et al. Hippocampal neu-
killer cell function parallel neurobehavioral fatigue induced by 64 rogenesis is reduced by sleep fragmentation in the adult rat. Neu-
hours of sleep deprivation. J Clin Invest 1994;93:1930-1939. roscience 2007;148:325-333.
80. Irwin MR, Wang M, Campomayor CO, et al. Sleep deprivation and 100. Phillipson EA, Bowes G, Sullivan CE, et al. The influence of sleep
activation of morning levels of cellular and genomic markers of fragmentation on arousal and ventilatory responses to respiratory
inflammation. Arch Intern Med 2006;166:1756-1762. stimuli. Sleep 1980;3:281-288.
81. Moldofsky H. Central nervous system and peripheral immune func- 101. Cassel W, Ploch T, Becker C, et al. Risk of traffic accidents in
tions and the sleep-wake system. J Psychiatr Neurosci 1994;19: patients with sleep-disordered breathing: reduction with nasal
368-374. CPAP. Eur Respir J 1996;9:2606-2611.
82. Boyum A, Wiik P, Gustavsson E, et al. The effect of strenuous 102. Carskadon MA, Dement WC. Sleep loss in elderly volunteers. Sleep
exercise, calorie deficiency and sleep deprivation on white blood 1985;8:207-221.
cells, plasma immunoglobulins and cytokines. Scand J Immunol 103. Johnson LC, Slye ES, Dement WC. Electroencephalographic and
1996;43:228-235. autonomic activity during and after prolonged sleep deprivation.
83. Brown R, Pang G, Husband AJ, et al. Suppression of immunity to Psychosom Med 1965;27:415-423.
influenza virus infection in the respiratory tract following sleep 104. Bonnet MH. The effect of varying prophylactic naps on perfor-
disturbance. Regional Immunology 1989;2:321-325. mance, alertness and mood throughout a 52-hour continuous oper-
84. Renegar KB, Crouse D, Floyd RA, et al. Progression of influenza ation. Sleep 1991;14:307-315.
viral infection through the murine respiratory tract: the protective 105. Borbely AA, Baumann F, Brandeis D, et al. Sleep deprivation: effect
role of sleep deprivation. Sleep 2000;23:859-863. on sleep stages and EEG power density in man. Electroencephalogr
85. Benca RM, Kushida CA, Everson CA, et al. Sleep deprivation in Clin Neurophysiol 1981;51:483-495.
the rat: VII. Immune function. Sleep 1989;12:47-52. 106. Moses J, Lubin A, Naitoh P, et al. Exercise and sleep loss: effects
86. Roehrs T, Hyde M, Blaisdell B, et al. Sleep loss and REM sleep loss on recovery sleep. Psychophysiol 1977;14:414-416.
are hyperalgesic. Sleep 2006;29:145-151. 107. Nakazawa Y, Kotorii M, Ohshima M, et al. Changes in sleep
87. Smith MT, Edwards RR, McCann UD, et al. The effects of sleep pattern after sleep deprivation. Folia Psychiatr Neurol Jpn 1978;32:
deprivation on pain inhibition and spontaneous pain in women. 85-93.
Sleep 2007;30:494-505. 108. Benoit O, Foret J, Bouard G, et al. Habitual sleep length and pat-
88. Schey R, Dickman R, Parthasarathy S, et al. Sleep deprivation is terns of recovery sleep after 24 hour and 36 hour sleep deprivation.
hyperalgesic in patients with gastroesophageal reflux disease. Gas- Electroencephalogr Clin Neurophysiol 1980;50:477-485.
troenterology 2007;133:1787-1795. 109. Bonnet MH. Effect of 64 hours of sleep deprivation upon sleep in
89. Nascimento DC, Andersen ML, Hipolide DC, et al. Pain hyper- geriatric normals and insomniacs. Neurobiol Aging 1986;7:89-96.
sensitivity induced by paradoxical sleep deprivation is not due to 110. Reynolds CFd, Kupfer DJ, Hoch CC, et al. Sleep deprivation in
altered binding to brain mu-opioid receptors. Behav Brain Res healthy elderly men and women: effects on mood and on sleep
2007;178:216-220. during recovery. Sleep 1986;9:492-501.
90. Schmid SM, Hallschmid M, Jauch-Chara K, et al. Sleep loss alters 111. Bonnet MH, Arand DL. Sleep loss in aging. In: Roth T, Roehrs T,
basal metabolic hormone secretion and modulates the dynamic editors. Clinics in geriatric medicine. 1989. p. 405-420.
counterregulatory response to hypoglycemia. J Clin Endocrinol 112. Reynolds CFd, Kupfer DJ, Hoch CC, et al. Sleep deprivation effects
Metab 2007;92:3044-3051. in older endogenous depressed patients. Psychiatry Res 1987;21:
91. Tobler I, Sigg H. Long-term motor activity recording of dogs 95-109.
and the effect of sleep deprivation. Experientia 1986;42:987- 113. Reynolds CF 3rd, Kupfer DJ, Hoch CC, et al. Sleep deprivation as
991. a probe in the elderly. Arch Gen Psychiatry 1987;44:982-990.
92. Plyley MJ, Shephard RJ, Davis GM, et al. Sleep deprivation and 114. Bonnet MH. Performance and sleepiness following moderate sleep
cardiorespiratory function. Influence of intermittent submaximal disruption and slow wave sleep deprivation. Physiol Behav 1986;
exercise. Eur J Appl Physiol 1987;56:338-344. 37:915-918.
93. Born J, Lange T, Hansen K, et al. Effects of sleep and circadian 115. Rosa RR, Bonnet MH, Warm JS. Recovery of performance during
rhythm on human circulating immune cells. J Immunol 1997; sleep following sleep deprivation. Psychophysiol 1983;20:152-159.
158:4454-4464. 116. Beaumont M, Batejat D, Coste O, et al. Recovery after prolonged
94. Bonnet M. Sleep fragmentation. In: Kushida CA, editor. Sleep sleep deprivation: residual effects of slow-release caffeine on recov-
deprivation: basic science, physiology, and behavior. New York: ery sleep, sleepiness and cognitive functions. Neuropsychobiology
Marcel Dekker; 2005. p. 503-513. 2005;51:16-27.
95. Stepanski E. The effect of sleep fragmentation on daytime function. 117. Lamond N, Jay SM, Dorrian J, et al. The dynamics of neurobehav-
Sleep 2002;25:268-276. ioural recovery following sleep loss. J Sleep Res 2007;16:33-41.
96. Martin SE, Wraith PK, Deary IJ, et al. The effect of nonvisible 118. Malmberg B, Kecklund G, Karlson B, et al. Sleep and recovery in
sleep fragmentation on daytime function. Am J Respir Crit Care physicians on night call: a longitudinal field study. BMC Health
Med 1997;155:1596-1601. Serv Res 2010;10(1):239.
Chronic Sleep Deprivation
Siobhan Banks and David F. Dinges
Abstract (3) do not result in profound subjective sleepiness or full self-
awareness of the cumulative deficits from sleep restriction.
Chronic sleep deprivation, also referred to as chronic sleep The mechanisms underlying the sleep doseresponse cumula-
restriction, is common, with a wide range of causes including tive neurobehavioral and physiologic alterations during
shift work and other occupational and economic demands, chronic sleep restriction remain unknown. Individual variabil-
medical conditions and sleep disorders, and social and domes- ity in neurobehavioral responses to sleep restriction appear
tic responsibilities. Sleep doseresponse experiments have to be as large as those in response to total sleep deprivation
found that chronic sleep restriction to less than 7 hours per and as stable over time, suggesting a traitlike (possibly
night resulted in cognitive deficits that (1) accumulate (i.e., genetic) differential vulnerability to the effects of chronic sleep
become progressively worse over time as sleep restriction restriction or differences in the nature of compensatory brain
persists), (2) are sleep-dose sensitive (i.e., the less sleep that responses to the growing sleep loss.
is obtained, the faster the rate at which deficits develop), and

Chronic sleep restriction occurs frequently and results by problems in experimental design.8 Since 1997, experi-
from a number of factors, including medical conditions ments that have corrected for these methodological weak-
(e.g., pain), sleep disorders, work demands (including nesses have found markedly different results from those
extended work hours and shift work), and social and earlier studies, and have documented cumulative objective
domestic responsibilities. Adverse effects on neurobehav- changes in neurobehavioral outcomes as sleep restriction
ioral functioning accumulate as the magnitude of sleep loss progressed.10 This chapter reviews the cognitive and neu-
escalates, and the result is an increased risk of on-the-job robehavioral consequences of chronic sleep restriction in
errors, injuries, traffic accidents, personal conflicts, health healthy individuals.
complaints, and drug use.
Chronic sleep restriction, or partial sleep deprivation,
has been thought to occur when one fails to obtain a usual INCIDENCE OF CHRONIC SLEEP
amount of sleep.1,2 Half a century ago, Kleitman first used RESTRICTION
the phrase sleep debt to describe the circumstances of delay- Human sleep need, or, more precisely, the duration of
ing sleep onset time while holding sleep termination time sleep needed to prevent daytime sleepiness, elevated sleep
constant.3 He described the increased sleepiness and propensity, and cognitive deficits, has been a long-standing
decreased alertness in individuals on such a sleepwake controversy central to whether chronic sleep restriction
pattern, and proposed that those subjects who were able may compromise health and behavioral functions. Self-
to reverse these effects by extending their sleep on week- reported sleep durations are frequently less than 8 hours
ends were able to liquidate the debt.3, p. 317 per night. For example, approximately 20% of more than
The term sleep debt is usually synonymous with chronic 1.1 million Americans indicated that they slept 6.5 hours
sleep restriction because it refers to the increased pressure or less each night.11 Similarly, in polls of 1000 American
for sleep that results from an inadequate amount of physi- adults by the National Sleep Foundation, 15% of subjects
ologically normal sleep.4 To determine the effects of (aged 18 to 84 years) reported sleeping less than 6 hours
chronic sleep loss on a range of neurobehavioral and physi- on weekdays, and 10% reported sleeping less than 6 hours
ologic variables, a variety of paradigms have been used, on weekends over the past year.12 Scientific perspectives on
including controlled, restricted time in bed for sleep the duration of sleep that defines chronic sleep restriction
opportunities in both continuous and distributed sched- have come from a number of theories.
ules,5 gradual reductions in sleep duration over time,6
selective deprivation of specific sleep stages,7 and limiting THEORETICAL PERSPECTIVES ON
the time in bed to a percentage of the individuals habitual SLEEP NEED AND SLEEP DEBT
time in bed.8 These studies have ranged from 24 hours9 to
8 months6 in length. Basal Sleep Need
Many reports published before 1997 concluded that The amount of sleep habitually obtained by an individual
chronic sleep restriction in the range commonly experi- is determined by a variety of factors. Epidemiologic and
enced by the general population (i.e., sleep durations of experimental studies point to a high between-subjects vari-
less than 7 hours per night but greater than 4 hours per ance in sleep duration, influenced by environmental,
night) resulted in some increased subjective sleepiness but genetic, and societal factors. Although not clearly defined
had little or no effect on cognitive performance capabili- in the literature, the concept of basal sleep need has been
ties. Consequently, there was a widely held belief that described as habitual sleep duration in the absence of pre-
individuals could adapt to chronic reductions in sleep existing sleep debt.13 Sleep restriction has been defined as
duration, down to 4 to 5 hours per day. However, nearly the fundamental duration of sleep below which waking
all of these reports of adaptation to sleep loss were limited deficits begin to accumulate.14 Given these definitions, the

68 PARTI / Section 1 Normal Sleep and Its Variations

basal need for sleep appears to be between 7.5 and 8.5 optional sleep. There is, instead, recovery sleep, which
hours per day in healthy adult humans. This number was may or may not be optional, although there have very few
based on a study in which prior sleep debt was completely studies of the sleep needed to recover from varying degrees
eliminated through repeated nights of long-duration sleep of chronic sleep restriction.
that stabilized at a mean of 8.17 hours.15 A similar value
was obtained from a large-scale doseresponse experiment Adaptation to Sleep Restriction
on chronic sleep restriction that statistically estimated One popular belief is that subjects may be acutely affected
daily sleep need to average 8.16 hours per night to avoid after initial restriction of sleep length and may then be able
detrimental effects on waking functions.4 to adapt to the reduced sleep amount, with waking neuro-
cognitive functions unaffected further or returned to base-
Core Sleep versus Optional Sleep line levels. Although several studies have suggested that
In the 1980s, it was proposed that a normal nocturnal sleep this is the case when sleep duration is restricted to approxi-
period was composed of two types of sleep relative to mately 4 to 6 hours per night for up to 8 months,6,9 there
functional adaptation: core and optional sleep.16,17 The is also evidence indicating that the adaptation is largely
initial duration of sleep in the sleep period was referred confined to subjective reports of sleepiness but not objec-
to as core, or obligatory, sleep, which was posited to tive cognitive performance parameters.4 This suggests that
repair the effects of waking wear and tear on the cere- the presumed adaptation effect is actually a misperception
brum.16, p. 57 Initially, the duration of required core sleep on the part of chronically sleep-restricted people regarding
was defined as 4 to 5 hours of sleep per night, depending how sleep restriction has affected their cognitive
on the duration of the sleep restriction.16 The duration of capability.
core sleep has subsequently been redefined as 6 hours of One factor thought to be important in adaptation to
(good-quality, uninterrupted) sleep for most adults.14 chronic sleep restriction is the abruptness of the sleep
Additional sleep obtained beyond the period of core sleep curtailment. One study examined the relationship between
was considered to be optional, or luxury, sleep, which fills rate of accumulation of sleep loss, to a total of 8 hours, and
the tedious hours of darkness until sunrise.16, p. 57 This core neurobehavioral performance levels.19 After 1 night of total
versus optional theory of sleep need is often presented as sleep deprivation (i.e., a rapid accumulation of 8 hours of
analogous to the concept of appetite: Hunger drives one sleep loss), neurobehavioral capabilities were significantly
to eat until satiated, but additional food can still be con- reduced. When the accumulation of sleep loss was slower,
sumed beyond what the body requires. It is unknown achieved by chronically restricting sleep to 4 hours per
whether the so-called optional sleep serves any function. night for 2 nights or 6 hours per night for 4 nights, neu-
According to the core sleep theory, only the core portion robehavioral performance deficits were evident, but they
of sleepwhich is dominated by slow-wave sleep (SWS) were of a smaller magnitude than those following the night
and slow-wave activity (SWA) on an electroencephalogram of total sleep loss. A greater degree of neurobehavioral
(EEG)is required to maintain adequate levels of daytime impairment was evident in those subjects restricted to 4
alertness and cognitive functioning.16 The optional sleep hours for 2 nights than in those subjects allowed 6 hours
does not contribute to this recovery or maintenance of per night, leading to the conclusion that during the slowest
neurobehavioral capability. This theory was strengthened accumulation of sleep debt (i.e., 6 hours per night for 4
by results from a mathematical model of sleep and waking nights), there was evidence of a compensatory adaptive
functions (the three-process model) that predicted that mechanism.19
waking neurobehavioral functions were primarily restored It is possible but not scientifically resolved that different
during SWS,18 which makes up only a portion of total sleep objective neurobehavioral measures may show different
time. However, if only the core portion of sleep is required, degrees of sensitivity and adaptation to chronic sleep
it would be reasonable to predict that there would be no restriction. For example, in the largest controlled study to
waking neurobehavioral consequences of chronically date with statistical modeling of adaptation curves, cogni-
restricting sleep to 6 hours per night, and that cognitive tive performance measures showed little adaptation across
deficits would be evident only when sleep durations were 14 days of sleep restriction to 4 or 6 hours per night, com-
reduced below this amount. Experimental data have not pared with 8 hours per night,4 whereas waking EEG mea-
supported this prediction.10 For example, findings from the sures of alpha and theta frequencies showed no systematic
largest sleep doseresponse study to date, which examined sleep dosedependent changes over days.14 Consequently,
the effects of sleep chronically restricted to 4, 6, or 8 hours different neurobehavioral outcomes showed markedly dif-
of time in bed per night,4 found that cognitive performance ferent responses to chronic sleep restriction, with neuro-
measures were stable across 14 days of sleep restriction to cognitive functions showing the least adaptation, subjective
8 hours time in bed, but when sleep was reduced to either sleepiness measures showing more adaptation, and waking
6 or 4 hours per night, significant cumulative (dose-depen- EEG measures as well as nonrapid eye movement (non-
dent) decreases in cognitive performance functions and REM), SWS measures showing little or no response.4,14
increases in sleepiness were observed.4 The reliability of the latter findings may depend on the
It appears, therefore, that the core sleep needed to dose of restricted sleep and other factors.
maintain stable waking neurobehavioral functions in
healthy adults aged 22 to 45 years is in the range of 7 to 8 Two-Process Model Predictions of
hours on average.18 Moreover, because extended sleep is Sleep Restriction
thought to dissipate sleep debt caused by chronic sleep Biomathematical models of sleepwake regulation have
restriction,3 it is not clear that there is such a thing as been used to make predictions about recovery in response
CHAPTER 6 Chronic Sleep Deprivation 69

to various sleep durations. The basis of almost all current studies examining the effects of chronic sleep restriction
biomathematical models of sleepwake regulation is the on cognitive performance were conducted outside a con-
two-process model of sleep regulation.20 This model pro- trolled laboratory setting, with little or no control over
poses that two primary components regulate sleep: (1) a potentially contaminating factors, such as the level of
homeostatic process that builds up exponentially during napping, extension of sleep periods, diet, stimulant use
wakefulness and declines exponentially during sleep (as (e.g., caffeine, nicotine), activity, or exposure to zeitgebers
measured by slow-wave energy or delta power in the (environmental time cues). The majority of these studies
non-REM sleep EEG), and (2) a circadian process, with concluded that there were few or no detrimental effects on
near24-hour periodicity. waking neurobehavioral capabilities, or subjective effects
Since its inception, the two-process model has gained of the sleep restriction. For example, restriction of noctur-
widespread acceptance for its explanation of the timing and nal sleep periods to between approximately 4 and 6 hours
structure of sleep. Its use has extended to predictions of per night for up to 8 months produced no significant
waking alertness and neurobehavioral functions in response effects on a range of cognitive outcomes, including vigi-
to different sleepwake scenarios.21 This extension of the lance performance,9 psychomotor performance,6 logical
two-process model was based on observations that as sleep reasoning, addition, or working memory.9 In addition, few
pressure accumulated with increasing time awake, so did effects on subjective assessments of sleepiness or mood
waking neurobehavioral or neurocognitive impairment, were reported.9
and as sleep pressure dissipated with time asleep, perfor- Later studies, however, with far greater experimental
mance capability improved during the following period of control and appropriate control groups, have demonstrated
wakefulness. In addition, forced-desynchrony experiments significant cumulative sleep doseresponse effects on a
revealed that the sleep homeostatic and circadian processes wide range of physiologic and neurobehavioral functions,
interacted to create periods of stable wakefulness and con- which we summarize here.
solidated sleep during normal 24-hour days.22 Hence, it
was postulated that waking cognitive function (alertness Sleep Architecture
variable A) could be mathematically modeled as the differ- Sleep restriction alters sleep architecture, but it does not
ence between the quantitative state for the homeostatic affect all sleep stages equally. Depending on the timing and
process (S) and the quantitative state for the circadian duration of sleep, and the number of days it is reduced,
process (C), and thus A = S C. Accordingly, predictions some aspects of sleep are conserved, occur sooner, or
for changes in the neurobehavioral recovery afforded by intensify, and other aspects of sleep time are diminished.
chronically restricted sleep of varying durations could be For example, studies examining sleep architecture during
made on the basis of sleepwake times and circadian phase chronic periods of sleep restriction have demonstrated a
estimates, using the quantitative version of the two-process consistent conservation of SWS at the expense of other
model. non-REM and REM sleep stages.4,24,25 In addition, eleva-
The validity of the various biomathematical models tions in SWA, derived from spectral analysis of the sleep
based on the two-process model, and their ability to predict EEG in the range of 0.5 to 4.5Hz, during non-REM sleep
actual experimental results of the neurobehavioral effects have also been reported during and after chronic sleep
of chronic sleep restriction have been evaluated in a blind restriction.4,25
test.23 Because all current models are based on the same Because of the conservation of the amount of SWS
underlying principles as the two-process model, all yielded and SWA during restricted sleep protocols, independent
comparable predictions for neurobehavioral functioning in of sleep duration (e.g., 8 hours of time in bed or 4
scenarios involving total sleep deprivation or chronic hours of time in bed), it has been proposed that, with
partial sleep restriction. All models accurately predicted regard to behavioral and physiologic outcomes, these
waking neurobehavioral responses to total sleep depriva- phenomena provide the recovery aspects of sleep. It
tion. However, they all failed to adequately predict sleepi- remains to be determined whether the lack of SWS and
ness and cognitive performance responses during chronic SWA response to chronic restriction of sleep to 4 hours
sleep restriction.4,23 Hence, it appears that the extension of a night, relative to steady increases in physiologic and
the two-process model to prediction of waking alertness21 neurobehavioral measures of sleepiness,4 can account for
does not account for the results of chronic sleep restric- the latter deficits. Consequently, although SWS and
tion. Because the two-process model has had a profound non-REM SWA may be conserved in chronic sleep restric-
theoretical influence on predictions of sleepiness based on tion (to 4 to 7 hours per night), they do not appear to
total sleep deprivation data, its failure to capture the reflect the severity of daytime cognitive deficits or to
dynamic changes in neurobehavioral measures during protect against these deficits, raising serious doubts about
chronic sleep restriction suggests that additional biological SWS and non-REM SWA being the only aspects of
factors are relevant to the brains response to chronic sleep sleep critical to waking functions in chronic sleep
restriction. restriction.4
Sleep Propensity
EFFECTS OF CHRONIC SLEEP With the development and validation of sleep latency
RESTRICTION measures as sensitive indices of sleep propensity,26 the
The effects of sleep loss may be quantified in a number of effects of chronic sleep restriction could be evaluated phys-
different ways, using a wide range of physiologic, neuro- iologically. Objective EEG measures of sleep propensity,
cognitive, behavioral, and subjective tools. Many early such as the multiple sleep latency test26 (MSLT) and the
70 PARTI / Section 1 Normal Sleep and Its Variations

