Visit https://ebookfinal.

com to download the full version and

explore more ebook

Fundamentals of Sleep Medicine Expert Consult Online

and Print 1 Har/Psc Edition Richard B. Berry Md

_____ Click the link below to download _____

https://ebookfinal.com/download/fundamentals-of-sleep-
medicine-expert-consult-online-and-print-1-har-psc-
edition-richard-b-berry-md/

Explore and download more ebook at ebookfinal.com

Here are some recommended products that might interest you.
You can download now and explore!

Movement Disorders in Childhood Expert Consult Online and

Print Expert Consult Title Online Print 1 Har/Psc Edition
Harvey S. Singer Md
https://ebookfinal.com/download/movement-disorders-in-childhood-
expert-consult-online-and-print-expert-consult-title-online-
print-1-har-psc-edition-harvey-s-singer-md/
ebookfinal.com

Neuromuscular Ultrasound Expert Consult Online and Print

1e 1 Har/Psc Edition Francis Walker Md

https://ebookfinal.com/download/neuromuscular-ultrasound-expert-
consult-online-and-print-1e-1-har-psc-edition-francis-walker-md/

ebookfinal.com

Atlas of Clinical Sleep Medicine Expert consult Online and

Print 1st Edition Meir H. Kryger

https://ebookfinal.com/download/atlas-of-clinical-sleep-medicine-
expert-consult-online-and-print-1st-edition-meir-h-kryger/

ebookfinal.com

Conn s Current Therapy 2012 Expert Consult Online and

Print 1 Har/Psc Edition Edward T. Bope Md

https://ebookfinal.com/download/conn-s-current-therapy-2012-expert-
consult-online-and-print-1-har-psc-edition-edward-t-bope-md/

ebookfinal.com
Manual of Surgical Pathology Expert Consult Online and
Print Expert Consult Title Online Print Third Edition
Susan C. Lester Md Phd
https://ebookfinal.com/download/manual-of-surgical-pathology-expert-
consult-online-and-print-expert-consult-title-online-print-third-
edition-susan-c-lester-md-phd/
ebookfinal.com

Aesthetic and Reconstructive Surgery of the Breast Expert

Consult Expert Consult Title Online Print 1 Har/Psc/
Edition Elizabeth Hall-Findlay
https://ebookfinal.com/download/aesthetic-and-reconstructive-surgery-
of-the-breast-expert-consult-expert-consult-title-online-print-1-har-
psc-edition-elizabeth-hall-findlay/
ebookfinal.com

Travel Medicine Expert Consult Online and Print 3e 3rd

Edition Jay S. Keystone Md

https://ebookfinal.com/download/travel-medicine-expert-consult-online-
and-print-3e-3rd-edition-jay-s-keystone-md/

ebookfinal.com

Developmental Behavioral Pediatrics Expert Consult Online

and Print 4th Edition William B. Carey Md

https://ebookfinal.com/download/developmental-behavioral-pediatrics-
expert-consult-online-and-print-4th-edition-william-b-carey-md/

ebookfinal.com

Emergency Medicine Expert Consult 1 Har/Psc Edition James

G. Adams Md Facep

https://ebookfinal.com/download/emergency-medicine-expert-
consult-1-har-psc-edition-james-g-adams-md-facep/

ebookfinal.com
Fundamentals of Sleep Medicine Expert Consult Online
and Print 1 Har/Psc Edition Richard B. Berry Md Digital
Instant Download
Author(s): Richard B. Berry MD
ISBN(s): 9781437703269, 1437703267
Edition: 1 Har/Psc
File Details: PDF, 14.36 MB
Year: 2011
Language: english
FUNDAMENTALS of

Sleep Medicine
FUNDAMENTALS of

Sleep Medicine
Richard B. Berry, MD
Professor of Medicine
University of Florida, Gainesville
Medical Director
University of Florida and Shands Sleep Disorder Center
Gainesville, Florida
1600 John F. Kennedy Blvd.
Ste 1800
Philadelphia, PA 19103-2899

FUNDAMENTALS OF SLEEP MEDICINE ISBN: 978-1-4377-0326-9

Copyright © 2012 by Saunders, an imprint of Elsevier Inc.

All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any
means, electronic or mechanical, including photocopy, 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 Publisher’s permissions policies and our arrangements with organizations such as the
Copyright Clearance Center and the Copyright Licensing Agency, can be found at our website: www.elsevier.
com/permissions.

This book and the individual contributions contained in it are protected under copyright by the Publisher
(other than as may be noted herein).