maintenance of wakefulness test27 (MWT), are frequently sleep during sleep restriction. EEG frequencies in the
used to evaluate sleepiness (see Chapters 4 and 143). slower range (0.5 to 14Hz)in sleep-deprived individuals
The daytime MSLT26 has been shown to vary linearly in particularmay herald an increased tendency for micro-
after 1 night of sleep restricted to between 1 and 5 hours sleeps. Significant increases in power densities in the delta
of time in bed.26 Progressive decreases in daytime sleep range (3.75 to 4.5Hz) and decreases in the alpha range
latency have been documented (i.e., increases in sleep pro- (9.25 to 10Hz) have been reported in subjects exposed to
pensity) across 7 days of sleep restricted to 5 hours per 4 hours of sleep restriction for 4 nights.25 In contrast, no
night in healthy young adults,28 a finding confirmed in a effect on waking alpha power (8 to 12Hz) across days of
later study using the psychomotor vigilance test (PVT).8 restriction was evident in subjects restricted to 4 or 6 hours
Doseresponse effects of chronic sleep restriction on of time in bed for sleep per night for 14 nights.14 An
daytime MSLT values have been reported in a controlled increase in theta power (4 to 8Hz) across days of the sleep
laboratory study in commercial truck drivers.24 A signifi- restriction protocol was evident; however, there was no
cant increase in sleep propensity across 7 days of sleep significant difference in theta power changes between
restricted to either 3 or 5 hours per night was observed, restriction conditions. It has been suggested that these
with no increase in sleep propensity found when sleep was changes in waking EEG frequency during sleep loss reflect
restricted to 7 or 9 hours per night.24 Similarly, sleep pro- spectral leakage indicative of elevated sleep pressure
pensity (as measured by the MWT29) during 7 days of sleep manifesting in wakefulness.37 Few studies have investigated
restriction to 4 hours per night was reported to increase, the leakage of slower EEG activity in humans, but animal
especially in subjects whose sleep was restricted by advanc- studies suggest that this process may enable the brain to
ing sleep offset.30 discharge some of the accumulated homeostatic sleep drive
An epidemiologic study of predictors of objective sleep without actually going to sleep.37
tendency in the general population31 also found a dose
response relationship between self-reported nighttime Cognitive Effects
sleep duration and objective sleep tendency as measured Reduced sleep time can adversely affect different aspects
by the MSLT. Persons reporting more than 7.5 hours of of waking cognitive performance, especially behavioral
sleep had significantly less probability of falling asleep on alertness, which is fundamental to many cognitive tasks.
the MSLT than those reporting between 6.75 and 7.5 Behavioral alertness can be measured with the PVT, which
hours per night (27% risk of falling asleep), and than those requires vigilant attention and has proved to be very sensi-
reporting sleep durations less than 6.75 hours per night tive to any reduction in habitual sleep time.38,39 Studies
(73% risk of falling asleep).31 Although the MWT has have consistently shown that sleep restriction increases
been used less in experimental settings than the MSLT, it PVT response slowing40 and lapses,38 which are thought to
has also been found to increase in experiments in which reflect microsleeps.3,41 As loss of sleep accumulates, rela-
adults were restricted to 4 hours for sleep for 7 nights,30 tively brief lapses of a half second can increase to well over
and for 5 nights.32 All of these studies suggest that chronic 10 seconds and longer.3,41,42 It is suggested that lapses pro-
curtailment of nocturnal sleep increases daytime sleep duced by sleep loss involve shifts in neuronal activity in
propensity. frontal, thalamic, and secondary sensory processing areas
Oculomotor responses have also been reported to be of the brain.43 Lapses of attention occur unpredictably
sensitive to sleep restriction.33 Eyelid closure and slow throughout cognitive performance in sleep-restricted sub-
rolling eye movements are part of the initial transition jects, and they increase in frequency and duration as a
from wakefulness to drowsiness. Eye movements and eye function of the severity of sleep restriction, which has led
closures have been studied during sleep-loss protocols, to the idea that they reflect underlying wake state instabil-
under the premise that changes in the number and rate of ity.38,42,43 This instability appears to involve moment-to-
movements and eyelid closures are a reflection of increased moment fluctuations in the relationship between
sleep propensity and precursors of the eventual onset of neurobiological systems mediating wake maintenance and
sleep.34 It has been demonstrated experimentally that slow sleep initiation.43
eyelid closures during performance are associated with One early study of chronic sleep restriction effects on
vigilance lapses and are sensitive indices of sleep depriva- cognitive performance examined the effects of reducing
tion, and slow eyelid closures have been found to be a sign habitual sleep time by 40% for 5 nights.44 Decreases in
of drowsiness while driving.33 performance on a vigilance and simple reaction time per-
Increased slow eye movements attributed to attentional formance task were observed across the protocol with sleep
failures have been reported to be increased by reduced restriction. Interestingly, however, there was no effect of
sleep time in medical residents.35 Sleep restriction has also sleep restriction on a choice reaction time task, suggesting
been found to decrease saccadic velocity and to increase that not all measures of performance are equally sensitive
the latency to pupil constriction in subjects allowed only 3 to chronic sleep restriction. This could result from any of
or 5 hours of time in bed for sleep over 7 nights.36 These a number of aspects of the psychometric properties of
changes in ocular activity were positively correlated with cognitive tests (e.g., learning curves) or from their neuro-
sleep latency, subjective sleepiness measures, and accidents biological substrates; negative findings provide no insight
on a simulated driving task.36 into the reason for lack of sensitivity.
Two large-scale experimental studies published in 2003
Waking Electroencephalogram described dose-related effects of chronic sleep restriction
Slowing in certain waking EEG frequencies has been on neurobehavioral performance measures.4,24 In one
thought to reflect the increased homeostatic pressure for study, truck drivers were randomized to 7 nights of 3, 5,
CHAPTER 6 Chronic Sleep Deprivation 71

7, or 9 hours of time in bed for sleep per night.24 Cognitive 16

performance was assessed using the PVT. Subjects in the
14 Differences among
3- and 5-hour time-in-bed groups experienced a decrease 0 hr TIB 4 hr TIB
in performance across days of the sleep restriction proto- 12 P = .036
col, with increases in the mean reaction time, in the number

PVT lapses (relative #)

10 Curvature (SEM)
of lapses, and in the speed of the fastest reaction on the = 0.78 (0.04)
PVT.24 In the subjects who were allowed 7 hours of time 8 6 hr TIB

Poor performance
Effect sizes
in bed per night, a significant decrease in mean response 4 hr vs. 8 hr: 1.45
speed was also evident, although no effect on lapses was 6
6 hr vs. 8 hr: 0.71
evident. Performance in the group allowed 9 hours of time 4
4 hr vs. 6 hr: 0.43
in bed was stable across the 7 days.
In an equally large experiment,4 young adults had their 2
8 hr TIB
sleep duration restricted to 4, 6, or 8 hours of time in bed 0
per night for 14 nights, and daytime deficits in cognitive
functions were observed for lapses on the PVT (Fig. 6-1), BL 1 2 3 4 5 6 7 8 9 10 11 12 13 14
Days of nocturnal sleep restriction
for a memory task, and for a cognitive throughput task.
These performance deficits accumulated across the experi- Figure 6-1 Psychomotor vigilance task (PVT) performance
mental protocol in those subjects allowed less than 8 hours lapses under varying dosages of daily sleep. Displayed are
of sleep per night.4 Data from this study demonstrate that group averages for subjects in the 8-hour (diamond), 6-hour
(light blue square), and 4-hour (circle) chronic sleep period time
sleep restrictioninduced deficits continued to accumulate in bed (TIB) across 14 days, and in the 0-hour (green square)
beyond the 7 nights of restriction used in other experi- sleep condition across 3 days. Subjects were tested every 2
ments,8,24 with performance deficits still increasing at day hours each day; data points represent the daily average (07:30
14 of the restricted sleep schedule. By the end of the to 23:30) expressed relative to baseline (BL). The curves through
14-day chronic partial-sleep restriction period, the level of the data points represent statistical nonlinear model-based
best-fitting profiles of the response to sleep deprivation for
cognitive impairments recorded in subjects in the 4-hour subjects in each of the four experimental conditions. The
sleep restriction condition was equivalent to the level of ranges (mean SE) of neurobehavioral functions for 1 and 2
impairment seen after 1 to 2 nights without any sleep (see days of 0 hours of sleep (total sleep deprivation) are shown as
Fig. 6-1). To understand the relationship between the dif- light and dark bands, respectively, allowing comparison of the
3-day total sleep deprivation condition and the 14-day chronic
ferent sleep-loss conditions and the equivalence in perfor- sleep restriction conditions. (Redrawn from Van Dongen HP,
mance impairment observed, the amount of cumulative Maislin G, Mullington JM, etal. The cumulative cost of addi-
sleep loss for subjects in each condition was calculated.4 tional wakefulness: dose-response effects on neurobehavioral
The degree of sleep loss was greater in subjects allowed 4 functions and sleep physiology from chronic sleep restriction
hours of sleep each night for 14 nights (i.e., losing approxi- and total sleep deprivation. Sleep 2003;26:117-126.)
mately 55 hours of sleep) than in subjects who remained
awake for 88 hours (i.e., losing approximately 25 hours of
sleep) (Fig. 6-2A), suggesting that impairments in waking It appears that the neurocognitive effects of restricting
performance should have been much worse in the 4-hour nocturnal sleep to 6 or 4 hours per night on a chronic basis
condition. However, this was not the case (see Fig. 6-1). are fundamentally the same as when sleep is chronically
To reconcile this paradox, wake time was defined as the restricted but split each day between a nighttime sleep and
difference between the duration of each continuous wake a daytime nap.45 Cognitive performance deficits also accu-
period and the duration of habitual wake time. Accord- mulate across consecutive days in which the restricted
ingly, cumulative wake-time extension was calculated as sleep occurs during the daytime and wakefulness occurs at
the sum of all consecutive hours of wakefulness extending night.46 The primary difference between the nocturnally4
beyond the habitual duration of wakefulness that each and diurnally46 placed restricted sleep periods is that the
subject was accustomed to at home. In the 4-, 6-, and magnitude of neurobehavioral impairment is significantly
8-hour sleep restriction conditions, this yielded the same greater with daytime sleep compared with nighttime sleep,
results as for cumulative sleep loss, because the definitions reflecting a combined influence of the homeostatic and
of cumulative wake extension and cumulative sleep loss circadian systems. In another experiment, when recovery
were arithmetically equivalent. However, for the 0-hour periods after total sleep time were restricted to 6 hours
total sleep deprivation condition, each day without sleep versus 9 hours time in bed for sleep per night, neither PVT
added 24 hours to the cumulative wake extension. Thus, performance nor sleepiness recovered to baseline levels,
over 3 days with 0 hours of sleep, cumulative wake exten- suggesting that restricting sleep can also reduce its recov-
sion was equal to 72 hours for each subject (see Fig. 6-2B), ery potential.47
whereas cumulative sleep loss was only 23 hours (see Fig. All these studies suggest that when time in bed for sleep
6-2A). These results illustrate that cumulative sleep loss is chronically restricted to less than 7 hours per night in
and cumulative wake extension are different constructs healthy adults (aged 21 to 64 years), cumulative deficits in
that can have different quantitative values, depending on a variety of cognitive performance functions become
the manner in which sleep loss occurs. They also suggest evident. These deficits can accumulate to levels of impair-
that sleep debt can also be understood as resulting in addi- ment equivalent to those observed after 1 or even 2 nights
tional wakefulness beyond an average of approximately 16 of total sleep deprivation.
hours a day, which has a neurobiological cost that accumu- These cognitive performance findings are consistent
lates over time.4 with those on the effects of sleep restriction on physiologic
72 PARTI / Section 1 Normal Sleep and Its Variations

80 80
70 70
60 60

Cumulative hours

Cumulative hours
50 50
40 40
30 30
20 20
10 10
0 0
0 2 4 6 8 10 12 14 0 2 4 6 8 10 12 14
A Days of sleep restriction B Days of sleep restriction

Figure 6-2 Cumulative buildup of sleep loss and wake time extension across days of sleep restriction and total sleep loss.
A, Cumulative sleep loss relative to habitual sleep durationthat is, all hours of sleep habitually obtained (as measured at home
during the 5 days prior to the experiment) but not received in the experiment because of sleep restriction. B, Cumulative wake
extension relative to habitual wake durationthat is, all consecutive hours of wakefulness in excess of the habitual duration of a
wakefulness period. Daily means are shown for subjects in the 8-hour (diamond), 6-hour (light blue square), 4-hour (circle), and
0-hour (green square) sleep period conditions. A also shows the range (orange band) of cumulative sleep loss (relative to habitual
sleep duration) after 3 days in the 0-hour sleep condition, which was 23.1 2.6 hours (mean SD). This was significantly less
than the cumulative sleep loss after 14 days in the 4-hour sleep period condition (t20 = 10.58, P < .001). (Redrawn from Van Dongen
HP, Maislin G, Mullington JM, etal. The cumulative cost of additional wakefulness: dose-response effects on neurobehavioral func-
tions and sleep physiology from chronic sleep restriction and total sleep deprivation. Sleep 2003;26:117-126.)

sleep propensity measures (MSLT, MWT).24,28,30,32 Col- and behaviorally less alert, but they thought themselves
lectively, they suggest that there is a neurobiological inte- only moderately sleepy. This suggests that individuals fre-
grator that accumulates either homeostatic sleep drive or quently underestimate the actual cognitive impact of sleep
the neurobiological consequences of excess wakefulness.4,24 restriction and believe themselves fit to perform. Experi-
There has as yet been no definitive evidence of what causes ments using driving simulators have found similar results,
this destabilization of cognitive functions, but one intrigu- with drivers unable to accurately perceive their level of
ing line of evidence suggests that it may involve extracel- fatigue and cognitive impairment.50
lular adenosine in the basal forebrain.48
Individual Differences in Responses to
Driving Performance Chronic Sleep Restriction
There is an increased incidence of sleep-related motor Although the majority of healthy adults develop cumula-
vehicle crashes in drivers reporting less than 7 hours of tive cognitive deficits and sleepiness with chronic sleep
sleep per night on average.49 Additional contributing restriction, interindividual variability in the neurobehav-
factors to these sleep- or sleepiness-related crashes included ioral and physiologic responses to sleep restriction is sub-
poor sleep quality, dissatisfaction with sleep duration (i.e., stantial.3,38,39,42 Sleep restriction increases neurobehavioral
undersleeping), daytime sleepiness, previous instances of performance variability in subjects,38,39,41 and it also reveals
driving drowsy, and time driving and time of day (driving clear neurobehavioral differences between subjects. This
late at night). It has been found that after 1 night of interindividual variability is quite apparent in sleep restric-
restricted sleep (5 hours), a decrease in performance on a tion studies. For example, not everyone was affected to the
driving simulator, with a concurrent increase in subjec- same degree when sleep duration was limited to less than
tively reported sleepiness, was found.50 In addition, during 7 hours per day in the studies described earlier.4,24 Some
chronic sleep restriction in a controlled laboratory, with people experience very severe impairments even with
sleep durations reduced to between 4 and 6 hours per 24 modest sleep restriction, whereas others show little effect
hours, placed either nocturnally or diurnally, significant until sleep restriction is severe. However, this difference is
increases in the rate of accidents on a driving simulator not always apparent to the individual. It has been postu-
occurred with decreased sleep durations, independent of lated that these individual differences are a result of state
the timing of the sleep period.51 (basal level of sleepiness/alertness and basal differences in
circadian phase) and trait differences (optimal circadian
Subjective Sleepiness and Mood phase for sleep/wakefulness, sensitivity/responsiveness of
In contrast to the continuing accumulation of cognitive sleep homeostat, and compensatory mechanisms),52 but
performance deficits associated with nightly restriction of these factors have not been widely researched. For studies
sleep to below 8 hours, subjective ratings of sleepiness, of the possible genetic contributors to differential vulner-
fatigue, and related factors repeatedly made by subjects on ability to sleep loss, it is significant that the neurobehav-
standardized sleepiness scales did not parallel performance ioral responses to sleep deprivation have been found to be
deficits.4 As a consequence, after 2 weeks of sleep restricted stable and consistent within subjects,23 suggesting they are
to 4 or 6 hours per night, subjects were markedly impaired traitlike.23
CHAPTER 6 Chronic Sleep Deprivation 73

interleukin-6,62 and soluble tumor necrosis factoralpha

PHYSIOLOGIC EFFECTS receptor 161 have all been reported with total sleep depriva-
There is increasing evidence of physiologic and health- tion and sleep restriction.
related consequences of chronic sleep restriction. Altera- One study reported that antibody titers were decreased
tions in other physiologic parameters, such as endocrine by more than 50% after 10 days in subjects who were vac-
(see Chapter 125) and immune function (see Chapters 25 cinated for influenza immediately after 6 nights of sleep
and 26), have been recognized and have implications for restricted to 4 hours per night, compared with those who
health status and risk. were vaccinated after habitual sleep duration.63 But by 3 to
Several anecdotal and longitudinal studies have reported 4 weeks after the vaccination, there was no difference in
an increased incidence and risk of medical disorders and antibody level between the two subject groups. Therefore,
health dysfunction related to shift work schedules, which sleep loss appeared to alter the acute immune response to
have been attributed to both circadian disruption and sleep vaccination.
disturbance. Further links between sleep disturbance and
health effects have been reported in studies examining Cardiovascular Effects
insomniac patients and patients with other sleep disorders Increased cardiovascular events and cardiovascular mor-
and medical disorders that disturb sleep.53 In addition, an bidity have been reported with reduced sleep durations.11,54
elevated mortality risk in those individuals who reported Additionally, this relationship has been found in a case-
sleeping less than 6.5 hours per night has been found.11 control study examining insufficient sleep resulting from
One provocative discovery from this study was the finding work demands.64 In the Nurses Health Study, Ayas and
that individuals sleeping more than 7.4 hours per night colleagues54 reported that coronary events were increased
were also at an elevated risk of all-cause mortality. This in female subjects obtaining 7 hours of sleep or less per
finding is similar to that reported from the Nurses Health night compared with those averaging 8 hours per night,
Study,54 where subjects reporting greater than 9 hours of and Liu and associates64 reported a twofold to threefold
sleep per night on average were at a higher risk for coro- increase in risk of cardiovascular events with an average
nary events than those sleeping 8 hours per night. In addi- sleep duration of 5 hours or less per night. Shift workers
tion, an increased risk of coronary events in women who typically experience chronic reductions in sleep time
obtaining 7 hours of sleep or less per night was observed.55 as well as circadian disruption have been found to have
Also, increasing epidemiologic, cross-sectional, and longi- reduced cardiovascular health.65
tudinal data suggest that reduced sleep duration is associ- The mechanism that links chronic sleep restriction and
ated with larger body mass index (BMI). A meta-analysis increased cardiovascular risk is unknown, but one potential
study found a consistent, increased risk of obesity among pathway may be via activation of inflammatory processes
short sleepersboth children (sleeping less than 10 hours during sleep loss. C-reactive protein (CRP) is a predictive
per night) and adults (sleeping less than 5 hours per inflammatory marker of increased risk for cardiovascular
night)but, as the authors pointed out, causal inference disease. Increased CRP levels have been found in patients
was difficult because of the lack of control of confounders with obstructive sleep apnea, who commonly experience
and inconsistency in the methodologies used.56 reduced sleep time as well as hypoxia,66 and an increase in
CRP levels was reported after both total sleep deprivation
Endocrine and Metabolic Effects and sleep restriction (4 hours in bed for sleep per night) in
A number of studies have examined the effects of sleep loss healthy subjects.67
on a range of neuroendocrine factors.57,57a Comparison of
sleep restriction (4 hours per night for 6 nights) with sleep
Clinical Pearl
extension (12 hours per night for 6 nights) revealed an
elevation in evening cortisol, increased sympathetic activa- Chronic sleep deprivation can be caused by sleep
tion, decreased thyrotropin activity, and decreased glucose disorders, work schedules, and modern lifestyles.
tolerance in the restricted as opposed to the extended sleep Regardless of its cause, chronic sleep deprivation
condition.55 Similarly, an elevation in evening cortisol results in cumulative adverse effects in daytime
levels and an advance in the timing of the morning peak awake functions, including sleep propensity, cogni-
in cortisol, so that the relationship between sleep termina- tive performance, driving safety, mood, and physio-
tion and cortisol acrophase was maintained, were found logic conditions.
after 10 nights of sleep restricted to 4.2 hours of time in
bed for sleep each night compared with baseline measures
and a control group allowed 8.2 hours of time in bed for Acknowledgments
sleep for 10 nights.58 Changes in the timing of the growth The substantive evaluation on which this chapter was
hormone secretory profile associated with sleep restriction based was supported by National Institutes of Health
to 4 hours per night for 6 nights, with a bimodal secretory grants NR04281 and CTRC UL1RR024134, and by the
pattern evolving, have also been reported.59 National Space Biomedical Research Institute through
NASA NCC 9-58.
Immune and Inflammatory Effects
The majority of studies examining sleep loss and immune REFERENCES
function have concentrated on total sleep deprivation or 1 1. Webb WB. Partial and differential sleep deprivation. In: Kales A,
night of sleep restriction. Changes in natural killer cell editor. Sleep: physiology and pathologya symposium. Philadelphia:
activity,60,61 lymphokine-activated killer cell activity,60 JB Lippincott; 1969. p. 221-231.
74 PARTI / Section 1 Normal Sleep and Its Variations

2. Dement WC, Vaughan C. The promise of sleep. New York: Dell; 30. Guilleminault C, Powell NB, Martinez S, Kushida C, et al. Prelimi-
1999. nary observations on the effects of sleep time in a sleep restriction
3. Kleitman N. Sleep and wakefulness. 2nd ed. Chicago: University of paradigm. Sleep Med 2003;4(3):177-184.
Chicago Press; 1963. 31. Punjabi NM, Bandeen-Roche K, Young T. Predictors of objective
4. Van Dongen HPA, Maislin G, Mullington JM, Dinges DF. The sleep tendency in the general population. Sleep Med 2003;26(6):678-
cumulative cost of additional wakefulness: dose-response effects on 683.
neurobehavioral functions and sleep physiology from chronic sleep 32. Banks S, Van Dongen H, Dinges DF. How much sleep is needed to
restriction and total sleep deprivation. Sleep 2003;26(2):117-126. recover from sleep debt?the impact of sleep dose on recovery.
5. Hartley LR. Comparison of continuous and distributed reduced sleep Sleep 2005;28:A138.
schedules. Q J Exp Psychol 1974;26:8-14. 33. Dinges DF, Mallis M, Maislin G, Powell JW. Evaluation of tech-
6. Friedmann J, Globus G, Huntley A, Mullaney D, et al. Performance niques for ocular measurement as an index of fatigue and the basis
and mood during and after gradual sleep reduction. Psychophysiol- for alertness management. Final report for the U.S. Department of
ogy 1977;14:245-250. Transportation, National Highway Traffic Safety Administration.
7. Ferrara M, De Gennaro L, Bertini M. The effects of slow-wave sleep 1998;1-112.
(SWS) deprivation and time of night on behavioral performance 34. Mallis MM, Dinges DF. Monitoring alertness by eyelid closure.
upon awakening. Physiol Behav 1999;68(1-2):55-61. In: Stanton N, Hedge A, Brookhuis K, Salas E, editors. The hand-
8. Dinges DF, Pack F, Williams K, Gillen KA, et al. Cumulative sleepi- book of human factors and ergonomics methods. New York: CRC
ness, mood disturbance, and psychomotor vigilance performance Press; 2005. p. 25.
decrements during a week of sleep restricted to 4-5 hours per night. 35. Lockley SW, Cronin JW, Evans EE, Cade BE, et al. Effect of reduc-
Sleep 1997;20(4):267-277. ing interns weekly work hours on sleep and attentional failures.
9. Webb WB, Agnew HW. Effects of a chronic limitation of sleep N Eng J Med 2004;351(18):1829-1837.
length. Psychophysiology 1974;11(3):265-274. 36. Russo M, Thomas M, Thorne D, Sing H, et al. Oculomotor impair-
10. Dinges DF. Sleep debt and scientific evidence. Sleep 2004;27(6): ment during chronic partial sleep deprivation. Clin Neurophysiol
1050-1052. 2003;114(4):723-736.
11. Kripke DF, Garfinkel L, Wingard DL, Klauber MR, et al. Mortality 37. Cirelli C, Tononi G. Is sleep essential? PLoS Biol 2008;6(8):e216.
associated with sleep duration and insomnia. Arch Gen Psychiatry 38. Dorrian J, Dinges DF. Sleep deprivation and its effects on cognitive
2002;59(2):131-136. performance. In: Dorrian JS, Dinges DF, editors. Encyclopedia of
12. 2002 Sleep in America PollNational Sleep Foundation. Washing- sleep medicine. Hoboken, NJ:1 John Wiley & Sons; 2005.
ton DC: 2002. 39. Durmer JS, Dinges DF. Neurocognitive consequences of sleep depri-
13. Dement W, Greenber S. Changes in total amount of stage 4 sleep as vation. Semin Neurol 2005;25:117-129.
a function of partial sleep deprivation. Electroencephalogr Clin Neu- 40. Lim J, Dinges DF. Sleep deprivation and vigilant attention. Molecu-
rophysiol 1966;20(5):523-526. lar and biophysical mechanisms of arousal, alertness, and attention.
14. Van Dongen HPA, Rogers NL, Dinges DF. Understanding sleep Ann N Y Acad Sci 2008;1129:305-322.
debt: Theoretical and empirical issues. Sleep and Biol Rhythms 2003; 41. Dinges DF, Kribbs NB. Performing while sleepy: effects of
1:4-12. experimentally-induced sleepiness. In: Monk TH, editor. Sleep,
15. Wehr TA, Moul DE, Barbato G, Giesen HA, et al. Conservation of sleepiness and performance. Chichester: John Wiley & Sons; 1991.
photoperiod-responsive mechanisms in humans. Am J Physiol p. 97-128.
1993;265(4):R846-R857. 42. Doran SM, Van Dongen HPA, Dinges DF. Sustained attention per-
16. Horne J, editor. Why we sleep: the functions of sleep in humans and formance during sleep deprivation: evidence of state instability. Arch
other mammals. Oxford: Oxford University Press; 1988. Ital Biol 2001;139:253-267.
17. Horne JA. Sleep function, with particular reference to sleep- 43. Chee MW, Tan JC, Zheng H, Parimal S, et al. Lapsing during sleep
deprivation. Ann Clin Res 1985;17(5):199-208. deprivation is associated with distributed changes in brain activation.
18. Akerstedt T, Folkard S. The three-process model of alertness and its J Neurosci 2008;28(21):5519-5528.
extension to performance, sleep latency, and sleep length. Chrono- 44. Herscovitch J, Broughton R. Performance deficits following short-
biol Int 1997;14(2):115-123. term partial sleep-deprivation and subsequent recovery over-
19. Drake CL, Roehrs TA, Burduvali E, Bonahoom A, et al. Effects of sleeping. Can J Psychol 1981;35(4):309-322.
rapid versus slow accumulation of eight hours of sleep loss. Psycho- 45. Mollicone DJ, Van Dongen HPA, Rogers NL, Dinges DF. Response
physiology 2001;38(6):979-987. surface mapping of neurobehavioral performance: testing the feasi-
20. Borbely AA. A two process model of sleep regulation. Hum Neuro- bility of split sleep schedules for space operations. Acta Astronautica
biol 1982;1(3):195-204. 2008;63(7):833-840.
21. Borbely AA, Achermann P. Sleep homeostasis and models of sleep 46. Rogers NL, Van Dongen HPA, Powell JW, Carlin MM, et al. Neu-
regulation. J Biol Rhythms 1999;14(6):557-568. robehavioural functioning during chronic sleep restriction at an
22. Dijk DJ, Czeisler CA. Paradoxical timing of the circadian-rhythm of adverse circadian phase. Sleep 2002;25(Abstract Suppl.):A126-A127.
sleep propensity serves to consolidate sleep and wakefulness in 47. Lamond N, Jay SM, Dorrian J, Ferguson SA, et al. The dynamics of
humans. Neurosci Let 1994;166(1):63-68. neurobehavioural recovery following sleep loss. J Sleep Res 2007;
23. Van Dongen HPA. Comparison of mathematical model predictions 16(1):33-41.
to experimental data of fatigue and performance. Aviat Space Environ 48. Porkka-Heiskanen T, Strecker RE, Thakkar M, Bjorkum AA, et al.
Med 2004;75(3):A15-A36. Adenosine: a mediator of the sleep-inducing effects of prolonged
24. Belenky G, Wesensten NJ, Thorne DR, Thomas ML, et al. Patterns wakefulness. Science 1997;276(5316):1265-1268.
of performance degradation and restoration during sleep restriction 49. Stutts JC, Wilkins JW, Osberg JS, Vaughn BV. Driver risk factors for
and subsequent recovery: a sleep dose-response study. J Sleep Res sleep-related crashes. Accid Anal Prev 2003;35(3):321-331.
2003;12(1):1-12. 50. Banks S, Catcheside P, Lack L, Grunstein RR, et al. Low levels of
25. Brunner DP, Dijk DJ, Borbely AA. Repeated partial sleep-depriva- alcohol impair driving simulator performance and reduce perception
tion progressively changes the EEG during sleep and wakefulness. of crash risk in partially sleep deprived subjects. Sleep 2004;
Sleep 1993;16(2):100-113. 27(6):1063-1067.
26. Carskadon MA Dement WC. The multiple sleep latency testwhat 51. Dorrian J, Dinges DF, Rider RL, Price NJ, et al. Simulated driving
does it measure. Sleep 1982;5:S67-S72. performance during chronic partial sleep deprivation. Sleep 2003;
27. Mitler MM, Gujavarty KS, Browman CP. Maintenance of wakeful- 26:A182-A183.
ness testa polysomnographic technique for evaluating treatment 52. Roth T, Roehrs T. The timing of sleep opportunities in a seven-night
efficacy in patients with excessive somnolence. Electroencephalogr sleep restriction. Sleep Med 2003;4(3):169-170.
Clin Neurophysiol 1982;53(6):658-661. 53. Bonnet MH. Sleep deprivation. In: Kryger MH, Roth T, Dement
28. Carskadon MA, Dement WC. Cumulative effects of sleep restriction WC, editors. Principles and practice of sleep medicine. Philadelphia:
on daytime sleepiness. Psychophysiology 1981;18(2):107-113. Saunders; 1994. p. 50-67.
29. Banks S, Barnes M, Tarquinio N, Pierce RJ, et al. The maintenance 54. Ayas NT, White DP, Manson JE, Stampfer MJ, et al. A prospective
of wakefulness test in normal healthy subjects. Sleep 2004;27(4): study of sleep duration and coronary heart disease in women. Arch
799-802. Int Med 2003;163(2):205-209.
CHAPTER 6 Chronic Sleep Deprivation 75

55. Spiegel K, Leproult R, Van Cauter E. Impact of sleep debt on meta- to the sleep deprivation model of spaceflight. J Allergy Clin Immunol
bolic and endocrine function. Lancet 1999;354(9188):1435-1439. 2001;107(1):165-170.
56. Cappuccio FP, Taggart FM, Kandala NB, Currie A, et al. Meta- 62. Redwine L, Hauger RL, Gillin JC, Irwin M. Effects of sleep and
analysis of short sleep duration and obesity in children and adults. sleep deprivation on interleukin-6, growth hormone, cortisol, and
Sleep 2008;31(5):619-626. melatonin levels in humans. J Clin Endocrinol Metab 2000;
57. Knutson KL, Van Cauter E. Associations between sleep loss and 85(10):3597-3603.
increased risk of obesity and diabetes. Ann N Y Acad Sci 2008; 63. Spiegel K, Sheridan JF, Van Cauter E. Effect of sleep deprivation on
1129:287-304. response to immunization. JAMA 2002;288(12):1471-1472.
57a. Leproult R, Van Cauter E. Role of sleep and sleep loss in hormonal 64. Liu Y, Tanaka H, the Fukuoka Heart Study Group. Overtime work,
release and metabolism. Endocr Dev 2010;17:11-21. insufficient sleep, and risk of non-fatal acute myocardial infarction in
58. Rogers NL, Price NJ, Mullington JM, Szuba MP, et al. Plasma cor- Japanese men. Occup Environ Med 2002;59:447-451.
tisol changes following chronic sleep restriction. Sleep 2000;23: 65. Knutsson A, Hallquist J, Reuterwall C, Theorell T, et al. Shiftwork
A70-A71. and myocardial infarction: a case-control study. Occup Environ Med
59. Spiegel K, Leproult R, Colecchia EF, LHermite-Baleriaux M, et al. 1999;56(1):46-50.
Adaptation of the 24-h growth hormone profile to a state of 66. Lee LA, Chen NH, Huang CG, et al. Patients with severe obstruc-
sleep debt. Ame J Physiol Regul Integr Comp Physiol 2000;279(3): tive sleep apnea syndrome and elevated high-sensitivity C-reactive
R874-R883. protein need priority treatment. Otolaryngol Head Neck Surg
60. Irwin M, McClintick J, Costlow C, Fortner M, et al. Partial night 2010;143(1):72-77.
sleep deprivation reduces natural killer and cellular immune responses 67. van Leeuwen WM, Lehto M, Karisola P, et al. Sleep restriction
in humans. FASEB J 1996;10(5):643-653. increases the risk of developing cardiovascular diseases by augment-
61. Shearer WT, Reuben JM, Mullington JM, Price NJ, et al. Soluble ing proinflammatory responses through IL-17 and CRP. PLoS One
TNF alpha receptor 1 and IL-6 plasma levels in humans subjected 2009;4(2):e4589.
Sleep Mechanisms Section

and Phylogeny 2
Jerome M. Siegel

7 Neural Control of Sleep in Mammals

8 REM Sleep
9 Phylogeny of Sleep Regulation
10 Sleep in Animals: A State of Adaptive Inactivity

Neural Control of Sleep in Mammals
Dennis McGinty and Ronald Szymusiak
Abstract Although several brain regions modulate sleep, the POA
Mammalian sleep and wake states are facilitated by multiple has a critical role in the control of NREM sleep. Groups of
brain regions. The lower brainstem is sufficient to generate neurons in the POA exhibit increased activity during NREM
wake, nonrapid eye movement (NREM)-like, and REM sleep and REM sleep and respond to physiologic signals, such as
states, but lesions in several brain regions, including sites in warming, that increase sleep. Several putative neurochemical
the medulla, mesencephalon, preoptic area (POA) of the hypo- sleep factors promote sleep through actions in the POA or
thalamus, thalamus, and neocortex, all markedly reduce the adjacent basal forebrain. Signals from the circadian clock
amounts of NREM and REM sleep. Similarly, arousal and originating in the hypothalamic suprachiasmatic nucleus
waking are facilitated by several chemically distinct neuronal regulate the circadian timing of sleep through connections in
groups localized in the midbrain, the posterior and lateral this brain region. POA sleep-active neurons send inhibitory
hypothalamus, and the basal forebrain. These include hista- gamma-aminobutyric acid (GABA)ergic projections to the his-
minergic, orexinergic, serotonergic, cholinergic, dopaminer- taminergic, orexinergic, serotonergic, and noradrenergic
gic, and noradrenergic neurons. These arousal systems share arousal systems. Through coordinated inhibition of these
the property of having long axons and extensive projections arousal systems, the key elements of NREM sleep onset are
to widespread brain regions, including the diencephalon, enabled, including EEG synchronization and suppression of
limbic system, and neocortex. These widespread projections motor activity.
can account for the global aspect of arousal from sleep, We take it for granted that the brain controls sleep and
characterized by concurrent changes in electroencephalo- waking, and this has been confirmed. Research spanning the
graphic (EEG), motor, sensory, autonomic, and integrative past 80 years has identified specific groups of neurons and
functions. Arousal systems also control the neuronal mem- neurochemical mechanisms that carry out the control of sleep
brane potentials in thalamocortical neurons; these potentials, and waking and generate core aspects of the phenomena of
in turn, regulate the oscillatory mechanisms intrinsic to thala- sleep and waking, such as the electroencephalographic (EEG)
mocortical networks that underlie NREM or synchronized EEG patterns that define these states. Many methods have been
patterns. Inhibition of arousal systems permits the emergence used. The story is complex, but a surprisingly coherent picture
of synchronized EEG patterns. of sleepwake control has emerged.