Notices

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

Berry, Richard B., 1947–

Fundamentals of sleep medicine / Richard B. Berry.
    p. ; cm.
Includes bibliographical references and index.
ISBN 978-1-4377-0326-9 (pbk. : alk. paper) 1. Sleep disorders. 2. Sleep–Physiological aspects. I. Title.
[DNLM: 1. Sleep–physiology. 2. Sleep Disorders. WL 108]
RC547.B47 2012
616.8′498—dc22 2011008073

Acquisitions Editor: Julie Goolsby

Developmental Editor: Jessica Pritchard
Publishing Services Manager: Pat Joiner-Myers
Project Manager: Marlene Weeks
Design Manager: Louis Forgione
Marketing Manager: Carla Holloway

Working together to grow

libraries in developing countries
Printed in the United States of America www.elsevier.com | www.bookaid.org | www.sabre.org

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

This book is dedicated to my wife Cathy, my son David,
and my daughter Sarah.
They are my greatest joy.
 Preface

The goal of this book is to provide the reader with a core of fundamental knowledge about
sleep medicine and polysomnography. I have tried to write the book so that a person
without training in sleep medicine can start reading chapter one and progress until the end
of the book. With the publication of the International Classification of Sleep Disorders,
Second Edition, and the American Academy of Sleep Medicine scoring manual, there is a
need for an up-to-date text using current terminology and diagnostic criteria. A single text
cannot hope to cover all aspects of sleep medicine and sleep physiology. Therefore, I have
tried to focus on information that I feel is the most clinically useful. The field of sleep
medicine is changing so rapidly that any text is “out-of-date” before it is even published.
To this end, there will be an associated website to allow for updates, corrections, review
questions, illustrative case studies, and some video clips of parasomnias.
My inspiration for writing this book has come in part from the satisfaction and appre-
ciation that I received from writing the text Sleep Medicine Pearls. Sleep Medicine Pearls
had short fundamentals chapters mixed with cases and was heavy on graphics and illustra-
tive sleep tracings. I have tried to amplify the fundamentals sections to provide what I hope
is a concise and useful introduction to the entire spectrum of sleep disorders. In particular,
I have tried to cover aspects of the technology of sleep monitoring and interpreting sleep
studies that many new to the sleep field find difficult. The challenging but enjoyable experi-
ence of teaching sleep fellows and residents about sleep medicine has also prompted me to
write a book covering the fundamentals both to serve as an introductory text and to assist
those physicians actively taking care of sleep patients.

Richard B. Berry, MD

vii
viii Chapter 21   

Acknowledgments

I would like to express my gratitude for the support and encouragement of the University
of Florida sleep physicians, including Dr. Abby Wagner, co-director of the University of
Florida Sleep Medicine Fellowship, Dr. Stephan Eisenschenk, and Dr. Craig Foster. It is a
pleasure to work with such a dedicated and talented group of individuals. I would also like
to thank Dr. Klark Turpen for her assistance in editing the book chapters. The patience and
assistance of the Elsevier editorial staff is also greatly appreciated. Jessica Pritchard helped
assemble the chapters and many figures. Dolores Meloni, Senior Acquisitions Editor, was
instrumental in developing the concept for the book and provided critical support in the
planning stages. Julie Goolsby, Associate Acquisitions Editor, provided encouragement
during the final stages of book preparation. I am also grateful for the patience and diligence
of Berta Steiner of Bermedica Production, Ltd. during the production process.