DIVERSE BRAIN REGIONS mid-pontine level, an activated or wakelike forebrain EEG

MODULATE WAKING AND state becomes predominant immediately after the transec-
NREM SLEEP tion, but with some residual episodes of EEG slow-wave
activity.2 In this preparation, the forebrain exhibits evi-
Isolated Forebrain dence of conditioning, and other signs of an integrated
The capacity of various brain regions to generate sleep and waking state. These studies argue that neuronal groups
wake states was first studied by isolating or removing major localized between the midpons and upper midbrain are
regions. The physiology of the chronically maintained iso- important for generating a waking-like state. After 5 to 9
lated forebrain or chronic cerveau isol preparation can be days of recovery from surgery, the chronic cerveau isol rat
examined in dogs and cats.1 Acutely after complete mid- preparation exhibits a circadian pattern of EEG activation
brain transections, the isolated forebrain exhibits continu- and synchronization.3 In this preparation, preoptic area
ous EEG slow waves and spindles. Thus, structures below the (POA) lesions are followed by a continuously activated
midbrain must normally facilitate awake-like EEG states. EEG, abolishing the circadian facilitation of synchroniza-
However, if a brainstem transection is made lower, at the tion. Thus, the isolated forebrain can generate a sustained wake-

CHAPTER 7 Neural Control of Sleep in Mammals 77

like state, and the POA must play a critical role in initiating tion in the nucleus of the solitary track produced neocortical
the sleeplike EEG state of the isolated forebrain (see later). EEG synchronization.7 Lesions or cooling of this site were
Wakelike and sleeplike EEG states appear to depend on a balance followed by EEG activation.8
between wake-promoting and sleep-promoting systems. In summary, widespread structures in the mammalian
nervous system, from the neocortex to the lower brainstem, have
Diencephalon the capacity to facilitate both sleeplike and waking-like states and
Chronic diencephalic cats, whose neocortex and striatum to modulate the amounts of sleep.
have been removed, exhibit behavioral waking with persis-
tent locomotion and orientation to auditory stimuli, a RETICULAR ACTIVATING SYSTEM
quiet sleeplike or nonrapid eye movement (NREM)-like
state with typical cat sleeping postures, and a REM-like AND DELINEATION OF
state including antigravity muscle atonia, rapid eye move- AROUSAL SYSTEMS
ments, muscle twitches, and pontine EEG spikes.4 EEG The transection studies just reviewed support the concept
patterns recorded in the thalamus showed increased ampli- of a pontomesencephalic wake-promoting or arousal
tude in conjunction with the NREM-sleeplike state, system. No discovery was historically more significant than
although true spindles and slow waves are absent. The the description of the reticular activating system (RAS) by
thalamic EEG exhibits desynchronization during the Moruzzi.9 Large lesions of the core of the rostral pontine
REM-like state. In summary, at least in the cat, the neocortex and mesencephalic tegmentum are followed by persistent
and striatum are not required for any behaviorally defined sleep somnolence and EEG synchronization, and electrical stim-
wake states, and an NREM-like state occurs in the absence of ulation of this region induces arousal from sleep. Interrup-
sleep spindles and slow waves. tion of sensory pathways does not affect EEG activation.
It was hypothesized that cells in the RAS generated fore-
Thalamus brain activation and wakefulness.
Cats subjected to complete thalamectomy (athalamic cats) The concept of the RAS has been superseded by the
continue to exhibit episodes of EEG and behavioral sleep finding that arousal is facilitated not by a single system but
and waking, although there is an absence of spindles in the instead by several discrete neuronal groups localized within
NREM sleep EEG,4 and they exhibit chronic insomnia, and adjacent to the pontine and midbrain reticular forma-
with reductions in both NREM and REM sleep. Fatal tion and its extension into the hypothalamus (Fig. 7-1).
familial insomnia,5 a neurodegenerative disease character- These discrete neuronal groups are identified and differ-
ized by progressive autonomic hyperactivation, motor dis- entiated by their expression of molecular machinery that
turbances, loss of sleep spindles, and severe NREM sleep synthesizes and releases specific neurotransmitters and
insomnia, typically begins after age 40 years. Neuropatho- neuromodulators. These include neuronal groups that
logic findings reveal initial severe cell loss and gliosis in synthesize serotonin, noradrenalin, histamine, acetylcho-
the anterior medial thalamus, including the dorsomedial line, and orexin/hypocretin (herein called orexin). Each of
nucleus. However, patients with paramedian thalamic these systems has been studied extensively in the context
stroke, with magnetic resonance imaging (MRI)-verified of the control of specific aspects of waking behaviors. Here
damage to the dorsomedial and centromedial nuclei, we will give only a brief overview of each, focusing on their
present with either severe hypersomnolence or increased contribution to generalized brain arousal or activation.
daytime sleepiness, not insomnia.6 In summary the thalamus Before proceeding, we point out certain general properties
plays a critical role in regulating cortical EEG patterns during of these neuronal systems.
waking and sleep, and portions of this structure appear to have 1. Arousal is a global process, characterized by concurrent
hypnogenic functions. changes in several physiologic systems, including auto-
nomic, motor, endocrine, and sensory systems, and in
Lower Brainstem EEG tracings. Thus, it is intriguing that most arousal
After recovery from the acute effects of the complete mid- systems share one critical property: the neurons give
brain transections (described earlier), the lower brainstem rise to long, projecting axons with extensive terminal
can generate rudimentary behavioral waking, a NREM- fields that impinge on multiple regions of the brainstem
like state, and a REM-like state.4 The behavior of the and forebrain. These diffuse projections enable the
midbrain-transected cats could be characterized as having systems to have multiple actions, as might be expected
three states, including waking (identified by crouching, of arousal systems. In this review, we emphasize the
sitting, attempts to walk, dilated pupils, and head orienta- ascending projectionsthat is, projections from the
tion to noises), and two sleeplike states. In the first sleep- brainstem and hypothalamus to the diencephalon,
like state, cats lay down in a random position, pupils limbic system, and neocortex, as these are particularly
exhibited reduced but variable miosis, and eyes exhibited germane to the generation of cortical arousal. Some
slow and nonconjugate movements, and they could be arousal systems also give rise to descending projections,
aroused by auditory or other stimuli. If this stage is not which are also likely to play a role in regulating certain
disturbed, cats enter another stage, characterized by com- properties of sleepwake states, such as changes in
plete pupillary miosis, loss of neck muscle tone, and rapid muscle tone.
eye movements, identifying a REM-like state. Additional 2. The release of neurotransmitters and neuromodulators
studies support the hypothesis that the lower brainstem at nerve terminals is initiated by the propagation of
contains sleep-facilitating processes. Low-frequency elec- action potentials to the terminals. Thus, neurotransmit-
trical stimulation of the dorsal medullary reticular forma- ter release is correlated with the discharge rate of
78 PARTI / Section 2 Sleep Mechanisms and Phylogeny

its action. Pharmacologic actions are usually mediated

by actions on specific receptor types or transporters (see
examples later).
4. Chronic lesions of individual arousal systems or genetic
knockout (KO) of critical molecules have only small or
sometimes no effect on sleepwake patterns (with the
exception of serotonin and orexin KOs; see later), even
BF Thalamus
though acute manipulations of these same systems have
strong effects on sleepwake. The absence of chronic
TMN PPT (ACh) lesions or KO effects is probably explained by the
(HA) SN Raphe LDT redundancy of the arousal systems, such that, over time,
VTA (5-HT)
LC deficiency in one system is compensated for by other
(DA) (NE) systems or by changes in receptor sensitivity. Electro-
physiologic studies show that the arousal systems are
A formation normally activated and deactivated within seconds or
minutes. Thus, effects of acute experimental manipula-
tions of particular arousal-related neurotransmitters, as
with administration of a drug, may better mimic the
Cortex normal physiologic pattern and be more informative as
to their function.
Orexin 5. REM sleep is, on one hand, a sleep state, but, on the
other hand, it is associated with neocortical EEG char-
BF acteristics of wake. In parallel with these two sides of
PPT REM, it has been shown that arousal systems can be
TMN Raphe LDT classified into two types, ones that are off in REM,
VTA LC befitting the sleeplike property of REM, and others that
are on in REM, befitting the wakelike properties of
REM. Some arousal-promoting systems (summarized
later) also play a role in REM control. Detailed analyses
B of the control of the role of these systems in REM sleep
can be found in Chapters 8 and 9.
Figure 7-1 A, Sagittal view of a generic mammalian brain
providing an overview of the wake-control networks described
in the text. The upper brainstem, posterior and lateral hypo- WAKE-ON, REM-OFF
thalamus, and basal forebrain contain several clusters of AROUSAL SYSTEMS
neuronal phenotypes, with arousal-inducing properties. These
clusters include neurons expressing serotonin (5-HT), nor- Serotonin
epinephrine (NE), acetylcholine (ACh) in both pontomesen-
cephalic and basal forebrain clusters, dopamine (DA), and Neurons containing serotonin, or 5-hydroxytryptamine
histamine (HA). B, Sagittal view of brainstem and diencepha- (5-HT), innervate the forebrain and are found in the dorsal
lon showing localization of orexin-containing neurons and raphe (DR) and median raphe (MR) nuclei of the mid-
their projections to both forebrain and brainstem. All of brain. These neurons project to virtually all regions of the
these groups facilitate EEG arousal (waking and REM) and/
or motor-behavioral arousal (waking). The arousal systems
diencephalon, limbic system, and neocortex. Although it
facilitate forebrain EEG activation both through the thalamus was initially hypothesized that serotonin might be a sleep-
and the basal forebrain and through direct projections to promoting substance,10 much evidence shows that the
neocortex. Arousal systems also facilitate motor-behavioral immediate effect of release of serotonin is arousal (reviewed
arousal through descending pathways. in reference 11). Although there is some heterogeneity,
the discharge rates of most DR and MR neurons are
highest during waking, lower during NREM, and there
is minimal discharge in REM; release of serotonin in the
neurons. Most arousal systems have been studied by forebrain is highest in waking. Because of the great diver-
recording the discharge patterns of neurons in freely sity of serotonin receptors (there are at least 14 types),
moving animals, in relationship to spontaneously the effects of serotonin on target neurons are complex.
occurring wake and sleep states. Increased discharge Some receptor types are inhibitory, some are excitatory.
during arousal or wake compared with sleep constitutes At least one class of receptors (5-HT2A) appears to facilitate
part of the evidence for an arousal system. NREM sleep; 5-HT2A KO mice have less NREM.12
3. The actions of a neurotransmitter on a target system Another type (5-HT1A) is inhibitory to REM sleep, as
are determined primarily by the properties of the recep- 5-HT1A KO mice have increased REM.13 Selective sero-
tors in the target. The neurotransmitters and neuro- tonin reuptake inhibitors (SSRIs) and serotoninnorepi-
modulators underlying arousal systems each act on nephrine reuptake inhibitors (SNRIs) are used to treat a
several distinct receptor types, with diverse actions. In variety of medical and psychiatric problems, and some
addition, postsynaptic effects are regulated by transmit- drugs in this class have arousing or alerting properties.
ter-specific reuptake molecules, which transport the Serotonin has a wide range of functions in addition to the
neurotransmitter out of the synaptic space, terminating modulation of sleepwake.
CHAPTER 7 Neural Control of Sleep in Mammals 79

cephalic reticular formation (RF) (the pedunculopontine

Norepinephrine tegmental [PPT] and laterodorsal tegmental [LDT]
Norepinephrine (NE)-containing neuronal groups in nuclei) and in the basal forebrain.25 The pontomesence-
mammals are found throughout the brainstem, but the phalic ACH neuronal group projects to the thalamus,
primary nucleus giving rise to ascending projections is hypothalamus, and basal forebrain; the basal forebrain
the locus coeruleus (LC). NE neurons in the LC project group projects to the limbic system and neocortex.
throughout the diencephalon, forebrain, and cerebellum. Neurons in both groups exhibit higher rates of discharge
LC neurons exhibit regular discharge during waking, in both waking and REM than in NREM sleep,26 and
reduced discharge during NREM sleep, and near-com- release of ACH is also increased in these states.27 Phar-
plete cessation of discharge in REM sleep, a pattern macologic blockade of ACH receptors induces EEG syn-
congruent with a role in behavioral arousal.14 Acute inac- chrony and reduces vigilance, and inhibition of the
tivation of the LC or a lesion in the ascending pathway ACH-degrading enzyme cholinesterase enhances arousal.28
from the LC increases slow-wave EEG activity during
sleep.15 Distinct roles for alpha-1, alpha-2, and beta NE Dopamine
receptor types are established. Direct application of Dopamine (DA)-containing neurons are primarily local-
alpha-1 and beta agonists in preoptic area and adjacent ized in the substantia nigra and the adjacent ventral
basal forebrain sites induces increased wakefulness tegmental area of the midbrain and the basal and medial
(reviewed in reference 16). The arousal-producing effects hypothalamus.29 Putative dopaminergic neurons exhibit
of psychostimulant drugs such as amphetamines depend highest activity in waking and REM sleep. Release of
partly on induction of increased NE release and inhibi- DA in the frontal cortex is higher during wakefulness
tion of NE reuptake, as well as on enhanced dopamine than during sleep.30 DA is inactivated primarily through
action (see later). reuptake by the dopamine transporter. Stimulant drugs
such as amphetamines and modafinil act primarily through
Histamine DA receptors, particularly by binding to and suppressing
Histamine (HA)-containing neurons in mammals are dis- the dopamine transporter, reducing reuptake.31 The
cretely localized in the tuberomamillary nucleus (TMN) degeneration of the nigrostriatal DA system is the primary
and adjacent posterior hypothalamus (PH). HA neurons neuropathologic basis of Parkinsons disease, and exces-
project throughout the hypothalamus and forebrain, sive daytime sleepiness is one manifestation of it.32 DA
including the neocortex, as well as to the brainstem and agonists used to treat periodic limb movement in sleep
spinal cord. Perhaps the most familiar evidence for an and restless leg syndrome do not usually induce arousal,
arousal-promoting action of central HA is that administra- probably because they have effects on both presynaptic
tion of histamine (H1)-receptor antagonists (antihista- and postsynaptic DA receptors; presynaptic receptors
mines) that penetrate the bloodbrain barrier cause inhibit transmitter release, counteracting postsynaptic
sedation. Transient inactivation of the TMN region results stimulation.
in increased NREM sleep.17 HA neurons exhibit regular
discharge during waking, greatly reduced discharge during Glutamate
NREM sleep, and cessation of discharge in REM sleep.18 Glutamate (Glu)-containing neurons are found through-
HA neurons express the inhibitory H3-type autoreceptors. out the brain, including in the core of the pontine and
Administration of an antagonist of this receptor causes midbrain RF.33 However, the projections of the brainstem
disinhibition and increased waking. Blockade of the critical Glu system have not been described. Glu is the primary
HA-synthesizing enzyme increases NREM sleep.19 excitatory neurotransmitter of the brain. Arousal is
increased by application of Glu to many sites.34 Glu
Orexin actions are mediated by receptors controlling membrane
The loss of orexin neurons is known to underlie the human ion flux, including the N-methyl-d-aspartate (NMDA)
disease narcolepsy, whose major symptoms are cataplexy receptor, and metabotrophic receptors controlling
and excessive sleepiness.19-20a Orexin-containing neurons intracellular processes. Humans may be exposed to sys-
are localized in the midlateral hypothalamus, and like temic administration of NMDA receptor antagonists in
other arousal systems, they give rise to projections to all the form of anesthetics (e.g., ketamine) or recreational
brain regions including the brainstem.21 Among the targets drugs (PCP). The effects are dosage dependent: low
of orexin terminals are other arousal-promoting neurons dosages produce arousal, and high dosages produce seda-
including HA, 5-HT, and NE neurons. Orexin-containing tion. In rats, exposure to NMDA antagonists induces
neurons are active in waking, and they are off in both a potent long-lasting enhancement of NREM slow-wave
NREM and REM sleep.22,23 Local administration of orexin activity.35
in several brain sites induces arousal.24 (See also Chapters
8 and 16.)
WAKE-ON, REM-ON AROUSAL We have reviewed evidence for multiple neurochemically
SYSTEMS specific arousal systems and noted that the activity of each
of these neuronal groups is reduced during NREM sleep.
Acetylcholine In most groups, the reduction in neuronal discharge pre-
Groups of acetylcholine (ACH)-containing neurons are cedes EEG changes that herald sleep onset. How is the
localized in two regions: in the dorsolateral pontomesen- process of sleep onset orchestrated?
80 PARTI / Section 2 Sleep Mechanisms and Phylogeny


A POA Sleep-Promoting System
Von Economo postulated a POA sleep-promoting area
more than 70 years ago on the basis of his observation that
postmortem examinations of patients who had had enceph-
alitis and severe insomnia showed inflammatory lesions in
this area of the brain.36 Patients with hypersomnia had
lesions in the vicinity of the PH. To von Economo, this
suggested the concept of opposing hypothalamic sleep- A B
promoting and wake-promoting systems. In rats, sym- WAKE WAKE
metrical bilateral transections of the POA resulted in
complete sleeplessness, and symmetrical bilateral transec-
tions of the PH caused continuous sleep.37 Rats with both
POA and PH transections exhibited continuous sleep, as
with PH transections alone. This was interpreted as
showing that the POA normally inhibits the PH wake-
promoting region. The PH wake-promoting system can
now be understood as the rostral extension of arousal-
promoting systems and pathways summarized earlier. 600 m 300 m
The existence of a sleep-promoting mechanism in the
POA has been confirmed by a variety of methods. Experi- ROSTRAL MnPN c-Fos COUNT CAUDAL MnPN c-Fos COUNT
mental lesions of this area result in insomnia. Bilateral VS. SLEEP PERCENTAGE VS. SLEEP PERCENTAGE
lesions of the POA with diameters of 1 to 2 mm in rats 160
r22 = 0.89
r22 = 0.8
and cats induce partial sleep loss. Larger bilateral lesions Cell count 120 p2 <0.0001 r12 = 0.35 p2 <0.0001

Cell count
p1 = 0.01 60
(3 to 5mm in diameter) that extend into the adjacent basal 80 40
forebrain yield more severe insomnia (reviewed in refer- 40 20 r12 = 0.67
ence 38). After POA lesions that result in partial sleep loss, 0 0 p1 = 0.0002
residual sleep is characterized by reduced slow-wave (delta) 0 20 40 60 80 100 0 20 40 60 80 100
EEG activity.39 Delta EEG activity in NREM sleep is A Sleep percentage (2 hours) B Sleep percentage (2 hours)
augmented by POA warming (see later). Because delta VLPO c-Fos COUNT VS.
activity is recognized as a marker of enhanced sleep drive, SLEEP PERCENTAGE
this finding suggests that POA output contributes to the r12 = 0.60
40 p1 = 0.0006
regulation of sleep drive. Electrical or chemical stimula-
Cell count

30 r22 = 0.08
tion of the POA evokes EEG synchronization and behav- 20
p2 = 0.37
ioral sleep. 10
c-Fos Mapping 0 20 40 60 80 100
C Sleep percentage (2 hours)
Lesion and stimulation studies argue that the POA must
contain sleep-active neurons. This has been confirmed in Control sleep (CS) Heat sleep (HS) Control sleep vs. control wake
Control wake (CW) Heat wake (HW) Heat sleep vs. heat wake
several species (reviewed in reference 38). The identifica-
tion of sleep-active POA neurons was advanced by the
Figure 7-2 Upper: Examples of c-Fos immunostaining of pre-
application of the c-Fos immunostaining method.40 Rapid optic area (POA) neuronal nuclei, identified by dark spots after
expression of the protooncogene c-fos has been identified either sustained spontaneous sleep or wake. c-Fos immunos-
as a marker of neuronal activation in many brain sites.41 taining is a marker of neuronal activation and is a method for
Thus, c-Fos immunostaining permits functional mapping mapping the localization of sleep-active neurons in the brain.
of neurons, identifying neurons that were activated in the After sleep, increased staining was seen in the midline (A) or
around the top of the third ventricle (B) compared with the
preceding interval. After sustained sleep, but not waking, wake samples (C and D). These sites correspond to the caudal
a discrete cluster of neurons exhibiting Fos immunoreac- and rostral median preoptic nucleus (MnPN). Similar results
tivity is found in the ventrolateral preoptic area (VLPO).40 were seen in the ventrolateral preoptic area (VLPO). In other
The VLPO is located at the base of the brain, lateral to POA sites, c-Fos immunostaining was seen following both
waking and sleep.
the optic chiasm. Sleep-related Fos immunoreactive Lower: Regression functions and correlations relating c-Fos
neurons are also localized in the rostral and caudal median counts and sleep amounts before sacrifice among individual
preoptic nucleus (MnPN).42 The MnPN is a midline cell animals. In all sites, high correlations were found between sleep
group that widens to form a cap around the rostral end amounts and c-Fos counts at a normal ambient temperature.
of the third ventricle. Examples of c-Fos immunostaining Groups of animals were studied in both normal and warm
ambient temperatures. In a warm ambient temperature, c-Fos
and the correlations between c-Fos counts and sleep counts after sleep and correlations between counts and sleep
amounts are shown in Figure 7-2. The number of sleep- amounts were increased in MnPN sites (A and B), but they were
related Fos immunoreactive neurons in the MnPN is suppressed in the VLPO (C). (From Gong H, Szymusiak R, King
increased in rats exposed to a warm ambient temperature, J, et al. Sleep-related c-Fos expression in the preoptic hypothala-
mus: effects of ambient warming. Am J Physiol Regul Integr
in association with increased NREM sleep.42 Comp Physiol 2000;279:R2079-R2088.)
The VLPO and the MnPN contain a high density of
neurons with sleep-related discharge.43 Most sleep-active
CHAPTER 7 Neural Control of Sleep in Mammals 81


500 V

Spikes sec1




40 sec

Figure 7-3 Example of sleep-active neurons in the median preoptic nucleus (MnPN). Shown is a continuous recording of discharge
of an MnPN neuron during a wake-NREM-REM cycle (top). Discharge rate increased at the onset of sleep, as indicated by the
increased amplitude of the EEG. Discharge rate increased further in association with REM sleep (right). Such sleep-active neurons
were the majority of neurons encountered in the median preoptic nucleus (MnPN) and the ventrolateral preoptic area (VLPO). The
presence of sleep-active neurons provides one critical piece of evidence for the importance of a brain region in the facilitation of
sleep. (From Suntsova N, Szymusiak R, Alam MN, et al. Sleep-waking discharge patterns of median preoptic nucleus neurons in
rats. J Physiol 2002;543:665-677.)

neurons in these nuclei are more activated during both rotransmitter, gamma-aminobutyric acid (GABA). The
NREM and REM sleep than in waking (Fig 7-3). A major- majority of MnPN neurons that exhibit sleep-related Fos-
ity of VLPO neurons exhibit an increase in activity during immunoreactivity also express GAD.44 VLPO neurons
the immediate transition periods between waking and send anatomic projections to HA neurons in the TMN.45
sleep onset and display a progressive increase in discharge Additional projections of the VLPO include the midbrain
rate from light to deep NREM sleep. In response to 12 to DR and the LC.46 The MnPN also projects to both the
16 hours of sleep deprivation, VLPO neurons exhibit DR and LC.47 Both MnPN and VLPO also project to the
increased activation during sleep, but rates during waking perifornical lateral hypothalamic area, the location of cell
remain the same as in control rats. Many MnPN neurons bodies of the orexin arousal system.48 Thus, discharge of
show a gradual increase in firing rates before sleep onset, VLPO and MnPN GABAergic neurons during sleep is
elevated discharge rates during NREM sleep, and a small expected to release GABA at these sites. Indeed, GABA
but significant additional increase in discharge rate during release is increased during NREM sleep and further
REM sleep. increased in REM sleep in the PH, DR, and LC (reviewed
Other sites in the POA also exhibit sleep-related c-Fos in reference 38).49 Sleep-active neurons in VLPO and
immunoreactivity. Electrophysiologic studies describe MnPN exhibit discharge-rate-change profiles across the
sleep-active neurons throughout the POA, and lesions wake-NREM-REM cycle that are reciprocal to those of
that do include the VLPO or MnPN are known to sup- wake-promoting HA, 5-HT, NE, and orexin neurons (Fig
press sleep. This suggests that much of the diffuse 7-4). These findings support a hypothesis that POA sleep-
sleep-related c-Fos immunoreactivity in the POA also active neurons, through release of GABA, inhibit multiple
labels neurons that are functionally important for sleep arousal systems.
regulation. The PH and LH areas also facilitate both motor and
autonomic activation, and these processes are under GAB-
Aergic control.50 Sleep-related GABAergic inhibition of
THE ORCHESTRATION OF SLEEP BY PH and LH provides a basis for sleep-related deactivation
THE POA HYPNOGENIC SYSTEM of autonomic and motor functions. POA lesions suppress
How do POA sleep-active neurons initiate and sustain REM as well as NREM sleep. REM sleep loss after POA
sleep? VLPO neurons that exhibit c-Fos immunoreactivity damage may be a secondary consequence of NREM sleep
following sleep express glutamic acid decarboxylase (GAD), disturbance. Alternatively, POA mechanisms may facilitate
the synthetic enzyme that produces the inhibitory neu- REM sleep as well as NREM sleep.
82 PARTI / Section 2 Sleep Mechanisms and Phylogeny

GABAergic neurons in the VLPO region are inhibited

VLPO by ACH, 5-HT, and NE, transmitters of the arousal
PFLH systems.51 MnPN neurons are inhibited by NE.52 Thus,
Discharge rate MnPN POA sleep-active neurons inhibit arousal systems, and
TMN arousal systems inhibit POA sleep-active neurons. These
mutually inhibitory processes are hypothesized to underlie
a sleep-wake switch or flip-flop model (see Fig 7-4).53
Activation of arousal systems would inhibit sleep-active
neurons, thereby removing inhibition from arousal systems,
facilitating stable episodes of waking. On the other hand,
activation of sleep-promoting neurons would inhibit
arousal-related neurons, removing inhibition from sleep-
A Wake Non-REM REM
promoting neurons and reinforcing consolidated sleep
episodes. This model provides a mechanism for the stabi-
lization of both sleep and waking states.
Abnormalities in one or more of the components of this flip-
flop system could result in less stable sleep and wake states, a
possible explanation for the fragmentation of sleep in sleep dis-
5-HT ac
orders in which insomnia is an element, and the fragmentation
ACH BF of waking in narcolepsy.
ACH, () Changes in cortical EEG patterns are usually considered
HIST, () VLPO to be the defining feature of NREM sleep in mammals.
5-HT, MnPN
Here we briefly review current understanding of thalamo-
cortical mechanisms underlying NREM sleep EEG pat-
C Arousal Sleep terns, and of how modulation of thalamocortical circuits
by arousal and sleep regulatory neuronal systems has an
Figure 7-4 Interactions of the preoptic area (POA) sleep-pro- impact on key features of the sleep EEG.
moting neuronal system with arousal systems that can account
for the orchestration of the sleep process. A, The neuronal Thalamocortical circuits exhibit two fundamentally dif-
discharge rates across the wake-NREM-REM cycle of sleep-active ferent modes of operation across the sleepwaking cycle:
neurons from the ventrolateral preoptic area (VLPO) and the a state of tonic activation, or desynchrony, during waking
median preoptic nucleus (MnPN), and from arousal-related and REM sleep, and a state of rhythmic, synchronized
(wake-active) neurons in the perifornical lateral hypothalamus
(PFLH) and tuberomammillary nucleus (TMN). These neuronal
activity that is characteristic of NREM sleep.54 The two
groups have generally reciprocal discharge patterns, although functional modes of thalamocortical activity are evident at
MnPN and VLPO neurons have peak activity at different times the level of single neurons. During waking and REM sleep,
during NREM episodes. The wake-active, NREM-diminished, thalamocortical neurons exhibit tonic firing of single action
REM-off discharge pattern shown for TMN and a subgroup of potentials (Fig. 7-5A) that are modulated by the levels of
PFLH neurons is also characteristic of putative serotoninergic
neurons of the dorsal raphe nucleus (DR) and putative norad- excitatory input from thalamic afferents, including specific
renergic neurons of the locus coeruleus (see also C). B, Sagittal sensory afferents.54,55 During NREM sleep, relay neurons
section of the diencephalon and upper brainstem of the rat discharge in high-frequency bursts of action potentials,
showing anatomic interconnections of MnPN and VLPO neurons followed by long pauses (see Fig. 7-5B).
with arousal-related neuronal groups. The MnPN and VLPO dis-
tribute projections to sites of arousal-related activity including
These two modes of action potential generation reflect
the (1) basal forebrain, (2) PFLH, which includes orexin-contain- the expression of intrinsic properties of thalamocortical
ing neurons, (3) histamine (HIST)-containing neurons of the neurons, and a specialized, voltage-sensitive Ca2+ current
tuberomammillary nucleus, (4) pontomesencephalic acetylcho- plays a critical role.56 This Ca2+ current, known as the low-
line (ACH)-containing neurons, (5) pontomesencephalic sero- threshold or transient Ca2+ current (It), is inactivated (non-
tonin (5-HT)-containing neurons, and (6) noradrenergic
(NE)-containing neurons of the pons, particularly the locus coe- functional) when the membrane potential of thalamic relay
ruleus (LC). 5-HT, NE, and ACH arousal-related neurons provide neurons is relatively depolarized (less negative than
inhibitory feedback to sleep-active neurons. The arousal-related 65mV). Thus, when depolarizing input is delivered to a
neuronal groups also have widespread additional ascending relay neuron that is resting at this level of membrane
and descending projections that control state-related functions
throughout the brain. C, The sleepwake switch or flip-flop
polarization, the cell responds with tonic single-spike
model characterized by mutually inhibitory sleep-promoting firing (see Fig. 7-5C). When relay neurons are hyperpolar-
and arousal-promoting neuronal groups, depicted as a seesaw, ized (membrane potential more negative than 65mV), It
which can promote stable sleep or waking states. Activity of becomes activated, and depolarizing input evokes a slow
either sleep-promoting or wake-promoting neurons inhibits the Ca2+-mediated depolarization (100 to 200 msec in dura-
neurons generating the opposing state. This network provides
a mechanism for the global control of brain activity in the tion) that is crowned by a burst of three to eight fast Na+
sleepwake cycle. (Modified from McGinty D, Szymusiak R. action potentials (see Fig. 7-5D). There is a pause in the
Hypothalamic regulation of sleep and arousal. Front Biosci generation of fast action potentials after the burst, because
CHAPTER 7 Neural Control of Sleep in Mammals 83


Cerebral cortex



20 mV
0.5 nA
C D 50 msec

Figure 7-5 Thalamic neurons exhibit distinct patterns of action

potential generation during waking/REM sleep and during RE
NREM sleep. A and B, Typical extracellularly recorded discharge
patterns of a neuron in the cat lateral geniculate nucleus during
waking and NREM sleep. Note the change from tonic, single-
spike firing during waking (A), to high-frequency-burst firing
during NREM sleep (B). Tonic versus burst firing reflects intrin- TC
sic, voltage-dependent properties of thalamic neurons, and it
can be recorded in neurons from isolated slices of thalamus. C
and D, In vitro intracellular recordings of a relay neuron from
guinea pig thalamus. In C, note direct current (DC) injections Thalamus PRE
(1 and 2) and recordings of intracellular voltage (3 and 4).
Spontaneous resting potential for the cell is indicated by the
dashed line. A depolarizing current step (1) delivered at resting Figure 7-6 Schematic representation of thalamic and cortical
potential (> 65 mV) evokes a tonic depolarizing response in cell types involved in the generation of sleep EEG rhythms, and
the neuron (3) that is subthreshold for action potential genera- of the synaptic connectivity among the cell types. Four cell
tion. When membrane potential is rendered more positive by types are shown: thalamocortical relay (TC) cells, thalamic
DC injection, the same depolarizing step (2) evokes tonic gen- reticular (RE) neurons, cortical pyramidal (PY) cells, and cortical
eration of fast action potentials (4) that persist for the duration interneurons (IN). TC cells receive excitatory inputs from
of the depolarizing pulse. In D, the neuron has been hyperpo- prethalamic afferent fibers (PRE) arising from specific sensory
larized below resting potential (< 65 mV) by negative DC systems and from cholinergic and monoaminergic arousal
injection, and the low threshold Ca2+ current (It) is activated. systems located in the brainstem and posterior hypothalamus.
When a depolarizing pulse is applied on the background of Activity in sensory systems is relayed to the appropriate cortical
hyperpolarization, a slow Ca2+ spike is evoked (arrow), and it area by ascending thalamocortical axons (up arrow). TC neurons
is crowned by a high-frequency burst of fast Na+ action poten- also send an axon collateral that makes synaptic contact with
tials. It is inactivated in response to the Ca2+-mediated depolar- RE neurons. RE neurons are GABAergic, and they send an inhibi-
ization, and membrane potential sags toward the hyperpolarized tory projection back to TC neurons. Corticothalamic feedback
level despite the continuance of the depolarizing current step. is mediated by layer VI, PY neurons that project back to the
EOG, electrooculogram; LGN, lateral geniculate nucleus. (A and same relay neurons from which they derive thalamic input, and
B modified from McCarley RW, Benoit O, Barrionuevo G. Lateral they send an axon collateral to RE neurons. Corticothalamic
geniculate nucleus unitary discharge during sleep and waking: projections are excitatory on both RE and TC cells, but cortical
state- and rate-specific effects. J Neurophysiol 1983;50:798- stimulation can evoke net inhibitory effects on TC neurons
818; C and D modified from Jouvet M. Neurophysiology of the because of activation of GABAergic RE neurons. (From Destexhe
states of sleep. Physiol Rev 1967;47:117-177.) A, Sejnowski TJ. Interactions between membrane conductances
underlying thalamocortical slow-wave oscillations. Physiol Rev

It is self-inactivated by the Ca2+-mediated depolarization,

and membrane potential falls below the threshold for tion of intrinsic neuronal properties (e.g., It) and the syn-
action potential generation and back to the resting, hyper- aptic organization of cortical and thalamic circuits.57 A
polarized state (see Fig. 7-5D). Thus, the properties of It schematic of the core circuitry responsible for the genera-
equip thalamocortical neurons with the ability to generate tion of sleep rhythms in the EEG is shown in Figure 7-6.
action potentials in two different modes: (1) tonic firing Thalamocortical relay neurons receive excitatory input
when stimulated from a relatively depolarized resting state, from sensory neurons and from several of the brainstem
and (2) burst-pause firing from a hyperpolarized resting arousal systems that function to promote depolarization
state.54,55 and tonic firing in relay cells during waking. During
The major NREM sleep EEG rhythms, spindles, delta waking, this excitation is faithfully conveyed by ascending
waves, and slow oscillations all arise through a combina- thalamocortical axons to the functionally relevant area of
84 PARTI / Section 2 Sleep Mechanisms and Phylogeny

cortex for processing and integration. Thalamic relay

neurons also send a collateral projection to the adjacent
portion of the thalamic reticular nucleus (RE), which is a
thin band of neurons that surrounds most thalamic relay
nuclei. RE neurons are GABAergic and they send an inhib- r
itory projection back to the relay neurons. These recipro-
cal connections between relay and RE neurons are thought Wake
to be important for aspects of waking thalamic function.
The final critical piece of the basic circuitry for consid- t
eration is the feedback projection from layer VI pyramidal
cells in cortex to both thalamic relay neurons and RE
neurons. Corticothalamic projections are topographically
organized, so that each cortical column has connectivity
with the same relay neurons from which they derive tha- Drowsy
lamic inputs, and with the corresponding sector of the RE.
In this anatomic/functional scheme, note the central loca-
tion of RE neurons, which receive copies of thalamocorti-
cal and corticothalamic activity and sends an inhibitory
projection back to the relay neurons. Although corticotha-
lamic projections have excitatory effects on their postsyn-
aptic targets in the thalamus, corticothalamic inputs to the
RE are so powerful that the net response evoked in relay Sleep
neurons by cortical stimulation is often inhibition.57
Sleep Spindles
In humans, EEG spindles are waxing and waning, nearly

0.1 mV
sinusoidal waves with a frequency profile of 10 to 15Hz.
Spindles are generated in the thalamus, as evidenced by
the fact that thalamectomy eliminates spindles in the sleep +
EEG.4 At the level of the thalamus, spindles are generated 2 msec
by an interplay between neurons in the RE and the relay
nuclei.54,55,57 RE neurons also possess It calcium channels Figure 7-7 Blockade of synaptic transmission in the thalamus
and exhibit high-frequency-burst firing from a hyperpolar- during drowsiness and sleep in the cat. Field potentials are
ized background. A high-frequency burst in RE neurons evoked in the ventrolateral thalamus by stimulation of the
will produce strong inhibitory postsynaptic potentials cerebellothalamic pathway. The evoked response consists of a
presynaptic volley (t) and the postsynaptic response (r). Note
(IPSPs) in relay neurons that are followed by rebound slow the variability of the postsynaptic response during drowsiness
Ca2+ spikes and a high-frequency burst. This burst of firing compared with during waking, and the complete absence of a
in the relay neuron is conveyed back to the RE, evoking postsynaptic response during sleep. (Modified from Steriade M,
an excitatory postsynaptic potential (EPSP) that triggers a Llins RR. The functional states of the thalamus and the associ-
calcium spike and a burst in RE neurons. In thalamic slices ated neuronal interplay. Physiol Rev 1988;68:649-742.)
that preserve connectivity between the RE and the adja-
cent relay nuclei, this disynaptic circuit can generate spon-
taneous spindle-like oscillations that can propagate across
the slice.55 In the intact brain, the spindle oscillation in the Delta Waves
thalamus is conveyed to the cortex by the pattern of burst The delta oscillation of NREM sleep appears to have both
firing in thalamic relay neurons.54,57,58 However, relay of cortical and thalamic components. This is evidenced by the
specific sensory information through the thalamus to the fact that cortical delta activity in the 1- to 4-Hz range
cortex is severely compromised during spindle oscillations persists after complete thalamectomy4 and by the demon-
(Fig. 7-7), because of the combination of (1) disfacilitation stration that isolated thalamic relay neurons can generate
of relay neurons resulting from loss of excitatory input a spontaneous clocklike delta oscillation resulting from the
from arousal systems and (2) the rhythmic IPSPs evoked interplay of It and a hyperpolarization-activated cation
by RE input. This sensory deafferentation of the cortex is current, known as the h-current.55 In the intact, sleeping
believed to play an important role in maintaining NREM brain, both sources of delta oscillation contribute to the
sleep continuity. frequency content of the cortical EEG. As discussed for
Although the spindle oscillation originates in the thala- spindles, corticothalamic and cortico-cortical connections
mus, cortical feedback to the thalamus and cortico-cortical function to synchronize delta oscillations over widespread
connections are important in synchronizing the occur- cortical areas.
rence of spindles over widespread thalamic and cortical
areas.57,58 Phasic activation of layer VI pyramidal neurons Slow Oscillations
excites neurons in the RE and synchronizes IPSPrebound Slow oscillations (less than 1 Hz) are a key aspect of the
burst sequences in thalamic relay neurons with cortical sleep EEG because they function to coordinate the occur-
activity. rence of other synchronous EEG events (e.g., delta waves,
CHAPTER 7 Neural Control of Sleep in Mammals 85

Slow wave sleep Wake HUMAN

KC + spindle KC


0.5 mV
20 mV
10 sec
P4 C4
10 O2 P3 P4
LFP power ratio,
<4 Hz to >4 Hz

5 1 sec
A O1 O2

0 S
B 0 30 60 90 sec FFT


0 1 2 3 4 HZ
1 sec
C1 C2

Figure 7-8 Slow oscillations in local cortical field potentials

(LFP) and in the membrane potential of a cortical neuron during
NREM sleep. A, Simultaneous intracellular, LFP, and EMG record- 0 5 10 15 20
ing during sleep and wakefulness. the animal is in NREM sleep
at the beginning of the recording with a transition to waking B HZ
after about 70 seconds (arrows indicate EMG activation). Action
potentials are truncated in the intracellular recording. B, Higher Figure 7-9 The cortical slow oscillation groups other sleep-
levels of delta power in the LFP are present during NREM sleep related EEG events in humans. A, Recordings of human EEG
than during waking. Plotted are 10-second bins of the ratio of during NREM sleep from three bilateral derivations. Note the
spectral power (<4 Hz/>4 Hz) recorded in the LFP. C, Intracel- synchronous occurrence of K-complexes (KC) and of spindles
lular activity and LFP recording from (A) shown at expanded (o) at multiple recording sites, and the regular recurrence of
timescale. Note clear fluctuations of the membrane potential KC/delta waves with a periodicity of less than 1Hz. B, Spectral
between depolarized (up) and hyperpolarized (down) states decomposition from the C3 lead showing peaks in spectral
during NREM sleep in association with the slow oscillation power in the spindle frequency range (12 to 15Hz), in the delta
(<1Hz) in the LFP. During wakefulness, cell is tonically depolar- power range (1 to 4 Hz), and in the slow oscillation range
ized and no sustained episodes of hyperpolarization are (<1Hz). (Modified from Sanchez-Vives MV, McCormick DA. Cel-
present. (From Mukovski M, Chauvette S, Timofeev I, Volgushev lular and network mechanisms of rhythmic recurrent activity in
M. Detection of active and silent states in neocortical neurons neocortex. Nat Neurosci 2000;3:1027-1034.)
from the field potential signal during slow-wave sleep. Cereb
Cortex 2007;17:400-414.)