viii
 Chapter 1

Sleep Stages and Basic

Sleep Monitoring
recordings are used to detect REM sleep, which is character-
Chapter Points ized by REMs and reduced muscle tone. Since 1968, sleep
• In the EEG or EOG derivation G1-G2, an upward deflection was usually staged according to A Manual of Standardized
in the tracing is noted if input G1 becomes negative with Terminology, Techniques and Scoring System for Sleep Stages
respect to input G2 (negative upward polarity). of Human Subjects, edited by Rechtschaffen and Kales
• To differentiate whether alpha waves or sleep spindles (R&K).1 In the R&K scoring manual,1 NREM sleep was
are present, change to a 10-second window and count divided into sleep stages 1, 2, 3, and 4. REM sleep was referred
the individual deflections in one second (see Fig. 1–3). to as stage REM. Sleep stage nomenclature has changed fol-
• K complexes and slow waves have the greatest lowing the publication of the American Academy of Sleep
amplitude in frontal derivations. Sleep spindles and Medicine (AASM) Manual for the Scoring of Sleep and Associ-
saw-tooth waves have the greatest amplitude in ated Events (hereafter referred to as the AASM scoring
central derivations. manual).2 The new nomenclature was introduced to denote
• Alpha activity is any wave form with a frequency of 8 sleep stages defined by new criteria. The old and new nomen-
to 13 Hz. Alpha rhythm has a frequency of 8 to 13 Hz, clatures are shown in Table 1–1. Stages 3 and 4 are combined
is most prominent in the occipital derivations, and is into stage N3.
enhanced by eye closure and attenuated by eye Today, digital polysomnography (sleep recording) has
opening. virtually replaced recording on paper. However, previously
• The recommended EEG derivations are F4-M1, C4-M1, sleep recording was performed with polygraphs using ink
and O2-M1. writing pens with a paper speed of 10 mm/sec. At this paper
• The recommended EOG derivations are E1-M2 and speed, a 30-cm page of paper contained 30 seconds of record-
E2-M2. Both eye electrodes are referred to a common ing. A sleep stage was identifed for each page (30 sec) termed
mastoid electrode M2. an epoch. The tradition of staging sleep in 30-second epochs
• The front of the eye (cornea) is positive with respect to has been retained in the recent AASM scoring manual. The
the back of the eye (retina). If the eyes move toward sleep stage assigned to each epoch is the stage occupying the
E1-M2 and away from E2-M2, this causes a downward majority of time within that epoch. Digital recording allows
deflection in E1-M2 and an upward deflection in E2-M2. display of data in one of several time windows (typically 5,
• In the recommended EOG derivations, eye movements 10, 30, 60, 90, 120, 240 sec). The 10-second window corre-
result in out-of-phase deflections. K complexes result sponds to a paper speed of 30 mm/sec and is used for clinical
in in-phase deflections. EEG monitoring. This time window also approximates elec-
• In stage R, the chin EMG amplitude is equal to or lower trocardiographic (ECG) recording that was typically per-
than the lowest level in NREM sleep. The chin EMG formed using a paper speed of 25 mm/sec before the current
activity can reach the REM level during NREM sleep. use of digital ECG recording.
Transitions from NREM to stage R are not always
associated with a drop in chin activity. Chin EMG
activity is useful in differentiating stage R from stage W EEG ELECTRODE PLACEMENT
with the eyes open (REMs present). Monitoring to detect the presence and stage of sleep requires
only a portion of the electrodes used in standard clinical
Sleep is divided into non–rapid eye movement (NREM) EEG recording (Table 1–2). The nomenclature for the EEG
and rapid eye movement (REM) sleep. Sleep staging is based electrodes follows the International 10–20 system.3 The “10–
on electroencephalographic (EEG), electro-oculographic 20” refers to the fact that the electrodes are positioned at
(EOG), and submental (chin) electromyographic (EMG) either 10% or 20% of the distance between landmarks. The
criteria. EOG (eye movement recording) and chin EMG major landmarks include the nasion (bridge of the nose),
1
2 Chapter 1 Sleep Stages and Basic Sleep Monitoring

TABLE 1–1 TABLE 1–2

Sleep Stage Nomenclature Electroencephalographic Electrode Nomenclature
R&K AASM LEFT RIGHT MIDLINE
Wake Stage W Stage W Frontopolar Fp1 Fp2 Fpz
NREM Stage 1 Stage N1 Frontal F3 F4 Fz
Stage 2 Stage N2
Central C3 C4 Cz
Stage 3 Stage N3
Stage 4 Occipital O1 O2 Oz
REM Stage REM Stage R Mastoid M1 M2
AASM = American Academy of Sleep Medicine ; NREM = non–rapid eye
2

movement; R&K = Rechtschaffen and Kales A1; REM = rapid eye movement;
stages 3 and 4 are combined into stage N3.