resulting from depression of excitatory synapses.61,62 Dis-

charge of layer IV corticothalamic projection neurons
spindles, and K-complexes). Slow oscillations are entirely during up phases can synchronize IPSPs in thalamic relay
of cortical origin. They are absent from the thalamus in neurons through activation of RE neurons, causing the
decorticate animals and are present in the cortex after expression of EEG spindles and delta activity of thalamic
thalamectomy as well as in isolated cortical slices.57 Under- origin on the background of the slow oscillation.58 Slow
neath the slow oscillations, fluctuations occur between two oscillations organize the synchronization and propagation
states of activity in nearly all cortical neurons (Figs. 7-8 of cortical delta activity through cortico-cortical connec-
and 7-9).59,60 The up state is characterized by depolariza- tions. Frequency spectra of the human NREM sleep EEG
tion and generation of trains of action potentials. The up reflect this dynamic, with prominent spectral peaks in the
state occurs simultaneously in all cell types, including delta (1 to 4 Hz) and slow oscillation (less than 1 Hz)
interneurons, and both fast IPSPs and EPSPs are charac- frequency ranges (see Fig. 7-9).63
teristic of cortical neuronal activity during this state. Up By orchestrating the temporal and spatial coherence of
states are followed by a prolonged period of hyperpolariza- rhythmic oscillations in thalamocortical circuits, slow
tion and quiescence, referred to as a down state (see oscillations are thought to be important in promoting the
Fig. 7-8). functional sensory deafferentation of the cortex during
Generation of up states occurs through recurrent excita- sleep, which in turn enhances sleep continuity and sleep
tion in local cortical circuits and depends on excitatory depth. The transitions between up and down states in
transmission through -amino-3-hydroxy-5-methyl-4- cortical neurons have been hypothesized to underlie
isoxazole propionic acid (AMPA) and NMDA receptors.61 changes in synaptic plasticity, or synaptic strength, during
Transitions from up to down states involve a combination sleep, and to contribute to sleep-dependent changes in
of activation of outward K+ currents and disfacilitation learning and memory.64
86 PARTI / Section 2 Sleep Mechanisms and Phylogeny

Spindle and delta oscillations are blocked by stimulation sleepwake cycles, internal desynchronization, and forced
of the rostral brainstem, which activates cholinergic, desynchronization, the interactions of the temperature
monoaminergic, orexinergic, and glutamatergic inputs to rhythm and sleep propensity can be studied and partially
the thalamus. Activity of cholinergic and monoaminergic isolated from effects of prior waking (see Chapter 35).
neurons facilitates thalamic depolarization through inhibi- Although sleep may occur at any circadian phase, sleep
tion of potassium channels.55 Thus, inhibition of these propensity is increased on the late descending phase of the
arousal systems facilitates hyperpolarization of thalamic circadian temperature rhythm and is highest when tem-
neurons, permitting the activation of voltage-dependent perature is low. Awakenings tend to occur as temperature
membrane currents underlying spindles and slow waves. increases, even if sleep time is short. This can be inter-
The POA sleep-promoting system can control EEG patterns preted as evidence that the thermoregulatory mechanism
through inhibitory modulation of these arousal systems, as that lowers body temperature also promotes sleep.
described earlier. The coupling of sleep propensity and body cooling is
controlled by the POA. The POA is recognized as a ther-
moregulatory control site on the basis of the effects of local
INTEGRATION OF CIRCADIAN warming and cooling, lesions, local chemical stimulation,
RHYTHMS AND SLEEP and neuronal unit recording studies (reviewed in reference
The suprachiasmatic nucleus (SCN) of the POA generates 71). In vivo and in vitro studies have confirmed that the
the signals that bring about the circadian patterns of sleep POA contains populations of warm-sensitive and cold-
waking.65 Direct projections from the SCN to areas of the sensitive neurons (WSNs and CSNs), which are identified
POA implicated in sleep regulation are sparse, but multi- by changes in neuronal discharge in response to locally
synaptic pathways by which SCN signals can control of applied mild thermal stimuli. Local POA warming pro-
sleep-active neurons have been described. Lesions of a motes NREM sleep and EEG slow-wave activity in cats,
primary SCN projection target, the subparaventricular rabbits, and rats (reviewed in reference 38). NREM sleep
zone (SPVZ), also eliminate circadian rhythms of sleep is increased for several hours during sustained POA
waking.66 The SPVZ projects directly to the MnPN and warming.72 Local POA warming also increases EEG delta
indirectly to the VLPO, MnPN, and other POA regions activity in sustained NREM. Because delta EEG activity
through the dorsomedial hypothalamic nucleus (DMH).67 in sustained sleep is considered to be a measure of sleep
Lesions of the DMH disrupt the circadian distribution of drive, this finding supports a hypothesis that sleep drive is
sleepwaking states in rats. In diurnal animals, activity of modulated by POA thermosensitive neurons. Augmenta-
SCN neurons could inhibit sleep-promoting neurons tion of sleep by ambient warming is prevented by coinci-
during the light phase, or they may facilitate sleep-pro- dent local POA cooling.72 Because POA cooling prevents
moting neurons in the dark phase, with the reciprocal activations of WSNs, this finding suggests that POA WSN
pattern occurring in nocturnal animals, or they may do activation mediates the sleep augmentation induced by
both. ambient warming.
Most POA WSNs exhibit increased discharge during
NREM sleep compared with waking; most CSNs are
THE POA, THERMOREGULATION, wake-active (reviewed in reference 38). Increases in WSN
AND CONTROL OF SLEEP discharge and decreases in CSN discharge anticipate EEG
Several types of evidence support the hypothesis that sleep changes at sleep onset by several seconds. Thus, activity of
onset is coupled to body cooling. Depending on ambient WSNs and CSNs is likely to modulate spontaneous sleep
conditions, sleep onset evokes heat losseffector processes wake as well as sleep induced by local POA warming. As
such as cutaneous vasodilation and sweating.68 In humans, summarized earlier, POA sleep-active neurons send affer-
sleep onset occurs soon after vasodilation of the hands and ents to putative arousal systems. Local POA warming sup-
feet, which increases heat loss. Sleep in humans and animals presses the discharge of arousal-related neurons in the PH,
is modulated by ambient temperature. Mild to moderate DR (5-HT neurons), LH orexin neuronal field, and basal
ambient temperature elevation increases coincident sleep forebrain cholinergic field (Fig. 7-10) (reviewed in refer-
as well as subsequent sleep (reviewed in reference 38). An ence 38). These studies demonstrate that functional inhibitory
increase in heat production by selective activation of brown regulation of arousal-regulatory neurons originates in WSNs
adipose tissue using a beta3 adrenoreceptor agonist also located in the POA, and they provide a mechanistic explanation
increases NREM sleep.69 Thus, it is reasonable to hypoth- for the coupling of sleep propensity to heat loss.
esize that one function of sleep is body cooling, particu-
larly after increased body warming during wake.
A circadian temperature rhythm is well documented. In HIERARCHICAL CONTROL MODEL
humans, sleep normally occurs on the descending phase of As reviewed in the beginning paragraphs of this chapter,
the circadian temperature cycle, but sleep onset evokes a there is evidence for the existence of sleep-promoting
further decrease in temperature, even during continuous mechanisms at all levels of the neuraxis, including both
bed rest (reviewed in reference 38). Self-selected human forebrain and brainstem. Sleeplike behavior may be gener-
bedtimes occur at the time of the maximal rate of decline ated by an isolated lower brainstem, and sleep is facilitated
of core body temperature.70 The association of sleep onset by midbrain, thalamus, and neocortex. Lesions encompass-
and the circadian temperature rhythm could represent two ing the ventral pontomesencephalic reticular formation
independent outputs of the circadian oscillator. However, and adjacent ventral structures, also produced severe sus-
under certain experimental conditions, including short tained sleep reductions.73 Both NREM and REM sleep
CHAPTER 7 Neural Control of Sleep in Mammals 87

were reduced by about 50% a month after such lesions 1.4

were created. However, a central role for the POA in the
orchestration of sleep is supported by a variety of findings. 1.2
To rationalize these diverse findings, it is useful to con-

Discharge rate (spikes/sec)

sider a hierarchical control concept, such as that applied 1.0
previously to thermoregulation.74 In this conceptualiza-
tion, a fundamental behavior such as restactivity or sleep 0.8
waking is organized at all levels of the neuraxis, just as
restactivity cycles are found in all orders of animals. The 0.6
POA may act as a master controller, integrating sleep-
promoting circuitry with sleep homeostatic processes, 0.4 *
other behavioral and hormonal regulatory systems, and
circadian signals. The limbic system and neocortex are 0.2
likely sources of additional controls, related to more *
complex behavioral contingencies including learning pro-
cesses, instinctive behaviors, and sensory stimuli. These
controls may be conveyed to the hypothalamus through
projections from neocortex and limbic system. 1.66 Hz 0.59 Hz 1.71 Hz


100 V
Conceptions of sleep control based on neuronal circuitry, 1 msec
as outlined earlier, are deficient in that they do not provide EEG
explanations for quantitative features of sleep or sleep
homeostasis. The control of neuronal activity by neurochemical temp. C
mechanisms, including the release of GABA, occurs in a time
frame of seconds. Sleep regulation and homeostasis has a time 36 50 sec
frame of days, not seconds. In addition, a complete descrip-
tion should account for biological variations in sleep such Figure 7-10 Effects of activation of preoptic area (POA) warm-
as the high daily sleep quotas in low-body-weight mam- sensitive neurons (WSNs) by local warming on discharge of a
malian species and low quotas in very large species, higher putative serotoninergic neuron of the dorsal raphe nucleus
sleep quotas in infants, sleep facilitation after body heating, (DRN) in the rat. A, Pattern of activity of DRN neurons across
the sleepwake cycle. Putative DRN serotoninergic neurons
and increased sleep propensity during acute infection. The exhibit reduced discharge during NREM compared with waking,
investigation of biochemical mechanisms with sustained and very low discharge in REM sleep. The identification of this
actions is needed to address these issues. Here we provide pattern of discharge as representing serotoninergic neurons is
a brief overview of this approach, focusing on neurochemi- supported by several types of indirect evidence. B, Effects of
POA warming on the discharge of a putative serotoninergic
cal agents with sleep-promoting properties. See Obal and REM-off neuron during waking. A POA warming pulse was
Kreuger75 for a detailed and complete review of biochemi- applied for about 100 seconds (bottom trace). During POA
cal mechanisms of sleepwake regulation. warming, the discharge was reduced by 64%, to a typical NREM
sleep rate. A waking state was maintained during warning, as
Adenosine shown by electroencephalographic and electromyographic
recordings. Increased discharge of POA WSNs during spontane-
Adenosine is recognized as an inhibitory neuromodulator ous NREM sleep is thought to contribute to the concurrent
in the CNS, whose role in sleep is suggested by the potent reduction of DRN discharge. The central role of temperature-
arousal-producing effects of caffeine, an antagonist of sensitive neurons in sleep control provides a mechanistic basis
adenosine A1 receptors. Adenosine and its analogues were for the coupling of sleep and the circadian temperature rhythm
(see text). (From Guzman-Marin R, Alam MN, Szymusiak R, et
found to promote sleep after systemic injection, intracere- al. Discharge modulation of rat dorsal raphe neurons during
broventricular (ICV) administration, and intra-POA sleep and waking: effects of preoptic/basal forebrain warming.
microinjection, and particularly after administration by Brain Res 2000;875:23-34.)
microdialysis in the basal forebrain (reviewed in reference
76). In the basal forebrain and, to a lesser extent, in
neocortex, adenosine recovered through microdialysis
increased during sustained waking in cats. No increase was by basal forebrain cholinergic neurons via A1 receptors.76
found in thalamus, POA, or brainstem sites. Application in The proposed basal forebrain cholinergic neuronal site of
the basal forebrain of an antisense oligonucleotide to the action is problematic because destruction of these neurons
adenosine A1 receptor mRNA, which blocks synthesis of has little effect on sleep.79 However, adenosine could act
receptor protein, slightly reduced spontaneous sleep, but at multiple sites. Adenosine A2A receptors are also present
it strongly reduced rebound after sleep deprivation.77 Ade- in restricted brain regions and seem to mediate some
nosine A1 agonists delivered via microdialysis adjacent to sleep-promoting effects of adenosine.
basal forebrain neurons inhibited wake-active neurons Brain adenosine levels rise when ATP production is
during both waking and sleep.78 A current hypothesis is reduced; under these conditions, inhibition of neuronal
that the effects of adenosine on sleepwaking are mediated activity is neuroprotective. It has been proposed that
88 PARTI / Section 2 Sleep Mechanisms and Phylogeny

increased adenosine is a signal of reduced brain energy

reserves that develop during waking, and that sleep is Growth HormoneReleasing Hormone
induced as an energy-restorative state.80 With respect to Growth hormonereleasing hormone (GHRH) is known
the brain energy restorative concept, some, but not all, primarily for its role in stimulating the release of growth
studies show reduced cerebral glycogen, an energy-supply hormone (GH). A surge in GH release occurs early in the
substrate, after sleep deprivation.81 There is no functional major circadian sleep period in humans and in rats, specifi-
evidence that brain energy supply is compromised after cally during the earliest stage 3/4 NREM sleep episodes in
sleep deprivation. This is a critical gap in the theory. Of humans.85 GHRH is a peptide with a restricted localization
course, adenosine could be a sleep-promoting signal based in neurons in the hypothalamic arcuate nucleus and in the
on functions other than energy supply limitation. adjacent ventromedial and periventricular nuclei. Neurons
in the latter location are thought to be the source of pro-
Proinflammatory Cytokines jections to the POA and basal forebrain. GHRH promotes
Several proinflammatory cytokines have sleep-promoting NREM sleep after intraventricular, intravenous, intrana-
properties. We summarize work on a well-studied and sal, or intraperitoneal administration, or after direct micro-
prototypic molecule, interleukin (IL)-1. When adminis- injection into the POA (reviewed in reference 75). Blockade
tered intravenously, intraperitoneally, or into the lateral of GHRH by administration of a competitive antagonist
ventricles, IL-1 increases sleep, particularly NREM sleep, or by immunoneutralization reduces baseline sleep and
(reviewed in reference 82). Basic findings have been con- rebound sleep after short-term sleep deprivation. Mutant
firmed in several species. REM is usually inhibited by mice with GHRH signaling abnormalities have lower
IL-1. Much additional evidence supports a hypothesis amounts of NREM sleep. GHRH stimulates cultured
that IL-1 modulates spontaneous sleep. Administration of GABAergic hypothalamic neurons; these may constitute
agents that block IL-1 reduce sleep. In rats, IL-1 mRNA the GHRH target in the sleep-promoting circuit.86 It has
is increased in the brain during the light phase, when rat been suggested that GHRH may elicit sleep onset in con-
sleep is maximal. Sleep deprivation also increases IL-1 junction with release of GH as a coordinated process for
mRNA in brain. On the basis of these studies and others, augmenting protein synthesis and protecting proteins
it has been hypothesized that IL-1 plays a role in spontane- from degradation during the fasting associated with sleep.
ous sleep. Increased sleep associated with peripheral infec- In brain, protein synthesis increases during sleep, and inhi-
tion may also be mediated by responses to circulating IL-1, bition of protein synthesis augments sleep.87 GHRH is
either through vagal afferents or via induction of central proposed to be one element of a multiple-element sleep-
IL-1 or other hypnogenic signals. POA sleep-active promoting system.75
neurons are activated by local application of IL-1, and
wake-active neurons are inhibited.83 IL-1 fulfills many Sleep as Detoxification or Protection
criteria for a sleep-promoting signal, and it is likely to from Oxidative Stress
be particularly important in facilitating sleep during Oxidized glutathione (GSSR) was identified as one of four
infection. sleep-promoting substances in brain tissue extracted from
sleep-deprived rats.88 Infusion of GSSR or its reduced
Prostaglandin D2 form (GSH) into the lateral ventricle of rats during the
Administration of prostaglandin D2 (PGD2) by ICV infu- dark phase increases both NREM and REM sleep (reviewed
sion or by microinjection into the POA increases sleep, but in reference 89). The sleep-enhancing effects of GSSR and
the most potent site for administration is the subarachnoid GSH could be based on their antioxidant functions (see
space ventral to the POA and basal forebrain (reviewed in later) or on their functions as inhibitory regulators of glu-
reference 84). Sleep induced by PGD2 administration is tamatergic transmission. Glutamatergic stimulation, in
indistinguishable from normal sleep on the basis of EEG excess, has known neurotoxic functions mediated by
analysis. The synthetic enzyme lipocalin-PGD synthase NMDA receptors. Thus, GSSR and GSH could function
(L-PGDS) is enriched in the arachnoid membrane and to protect the brain against excess glutamatergic stimula-
choroid plexus, but the receptor (the D-type prostanoid tion and potential neurotoxicity, inducing sleep through
receptor) is localized more locally in the leptomeninges NMDA receptor blockade.
under the basal forebrain and the TMN. Knockout mice Reactive oxygen species (ROS) include superoxide anion
for L-PGDS had normal baseline sleep, but, unlike the (O2), hydrogen peroxide (H2O2), hydroxyl radical (OH),
controls, they had no rebound increase in NREM sleep nitric oxide (NO), and peroxynitrite (OONO). ROS are
after sleep deprivation. The PGD2 concentration was generated during oxidation reactions or reactions between
higher in the cerebrospinal fluid during NREM sleep than O2 and H2O2 or NO. ROS are normally reduced by con-
in waking and was higher in the light phase in the rat, when stitutive antioxidants such as GSSR, GSH, and different
sleep amounts are high. Sleep deprivation increased the forms of superoxide dismutase (SOD). GSH is lower in the
PGD2 concentration in the cerebrospinal fluid. On this hypothalamus of rats after 96 hours of sleep deprivation
basis, it was proposed that PGD2 plays a central role in using the platform-over-water method.90 Five to 11 days
sleep homeostasis. The hypnogenic action of PGD2 is of deprivation using the disk-over-water sleep deprivation
hypothesized to be mediated by adenosine release acting method reduces Cu/Zn SOD (cytosolic SOD) as well as
on an adenosine A2A receptor. Administration of A2A antag- glutathione peroxidase in the hippocampus and brain-
onists blocked the effects of PGD2. A functional basis for stem.91 These studies suggest that, by reducing antioxidant
the role of PGD2 in sleep homeostasis or its primary local- availability, sleep deprivation can increase the risk of oxida-
ization in the meninges has not been identified. tive damage, and that sleep is protective against actions of
CHAPTER 7 Neural Control of Sleep in Mammals 89

ROS. The possibility that sleep deprivation could cause systems would determine sleepwaking state. The circa-
neuronal damage was suggested by a finding that supraop- dian clock in the suprachiasmatic nucleus has direct and
tic nucleus neurons exhibited signs of subcellular damage multisynaptic connections, with POA sleep-promoting
after exposure to sleep deprivation.92 Supraoptic nucleus neurons providing a mechanistic basis for generation of the
neurons may be sensitive to sleep deprivation because of daily rhythm of sleepwaking. Both sleep-promoting and
their high rate of protein synthesis.92 arousal-promoting neuronal groups are modulated by a
What are the signaling molecules related to oxidative host of processes, including sensory, autonomic, endo-
stress that increase as a function of sustained wakefulness crine, metabolic, and behavioral influences, accounting for
and can promote sleep? One possible signal is NO produc- the sensitivity of sleep to a wide range of centrally acting
tion, which can be a response to glutamatergic stimulation, drugs and behavioral manipulations.
to cytokines and inflammation, and to oxidative stress.93 The long-term regulation of sleepsleep homeosta-
Activity of one NO synthetic enzyme, cytosolic nitric oxide sisis the subject of competing and still incomplete
synthase (NOS), is increased during the dark phase in rats, hypotheses. Long-term sleep homeostasis may reflect the
most strongly in the hypothalamus (reviewed in reference actions of several neurochemical processes that express
94). Intravertebroventricular or IV administration of a sleep factors. Some of these sleep factors, including ade-
NOS inhibitor strongly reduced sleep in rabbits and rats nosine, PGD2, IL-1, and GHRH, are known to act
and suppressed NREM sleep response to sleep depriva- directly on the POA or adjacent basal forebrain neuronal
tion. Administration of NO donors increased sleep. NO targets. Sleep factors have been linked to distinct func-
inhibits oxidative phosphorylation and may stimulate the tional models of sleep homeostasis, including brain energy
production of adenosine. NO production could, therefore, supply (adenosine), control of protein synthesis (GHRH),
be a mediator of several sleep factors. brain cell group use or temperature elevation (IL-1), or
ROS may play a role in the pathology associated with protection against oxidative or glutamatergic stress (NO,
patients with obstructive sleep apnea. These patients antioxidant enzymes). All of these factors may be involved
exhibit signs of increased oxidative stress, including in sleep homeostasis, but the relative importance of each
increased O2 production by neutrophils, monocytes, and factor for daily sleep is not established.
granulocytes derived from patients (see Chapter 20). They
also exhibit elevated levels of proteins, such as vascular
endothelial growth factor, that are normally induced by Clinical Pearls
ROS. These changes were reversed by treatment with con- Sleep is reduced by a wide variety of localized brain
tinuous positive airway pressure. On the basis of these and lesions. This may account for the association of
several additional findings, it has been proposed that the insomnia with a variety of neuropathologies. Degen-
increased cardiovascular disease in patients with obstruc- eration of the hypothalamic wake-promoting cell type
tive sleep apnea results from oxidative damage to vascular that expresses the neuropeptide orexin is known to
walls. Oxidative stress is hypothesized to play a central role underlie narcolepsy.
in both vascular disease and neurodegenerative disease.
It is important to note that adenosine, ROS, glutamate,
and NO have brief lives in the synaptic spaceno more Acknowledgments
than a few seconds. If these molecules regulate sleep, Supported by PHS grants MH 075076, MH 04708, MH
they must have sustained release, or they may regulate 61354, and MH 63323, and the Veterans Administration.
the gene expression to generate sustained downstream
effects. These mechanisms are the subject of current REFERENCES
investigations. 1. Villablanca J. Electroencephalogram in the permanently isolated
forebrain of the cat. Science 1962;138:44-46.
2. Zernicki B. Pretrigeminal cat. Brain Res 1968;9:1-14.
SUMMARY 3. Nakata K, Kawamura H. ECoG sleep-waking rhythms and
bodily activity in the cerveau isole rat. Physiol Behav 1986;36:
There has been rapid progress in the elucidation of the 1167-1172.
neural circuitry underlying the facilitation of sleep and the 4. Villablanca JR. Counterpointing the functional role of the forebrain
orchestration of NREM sleep as well as the facilitation of and of the brainstem in the control of sleep-waking system. J Sleep
arousal. We have detailed models that can explain many of Res 2004;13:179-208.
the phenomena of sleepwake. Wake and arousal are facili- 5. Montagna P, Gambetti P, Cortelli P, et al. Familial and sporadic fatal
insomnia. Lancet Neurol 2003;2:167-176.
tated by several chemically distinct neuronal groups, 6. Lovblad KO, Massetti C, Mathis J, Schroth G. MRI of paramedian
including groups synthesizing and releasing acetylcholine, thalamic stroke with sleep disturbance. Neuroradiology 1997;39:
serotonin, norepinephrine, dopamine, histamine, orexin/ 693-698.
hypocretin, and glutamate. These neuronal groups distrib- 7. Magnes J, Moruzzi G, Pomeiano O. Synchronization of the EEG
produced by low-frequency electrical stimulation of the region of the
ute axons and terminal throughout the brain, providing a solitary tract. Arch Ital Biol 1961;99:33-67.
basis for concurrent changes in physiology associated with 8. Berlucchi G, Maffei LMG, Stratta P. EEG and behavioral effects
arousal. At the center of the sleep-promoting circuitry is elicited by cooling of medulla and pons. Arch Ital Biol 1964;102:
the POA of the hypothalamus, confirming a hypothesis 372-392.
that is more than 70 years old. The POA sleep-promoting 9. Moruzzi G. The sleep-waking cycle. Ergeb Physiol 1972;64:1-165.
10. Jouvet M. The role of monoamines and acetylcholine-containing
circuitry has reciprocal inhibitory connections with several neurons in the regulation of the sleep-waking cycle. Ergeb Physiol
arousal-promoting systems. A balance between the activi- 1972;64:166-307.
ties of sleep-promoting and arousal-promoting neuronal 11. Ursin R. Serotonin and sleep. Sleep Med Rev 2002;6:57-69.
90 PARTI / Section 2 Sleep Mechanisms and Phylogeny

12. Popa D, Lena C, Fabre V, et al. Contribution of 5-HT2 receptor 37. Nauta WJH. Hypothalamic regulation of sleep in rats. An experi-
subtypes to sleep-wakefulness and respiratory control, and functional mental study. J Neurophysiol 1946;9:285-316.
adaptation in knock-out mice. J Neurosci 2005;25:11231-11238. 38. McGinty D, Szymusiak R. Brain structures and mechanisms involved
13. Boutrel B, Monaca C, Hen R, et al. Involvement of 5HT1A receptors in the generation of NREM sleep: focus on the preoptic hypothala-
in homeostatic and stress-induced adaptive regulation of paradoxical mus. Sleep Medicine Rev 2001;5:323-342.
sleep: studies in 5HT1A knock-out mice. J Neurosci 2002;22:4686- 39. Lu J, Greco MA, Shiromani P, Saper CB. Effect of lesions of the
4692. ventrolateral preoptic nucleus on NREM and REM sleep. J Neuro-
14. Aston-Jones G, Bloom FE. Activity of norepinephrine-containing science 2000;20:3830-3842.
locus coeruleus neurons in behaving rats anticipates fluctuations in 40. Sherin JE, Shiromani PJ, McCarley RW, et al. Activation of ven-
the sleep-waking cycle. J Neurosci 1981;1:876-886. trolateral preoptic neurons during sleep. Science 1996;271:216-
15. Lidbrink P. The effects of lesions of ascending noradrenaline path- 219.
ways on sleep and waking in the rat. Brain Res 1974;74:19-40. 41. Morgan J, Curran T. Stimulus-transcription coupling in the nervous
16. Berridge CW. Noradrenergic modulation of arousal. Brain Res Rev system: Involvement of the inducible proto-oncogenes fos and jun.
2008;58:1-17. Annu Rev Neurosci 1991;14:421-451.
17. Lin JS, Sakai K, Vanni-Mercier G, et al. A critical role of the poste- 42. Gong H, Szymusiak R, King J, et al. Sleep-related c-Fos expression
rior hypothalamus in the mechanisms of wakefulness determined by in the preoptic hypothalamus: effects of ambient warming. Am J
microinjection of muscimol in freely moving cats. Brain Res Physiol Regulatory Integrative Comp Physiol 2000;279:R2079-
1989;479:225-240. R2088.
18. Steininger TL, Alam MN, Gong H, et al. Sleep-waking discharge of 43. Suntsova N, Szymusiak R, Alam MN, et al. Sleep-waking discharge
neurons in the posterior lateral hypothalamus of the albino rat. Brain patterns of median preoptic nucleus neurons in rats. J Physiol
Res 1999;840:138-147. 2002;543:665-677.
19. Kiyono S, Seo ML, Shibagaki M, et al. Effects of alpha-fluorometh- 44. Gong H, McGinty D, Guzman-Marin R, et al. Activation of GAB-
ylhistidine on sleep-waking parameters in rats. Physiol Behav Aergic neurons in the preoptic area during sleep and in response to
1985;34:615-617. sleep deprivation. J Physiol 2004;556:935-946.
20. Thannickal TC, Moore RY, Nienhuis R, et al. Reduced number of 45. Sherin JE, Elmquist JK, Torrealba F, et al. Innervation of histamin-
hypocretin neurons in human narcolepsy. Neuron 2000;27:469-474. ergic tuberomammillary neurons by GABAergic and galaninergic
20a. Nakamura M, Kanbayashi T, Sugiura T, Inoue Y. Relationship neurons in the ventrolateral preoptic nucleus of the rat. J Neurosci
between clinical characteristics of narcolepsy and CSF orexin-A 1998;18:4705-4721.
levels. J Sleep Res 2010 Jul 19 (Epub ahead of print). 46. Steininger TL, Gong H, McGinty D, et al. Subregional organization
21. Peyron C, Tighe DK, van den Pol AN, et al. Neurons containing of preoptic area/anterior hypothalamic projections to arousal-
hypocretin (orexin) project to multiple neuronal systems. J Neurosci related monoaminergic cell groups. J Comp Neurol 2001;429:
1998;18:9996-10015. 638-653.
22. Mileykovskiy BY, Kiyashchenko LI, Siegel JM. Behavioral correlates 47. Uschakov A, Gong H, McGinty D, et al. Efferent projections from
of activity in identified hypocretin/orexin neurons. Neuron 2005; the median preoptic nucleus to sleep- and arousal regulatory nuclei
46:787-798. in the rat brain. Neurosci 2007;150:104-120.
23. Lee MG, Hassani OK, Jones BE. Discharge of identified orexin/ 48. Uschakov A, Gong H, McGinty D, et al. Sleep-active neurons in the
hypocretin neurons across the sleep-wake cycle. J Neurosci 2005; preoptic area project to the hypothalamic paraventricular nucleus and
25:6716-6720. the perifornical lateral hypothalamus. Eur J Neurosci 2006;23:
24. Methippara MM, Alam MN, Szymusiak R, et al. Effects of lateral 3284-3296.
preoptic area application of orexin-A on sleep-wakefulness. Neuro- 49. Nitz D, Siegel JM. GABA release in posterior hypothalamus across
report 2002;11:3423-3426. sleep-wake cycle. Am J Physiol 1996;271:R1707-R1712.
25. Mesulam MM, Mufson EJ, Wainer BH, et al. Central cholinergic 50. Waldrop TG, Bauer RM, Iwamoto GA. Microinjections of GABA
pathways in the rat: an overview based on an alternative nomencla- antagonists into the posterior hypothalamus elicits locomotor
ture. Neurosci 1983;10:1185-1201. movements and cardiorespiratory activation. Brain Res 1988;444:
26. Duque A, Balatoni B, Detari L, et al. EEG correlation of the dis- 84-94.
charge properties of identified neuron in the basal forebrain. 51. Gallopin T, Fort P, Eggermann E, et al. Identification of sleep-
J Neurophysiol 2000;84:1627-1635. promoting neurons in vitro. Nature 2000;404:992-995.
27. Marrosu F, Portas C, Mascia MS, et al. Microdialysis measurement 52. Bai D, Renaud LP. Median preoptic nucleus neurons: An in vitro
of cortical and hippocampal acetylcholine release during sleep-wake patch-clamp analysis of their intrinsic properties and noradrenergic
cycle in freely moving cats. Brain Res 1995;671:329-332. receptors in the rat. Neurosci 1998;83:905-916.
28. Karczmar AG, Longo A, de Carolis S. A pharmacological model of 53. Saper CB, Chou TC, Scammel TE. The sleep switch: hypothalamic
paradoxical sleep: the role of cholinergic and monoaminergic control of sleep and wakefulness. Trends Neurosci 2001;24:726-731.
neurons. Physiol Behav 1970;5:175-182. 54. Steriade M, Llins RR. The functional states of the thalamus and the
29. Lindvall O, Bjorklund A. Neuroanatomy of central dopamine path- associated neuronal interplay. Physiol Rev 1988;68:649-742.
ways: review of recent progress. In: Keal M, editor. Advances in 55. McCormick DA, Bal T. Sleep and arousal: thalamocortical mecha-
dopamine research. New York: Pergamon; 1982. pp. 297-311. nisms. Annu Rev Neurosci 1997;20:185-215.
30. Feenstra M, Botterblom M, Mastenbrook S. Dopamine and nor- 56. Jahnsen H, Llinas R. Electrophysiological properties of guinea-
adrenaline efflux in the prefrontal cortex in the light and dark period: pig thalamic neurons: an in vitro study. J Physiol 1984;349:
effects of novelty and handling and comparison to the nucleus 205-226.
accumbens. Neurosci 2000;100:741-748. 57. Destexhe A, Sejnowski TJ. Interactions between membrane conduc-
31. Wisor JP, Eriksson KS. Dopaminergic-adrenergic interactions in tances underlying thalamocortical slow-wave oscillations. Physiol
the wake promoting mechanism of modafinil. Neurosci 2005;132: Rev 2003;83:1401-1453.
1027-1034. 58. Steriade M. The corticothalamic system in sleep. Front Biosci
32. Rye DB, Biwise DL, Dihenia B, et al. Daytime sleepiness in Parkin- 2003;8:878-899.
sons disease. J Sleep Res 2000;9:63-69. 59. Steriade M, Nunez A, Amzica F. A novel (<1Hz) oscillation of neo-
33. Jones BE. Reticular formation: cytoarchitecture, transmitters, and cortical neurons in vivo: depolarizing and hyperpolarizing compo-
projections. In: Paxinos G, editor. The rat nervous system. Sydney: nents. J Neurosci 1993;13:3252-3265.
Academic Press; 1995. pp. 155-171. 60. McCormick DA. Cortical and subcortical generators of normal and
34. Manfridi A, Brambilla D, Mancia M. Stimulation of NMDA and abnormal rhythmicity. Int Rev Neurobiol 2002;49:99-114.
AMPA receptors in the rat nucleus basalis of Meynert affects sleep. 61. Sanchez-Vives MV, McCormick DA. Cellular and network mecha-
Am J Physiol Regul Integr Comp Physiol 1999;277:R1488-R1492. nisms of rhythmic recurrent activity in neocortex. Nat Neurosci
35. Campbell IG, Feinberg I. Noncompetitive NMDA channel blockade 2000;3:1027-1034.
during waking intensely stimulates NREM delta. J Pharmacol Exp 62. Steriade M, Curro-Dossi R, Nunez A. Network modulation of a slow
Ther 1996;276:737-742. intrinsic oscillation of cat thalamocortical neurons implicated in sleep
36. von Economo C. Sleep as a problem of localization. J Nerv Ment delta waves: cortically induced synchronization and brainstem cho-
Dis 1930;71:249-259. linergic suppression. J Neurosci 1991;11:3200-3217.
CHAPTER 7 Neural Control of Sleep in Mammals 91

63. Amzica F, Steriade M. The k-complex: its slow (<1Hz) rhythmicity 79. Bassant MH, Apartis E, Jazat-Poindessous FR, et al. Selective immu-
and relation to delta waves. Neurology 1997;49:952-999. nolesion of the basal forebrain cholinergic neurons: effects on
64. Tononi G, Cirelli C. Sleep function and synaptic homeostasis. Sleep hippocampal activity during sleep and wakefulness in the rat. Neu-
Med Rev 2006;10:49-62. rodegeneration 1995;4:61-70.
65. Ibuka N, Inouye ST, Kawamura H. Analysis of sleep-wakefulness 80. Benington JH, Heller HC. Restoration of brain energy metabolism
rhythms in male rats after suprachiasmatic nucleus lesions and ocular as the function of sleep. Prog Neurobiol 1995;45:347-360.
enucleation. Brain Res 1977;122:33-47. 81. Franken P, Gip P, Hagiwara G, et al. Changes in brain glycogen
66. Lu J, Zhang YH, Chou TC, et al. Contrasting effects of ibotenate after sleep deprivation vary with genotype. Am J Physiol 2003;
lesions of the paraventricular nucleus and subparaventricular zone on 285:P413-P419.
sleep-wake cycle and temperature regulation. J Neurosci 2001;21: 82. Kreuger JM, Rector DM, Churchill L. Sleep and cytokines. Sleep
4864-4874. Med Clin 2007;2:161-169.
67. Deurveilher S, Semba K. Indirect projections from the suprachias- 83. Alam MN, McGinty D, Imeri L, et al. Effects of interleukin-1 beta
matic nucleus to the median preoptic nucleus in rat. Brain Res on sleep- and wake-active preoptic anterior hypothalamic neurons in
2003;987:100-106. unrestrained rats. Sleep 2001;24:A59.
68. Heller HC, Glotzbach S, Grahn D, et al. Sleep-dependent changes 84. Huang Z-L, Urade Y, Hayaishi O. Prostaglandins and adenosine in
in the thermoregulatory system. In: Lydic R, Biebuyck J, editors. the regulation of sleep and wakefulness. Curr Opin Pharmacol
Clinical physiology of sleep. Washington, DC: American Physiologi- 2007;7:33-38.
cal Society; 1988. pp. 145-158. 85. Mitsugi N, Kimura F. Simultaneous determination of blood levels of
69. Dewasmes G, Loos N, Delanaud S, et al. Activation of brown adipose corticosterone and growth hormone in the male rat: relation to the
tissue thermogenesis increases slow wave sleep in rat. Neurosci Lett sleep-wakefulness cycle. Neuroendocrinology 1985;41:125-130.
2003;339:207-210. 86. De A, Churchill L, Obal F Jr, et al. GHRH and IL1 beta increase
70. Campbell SS, Broughton RJ. Rapid decline in body temperature cytoplasmic Ca2+ levels in cultured hypothalamic GABAergic
before sleep: fluffing the physiological pillow. Chronobiol Intern neurons. Brain Res 2002;949:209-212.
1994;11:126-131. 87. Methippara MM, Alam MN, Kumar S, et al. Administration of the
71. Boulant JA. Hypothalamic mechanisms in thermoregulation. Federa- protein synthesis inhibitor, anisomycin, has distinct sleep-promoting
tion Proc 1981;40:2843-2850. effects in lateral preoptic and perifornical hypothalamic sites in rats.
72. Sakaguchi S, Glotzbach SF, Heller HC. Influence of hypothalamic Neurosci 2008;151:1-11.
and ambient temperatures on sleep in kangaroo rats. Am J Physiol 88. Inoue S, Honda K, Komoda Y. Sleep as neuronal detoxification and
1979;237:R80-R88. restitution. Behavi Brain Res 1995;69:91-95.
73. Lai LL, Shalita T, Hajnik T, et al. Neurotoxic N-methyl-D-aspartate 89. Ikeda M, Ikeda-Sagara M, Okada T, et al. Brain oxidation is an initial
lesion of the ventral midbrain and mesopontine junction alters sleep- process in sleep induction. Neurosci 2005;130:1029-1040.
wake organization. Neurosci 1999;90:469-483. 90. DAlmeida V, Lobo LL, Hipolide DC, et al. Sleep deprivation
74. Satinoff E. Neural organization and evolution of thermal regulation induces brain region-specific decreases in glutathione levels. Neuro-
in mammals. Science 1978;201:16-22. report 1998;9:2853-2856.
75. Obal F Jr, Krueger JM. Biochemical regulation of non-rapid- 91. Ramanathan L, Gulyani S, Nienhuis R, et al. Sleep deprivation
eye-movement sleep. Front Biosci 2003;8:D520-D550. decreases superoxide dismutase activity in rat hippocampus and
76. McCarley RW. Neurobiology of REM and NREM sleep. Sleep Med brainstem. Neuroreport 2002;13:1387-1390.
2007;8:302-330. 92. Eiland MM, Ramanathan LGS, Gilliland M, et al. Increases in
77. Thakkar MM, Winston S, McCarley RM. A1 receptor and adenosin- amino-cupric-sliver staining of the supraoptic nucleus after sleep
ergic homeostatic of sleep-wakefulness: effects of antisense to the A1 deprivation. Brain Res 2002;945:1-8.
receptor in the basal forebrain. J Neurosci 2003;23:4278-4287. 93. Olivenza R, Mora MA, Lizasoain L, et al. Chronic stress induces
78. Alam MN, Szymusiak R, Gong H, et al. Adenosinergic modulation the expression of inducible nitirc oxide synthase in rat brain cortex.
of rat basal forebrain neurons during sleep and waking: neuronal J Neurochem 2000;74:785-791.
recording with microdialysis. J Physiol (London) 1999;521.3: 94. Gautier-Sauvigne S, Colas D, Parmantier P, et al. Nitric oxide and
679-690. sleep. Sleep Med Rev 2005;9:101-113.
REM Sleep*
Jerome M. Siegel
Abstract REM sleep. Of particular interest are manipulations that affect
the regulation of muscle tone in REM sleep. Lesions of several
Rapid eye movement (REM) sleep was first identified by its regions in the pons and medulla can cause REM sleep to occur
most obvious behavior: rapid eye movements during sleep. In without the normal loss of muscle tone. In REM sleep without
most adult mammals, the electroencephalogram (EEG) of the atonia, animals exhibit locomotor activity, appear to attack
neocortex exhibits low voltage during REM sleep. The hippo- imaginary objects, and execute other motor programs during
campus has regular high-voltage theta waves throughout REM a state that otherwise resembles REM sleep. This syndrome
sleep. may have some commonalties with the REM sleep behavior
The key brain structure for generating REM sleep is the disorder seen in humans. Stimulation of portions of the REM
brainstem, particularly the pons and adjacent portions of the sleep-controlling area of the pons can produce a loss of muscle
midbrain. These areas and the hypothalamus contain cells tone in antigravity and respiratory musculature, even without
that are maximally active in REM sleep, called REM-on cells, eliciting all aspects of REM sleep.
and cells that are minimally active in REM sleep, called REM-off Narcolepsy is characterized by abnormalities in the regula-
cells. Subgroups of REM-on cells use the transmitter gamma- tion of REM sleep. Most cases of human narcolepsy are caused
aminobutyric acid (GABA), acetylcholine, glutamate, or glycine. by a loss of hypocretin (orexin) neurons. Hypocretin neurons,
Subgroups of REM-off cells use the transmitter norepineph- which are located in the hypothalamus, contribute to the regu-
rine, epinephrine, serotonin, histamine, or GABA. lation of the activity of norepinephrine, serotonin, histamine,
Destruction of large regions in the midbrain and pons can acetylcholine, glutamate, and GABA cell groups. These trans-
prevent the occurrence of REM sleep. Damage to portions of mitters have potent effects on alertness and motor control and
the brainstem can cause abnormalities in certain aspects of are normally activated in relation to particular emotions.

Rapid eye movement, or REM, sleep was discovered by typically report dream mentation, perhaps because these
Aserinsky and Kleitman in 1953.1 In a beautifully written cortical regions have not yet developed.7 The physiologic
paper that has stood the test of time, they reported that signs of REM sleep in both the platypus, the animal
REM sleep was characterized by the periodic recurrence showing the most REM sleep,8 and a related monotreme,
of rapid eye movements, linked to a dramatic reduction in the short-nosed echidna,9 are largely restricted to the
amplitude from the higher-voltage activity of the prior brainstem, in contrast to their propagation to the forebrain
non-REM sleep period, as seen on the electroencephalo- in adult placental and marsupial mammals. These findings
gram (EEG). They found that the EEG of subjects in make it questionable whether all or any nonhuman
REM sleep closely resembled the EEG of alertwaking mammals that have REM sleep, all of which have cortical
subjects, and they reported that subjects awakened from regions whose structure differs from that of adult humans,
REM sleep reported vivid dreams. Dement identified a have dream mentation.
similar state of low-voltage EEG with eye movements in This chapter will review the following: (1) the defining
cats.2 Jouvet then repeated this observation, finding in characteristics of REM sleep, including its physiology and
addition a loss of muscle tone (i.e., atonia) in REM sleep neurochemistry, (2) the techniques used to investigate the
and using the term paradoxical sleep to refer to this state. mechanisms generating REM sleep and the conclusions of
The paradox was that the EEG resembled that of waking, such investigations, (3) the mechanisms responsible for the
but behaviorally the animal remained asleep and unrespon- suppression of muscle tone during REM sleep, and the
sive.3-5 Subsequent authors have described this state as pathologic effects of the disruption of these mechanisms,
activated sleep, or dream sleep. More recent work in (4) narcolepsy and its link to mechanisms involved in REM
humans has shown that some mental activity can be present sleep control, and especially to the peptide hypocretin, and
in non-REM sleep but has supported the original finding (5) the functions of REM sleep.
that linked our most vivid dreams to the REM sleep state.
However, we cannot assume that any person or animal
who has REM sleep also dreams. Lesions of parietal cortex CHARACTERISTICS OF REM SLEEP
and certain other regions prevent dreaming in humans, The principal electrical signs of REM sleep include a
even in individuals continuing to show normal REM sleep reduction in EEG amplitude, particularly in the power of
as judged by cortical EEG, suppression of muscle tone, and its lower-frequency components (Fig. 8-1). REM sleep is
rapid eye movements.6 Children younger than 6 years, who also characterized by a suppression of muscle tone (atonia),
have larger amounts of REM sleep than adults, do not visible in the electromyogram (EMG). Erections tend to
occur in males.10 Thermoregulation (e.g., sweating and
shivering) largely ceases in most animals, and body tem-
*Supported by the Medical Research Service of the Department of Veterans
Affairs, NIH grants NS14610, HL41370, HL060296, and MH64109. (Portions peratures drift toward environmental temperatures, as in
of this review have appeared elsewhere in another review by the author.) reptiles.11 Pupils constrict, reflecting a parasympathetic

CHAPTER 8 REM Sleep* 93


EMG 50 V




5 sec

Figure 8-1 Top, Polygraph tracings of states seen in the intact cat. Bottom, States seen in the forebrain 4 days after transection
at the pontomedullary junction. EEG, sensorimotor electroencephalogram; EMG, dorsal neck electromyogram; EOG, electrooculo-
gram; HIPP, hippocampus; LGN, lateral geniculate nucleus; OLF, olfactory bulb; PGO, ponto-geniculo-occipital.

dominance in the control of the iris.12 These changes that occur episodically in REM sleep have been called its phasic
are present throughout the REM sleep period have been features.
termed its tonic features. As we will see later, certain manipulations of the brain-
Also visible are electrical potentials that can be most stem can eliminate only the phasic events of REM sleep,
easily recorded in the lateral geniculate nucleus of the cat.13 whereas others can cause the phasic events to occur in
These potentials originate in the pons, appear after a few waking; yet other manipulations can affect tonic compo-
milliseconds in the lateral geniculate nucleus, and can be nents. These tonic and phasic features are also expressed
observed with further delay in the occipital cortex, leading to varying extents in different species, and not all of these
to the name ponto-geniculo-occipital (PGO) spikes. They features are present in all species that have been judged to
occur as large-amplitude, isolated potentials 30 or more have REM sleep.18
seconds before the onset of REM sleep as defined by EEG
and EMG criteria. After REM sleep begins, they arrive in REM GENERATION MECHANISMS
bursts of three to ten waves, usually correlated with rapid
eye movements. PGO-linked potentials can also be Technical Considerations
recorded in the motor nuclei of the extraocular muscles, The identification of sleep-generating mechanism can be
where they trigger the rapid eye movements of REM sleep. achieved by inactivation or destruction of particular brain
They are also present in thalamic nuclei other than the regions or neurons, by the activation of populations of
geniculate and in neocortical regions other than the occipi- neurons, or by recording the activity of neurons or measur-
tal cortex. ing the release of neurotransmitters. Each approach has its
In humans, rapid eye movements are loosely correlated advantages and limitations.
with contractions of the middle ear muscles of the sort that
accompany speech generation and that are part of the Inactivation of Neurons by Lesions, Inhibition,
protective response to loud noise.14 Other muscles also Antisense Administration, or Genetic Manipulation
contract during periods of rapid eye movement, briefly More has been learned about brain function and about
breaking through the muscle atonia of REM sleep. There sleep control from brain damage caused by stroke, injury,
are periods of marked irregularity in respiratory and heart or infection in patients, and by experimentally induced
rates during REM sleep, in contrast to non-REM sleep, brain lesions in animals, than by any other technique.
during which respiration and heart rate are highly regular. However, some basic principles need to be borne in mind
No single pacemaker for all of this irregular activity has when interpreting such data.
been identified. Rather, the signals producing twitches of Brain lesions can result from ischemia, pressure, trauma,
the peripheral or middle ear muscles may lead or follow and degenerative or metabolic changes. In animals, experi-
PGO spikes and rapid eye movements. Bursts of brainstem mental lesions are most commonly induced by aspiration,
neuronal activity may likewise lead or follow the activity transection of the neuraxis, electrolysis, local heating by
of any particular recorded muscle.15-17 These changes that radio frequency currents, or the injection of cytotoxins.
94 PARTI / Section 2 Sleep Mechanisms and Phylogeny

The latter include substances such as N-methyl-d-aspartic rebound, altered responses of sleep to environmental vari-
acid (NMDA) and kainate that cause cell death by excito- ables, and altered changes in sleep with development and
toxicity, and targeted cytotoxins such as saporin coupled aging. The same interpretive considerations long appreci-
to particular ligands, which will kill only cells containing ated in lesion studies apply to the interpretation of manip-
receptors for that ligand. Cytotoxic techniques have the ulations that inactivate genes or prevent gene expression,
considerable advantage of sparing axons passing through with the additional possibility of direct effects of genetic
the region of damage, so that deficits will be attributable manipulation on tissues outside the brain.
to the loss of local neurons rather than to interruption of
these axons. Activation of Neurons by Electrical or
If damage to a brain region causes the loss of a sleep Chemical Stimulation, Gene Activation, or
state, it cannot be concluded that this is where a center for Ion Channel Manipulation
the state resides. Lesion effects are usually maximal imme- Sites identified by lesion or anatomic studies can be stimu-
diately after the lesion is created. Swelling and circulatory lated to identify their roles in sleep control. Older studies
disruption make the functional loss larger than will be used electrical stimulation and were successful in identify-
apparent from standard postmortem histologic techniques. ing the medial medulla as a region mediating the suppres-
The loss of one brain region can also disrupt functions that sion of muscle tone,24 and the basal forebrain as a site
are organized elsewhere. For example, spinal shock, a phe- capable of triggering sleep.25 Electrical stimulation is
nomenon in which severing the spinal cords connection clearly an aphysiologic technique, involving the forced
to more rostral brain regions causes a loss of functions, is depolarization of neuronal membranes by ion flow at a
known to be mediated by circuits intrinsic to the spinal frequency set by the stimulation device, rather than by the
cord. patterned afferent impulses that normally control neuronal
On the other hand, with the passage of time, this sort discharge. For this reason, it has been supplanted for many
of denervation-induced shock dissipates. In addition, adap- purposes by administration of neurotransmitter agonists,
tive changes occur that allow other regions to take over either by direct microinjection or by diffusion from a
lost functions. This is mediated by sprouting of new con- microdialysis membrane that is placed in the target area
nections to compensate for the loss. A striking phenome- and perfused with high concentrations of agonists.
non seen after placement of lesions aimed at identifying One cannot assume that responses produced by such
the brain regions responsible for REM and non-REM agonist administration demonstrate a normal role for the
sleep is that even massive lesions targeted at putative sleep- applied ligand. For example, many transmitter agonists
generating centers often produce only a transient disrup- and antagonists have been administered to the pontine
tion or reduction of sleep, presumably as a result of this regions thought to trigger REM sleep. In some cases, this
compensation. administration has increased REM sleep. But we can only
A particularly useful approach to the understanding of conclude from this that cells in the region of infusion have
REM sleep generation has been the transection technique. receptors for the ligand and have connections to REM
In this approach, the brain is cut at the spinomedullary sleepgenerating mechanisms. Under normal conditions,
junction, at various brainstem levels, or at forebrain levels these receptors may not have a role in triggering the state.
by passing a knife across the coronal plane of the neuraxis. Only by showing that the administration duplicates the
Regions rostral to the cut may be left in situ or may be normal pattern of release of the ligand in this area, and
removed. It might be expected that such a manipulation that blockade of the activated receptors prevents normal
would completely prevent sleep phenomena from appear- REM sleep, can a reasonable suspicion be raised that a part
ing on either side of this cut. However, to a surprising of the normal REM sleep control pathway has been
extent this is not the case. As we will review later, REM identified.
sleep reappears within hours after some of these lesions. Because it is far easier to inject a substance than to
When both parts of the brain remain, signs of REM sleep collect and quantify physiologically released ligands, there
usually appear on only one side of the cut. This kind of have been many studies implying that various substances
positive evidence is much more easily interpreted than loss are critical for REM sleep control based solely on micro-
of function after lesions, because we can state with cer- injection. These results must be interpreted with caution.
tainty that the removed regions are not essential for the For example, hypocretin is known to depolarize virtually
signs of REM sleep that survive. all neuronal types. It should therefore not be surprising
It will soon be possible to acquire mutant mice in which to find that hypocretin microinjection into arousal systems
any one, or several, of more than 20,000 genes are inacti- such as the locus coeruleus produces arousal,26 that micro-
vated, to the extent that such deletions are not lethal. injection of hypocretin into sites known to control feeding
Investigation of two mutants19,20 led to major insights into increases food intake,27 that injection into regions known
the etiology of human narcolepsy.21-23 Techniques for the to contain cells that are waking active increase waking,28
postnatal inactivation of genes permit investigation of gene that injection into regions known to contain cells selec-
deletions without the developmental effect of these dele- tively active in REM sleep will increase the occurrence
tions. They can also be used for investigation of the effects of this state,29,30 that injection into regions known to
of gene inactivation in particular brain regions. A similar facilitate muscle tone will increase tone, that identical
inactivation can be achieved by localized microinjections injections into regions known to suppress tone will
of antisense. Many if not most such mutants can be decrease tone,31 and that intracerebroventricular injection
expected to have some sleep phenotype, such as increases of hypocretin can increase water intake32 and can activate
or decreases in total sleep or REM sleep time, altered sleep other periventricular systems.29 Such types of findings do
CHAPTER 8 REM Sleep* 95

not by themselves demonstrate a role for hypocretin (or executive neurons for REM sleep control on the basis of
any other neurotransmitter) in the observed behavior. It their tonic latencies were later found to have no essential
is necessary to obtain data on the effects of inactivation role in the generation of REM sleep.35-37 The more appro-
of, for example, hypocretin or hypocretin receptors, and priate comparison of the unit activity cycle to state control
to record evidence that indicates activity of hypocretin is to compare two different cell types to see what the phase
neurons at the appropriate times before seriously enter- relationship of the peaks or troughs of their activity is,
taining such conclusions. under similar conditions. This kind of study is difficult,
Genetic manipulations enable activation of neurons or involving the simultaneous long-term recording of multi-
nonneuronal cells of a particular type. A recent example ple cells, and it is rarely performed. Even in this case, a
of a genetic approach is the insertion of a light-sensitive phase lead does not by itself prove that the lead neuron
ion channel into hypocretin cells using a lentivirus. Fiber- is driving activity seen in the following neuron, but it
optic delivery of light could then be used to activate just does indicate that the reverse is not the case. However,
these cells and determine the effect on sleepwaking awakening is a process that can be studied in this way,
transitions.33 because it can be elicited by stimuli, and it appears to be
preceded by abrupt changes in the activity of many neuro-
Observation of Neuronal Activity nal groups.38 A major advantage of unit recording
Recording the activity of single neurons in vivo can provide approaches in the intact animal to understand sleep and
a powerful insight into the precise time course of neuronal other behavioral processes is their high level of temporal
discharge. Unit activity can be combined with other tech- resolution.
niques to make it even more useful. For example, electrical Observation of the normal pattern of neurotransmitter
stimulation of potential target areas can be used to anti- release and neuronal activity can help determine the neu-
dromically identify the axonal projections of the recorded rochemical correlates of sleep states. The natural release
cell. Intracellular or juxtacellular34 labeling of neurons with of neurotransmitters can be most easily determined by
dyes, with subsequent immunolabeling of their transmit- placing a tubular dialysis membrane 1 to 5mm in length
ter, can be used to determine the neurotransmitter pheno- in the area of interest and circulating artificial cerebrospi-
type of the recorded cell. Combined dialysis and unit nal fluid through it. Neurotransmitters released outside
recording or iontophoresis of neurotransmitter from mul- the membrane will diffuse through the membrane and can
tiple-barrel recording and stimulating micropipettes can be collected. Each sample is collected at intervals, typically
be used to determine the transmitter response of the ranging from 2 to 10 minutes. The collected dialysates can
recorded cell, although it cannot be easily determined be analyzed by chromatography, radioimmunoassay, mass
whether the effects seen are the direct result of responses spectroscopy, or other means. The temporal resolution of
in the recorded cell or are mediated by adjacent cells pro- this technique is typically on the order of a few minutes
jecting to the recorded cell. Such distinctions can be made for each sample.39-41
in in vitro studies of slices of brain tissue by blocking syn- Unit recording and dialysis approaches require a sharp
aptic transmission or by physically dissociating studied research focus on a particular neurotransmitter or neuro-
cells, but in this case their role in sleep may not be easily nal group. In contrast, histologic approaches can be used
determined. to measure the activity of the entire brain at cellular levels
Although the role of a neuron in fast, synaptically medi- of resolution. The most popular such approach in animal
ated events happening in just a few milliseconds can be studies labels the activation of immediate early genes.
traced by inspection of neuronal discharge and comparison These genes are expressed when a neuron is highly active,
of that discharge with the timing of motor or sensory and their expression is an early step in the activation of
events, such an approach may be misleading when applied other downstream genes mobilizing the response of the
to the analysis of sleep-state generation. The sleep cycle cell to activation. The likelihood of such expression may
consists of a gradual coordinated change in EEG, EMG, vary between neuronal types and may not be detectable
and other phenomena over a period of seconds to minutes, below some activity-dependent threshold. Activation of
as waking turns into non-REM sleep and then as non-REM these genes can be detected by immunohistochemistry,
sleep is transformed into REM sleep. most commonly by staining for the production of the Fos
Despite this mismatch of time courses, the tonic latency, protein or the mRNA used to synthesize this protein.42
a measure of how long before REM sleeponset activity in Neurons can be double-labeled to determine the transmit-
a recorded cell changes, has been computed in some ter they express, allowing investigators to determine, for
studies. Neurons purported to show a significant change example, whether histaminergic neurons in the posterior
in activity many seconds or even minutes prior to REM hypothalamus were activated in a particular sleep or waking
sleep onset have been reported. However, such a measure state. Metabolic labels such as 2-deoxyglucose can also
is of little usefulness, because at the neuronal level, the provide an indication of which neurons are active.42,43
activity of key cell groups can best be seen as curvilinear Similar techniques using radioactive ligands in positron
over the sleep cycle, rather than changing abruptly in the emission tomography (PET) studies can be used in living
way that activity follows discrete sensory stimulation. A humans or animals. In vivo measurements of blood flow
major determinant of the tonic latency, computed as can be made throughout the brain with functional mag-
defined here, is the level of noise, or variability in the cells netic resonance imaging (fMRI). All of these techniques
discharge, which affects the difficulty of detecting a signifi- have in common an ability to make anatomically driven
cant underlying change in rate in a cell population. It is discoveries of brain region involvement in particular states,
therefore not surprising that cell groups designated as independent of specific hypotheses, thus leading to major