FIGURE 1–1 Electrode positions using the 10–20 Nasion

system.
10%

20% Fpz Fp2

Fp1

F3 Fz F4
20%

C3 Cz C4

M1 M2
20%

O1 O2

20% Oz

10%

Inion

inion (prominence at base of the occiput), and preauricular

EEG Derivations
points (Figs. 1–1 and 1–2). In the 10–20 system, even-
numbered subscripts refer to the right side of the head and EEG signals are displayed as voltage differences between two
odd-numbered subscripts to the left. Electrodes are named electrodes. The term derivation refers to a set of two elec-
for the part of the brain they are over. For example, Fp1 and trodes (and the voltage difference between the electrodes).
Fp2 are the left and right frontal pole electrodes, F3 and F4 The term montage refers to a particular set of derivations.
are the left and right frontal electrodes, C3 and C4 are the In sleep monitoring, electrodes in the frontal, central, and
left and right central electrodes, and O1 and O2 are the left occipital electrodes are referenced against the opposite
and right occipital electrodes. Electrodes in the midline in mastoid electrode. The AASM scoring manual recommends
the frontopolar, frontal, central, and occipital regions are that all of the following electrodes be placed (F3, F4, C3, C4,
named Fpz, Fz, Cz, and Oz, respectively. The position of the O1, O2, M1, and M2). The recommended derivations and the
electrode Cz is at the top of the head and is called the vertex. alternative derivations are listed in Table 1–3. The backup
The left and right mastoid electrodes in the new AASM derivations are displayed if one of the electrodes in the rec-
scoring manual nomenclature are named M1 and M2, respec- ommended derivation fails. For example, if electrode F4 fails,
tively. They were previously named A1 and A2. The nomen- the derivation F3–M2 is used. In digital recording, one can
clature of EEG electrodes used for sleep monitoring is listed easily display all six derivations if desired at the same time.
in Table 1–2. In the original R&K scoring manual, only central derivations
 Chapter 1    Sleep Stages and Basic Sleep Monitoring 3

Cz FIGURE 1–2 Side, front, and back views

C3
of electroencephalographic (EEG) electrode
C4 F4 F3
Vertex placement using the 10–20 system.
20% O2 O1
20% Oz
20%
F4
C4 20% M2
20% M1

10% 20%
10%
10% O2 10% Nasion

Inion Preauricular F4 F3
M2 Fpz
point
Fp2 Fp1
O1 Oz O2

Inion

TABLE 1–3 term derivation is used to describe the differential signal

Recommended Electroencephalographic between two inputs. For example, in the derivation C4-M1, a
Derivations with Backup Derivations change in the voltage between these electrodes results in an
upward deflection if C4 is negative with respect to M1. EEG
RECOMMENDED DERIVATIONS2 BACKUP DERIVATIONS
activity is described by frequency in cycles per second
• F4-M1 • F3-M2 (hertz = Hz), amplitude (microvolts [µV]), and shape. The
• C4-M1 • C3-M2 classically described EEG frequency ranges are delta (<4 Hz),
theta (4–7 Hz), alpha (8–13 Hz), and beta (>13 Hz). Activity
• O2-M1 • O1-M2
that is faster results in narrower deflections and slower fre-
quency results in wider deflections. Sharp waves are narrow
were utilized. However, additional derivations allow better waves of 70 to 200 msec duration and spikes have a shorter
visualization of the EEG patterns used to stage sleep. duration of 20 to 70 msec.
Although the derivations cited previously are the most Some of the characteristics of EEG patterns important for
widely used, the AASM scoring manual lists alternative sleep staging are listed in Tables 1–5 through 1–9. In addi-
acceptable derivations (Table 1–4). The alternative deriva- tion to frequency, the region of highest activity (amplitude)
tions use the electrodes Fz, Cz, Oz, C4, and M1 with the backup and the effects of maneuvers on the EEG activity are also
electrodes Fpz (to replace Fz), C3 (to replace Cz or C4), O1 (to important. For example, one could use the term “alpha activ-
replace Oz), and M2 (to replace M1). ity” to describe any EEG activity with a frequency in the
It is common to place additional electrodes beyond those alpha range (8–13 Hz). However, alpha rhythm consists of
discussed in the AASM scoring manual to serve as a ground activity most prominent in occipital derivations that is atten-
electrode and common reference. In sleep monitoring, a uated by eye opening and increased by eye closure (Fig. 1–3).
ground electrode is usually placed at or near Fpz and con- An important part of biocalibrations (see Chapter 4) at the
nected to the ground (or iso-ground) input on the electrode start of sleep recording is to ask patients to close and then
box. As discussed in Chapter 2, the ground is used to balance open their eyes to document that they produce alpha rhythm.
the individual AC differential amplifiers. One electrode (or Bursts of alpha waves can also occur during stage R typically
two linked electrodes) is also placed to serve as a reference at a frequency 1 to 2 Hz slower than during wakefulness.
for referential recording (see Chapter 2). The reference elec- Sleep spindles7–9 are bursts of activity with a frequency
trode is commonly placed at or near Cz depending on which range of 11 to 16 Hz (usually 12–14) with a duration of 0.5 sec
EEG electrodes are to be recorded for sleep monitoring. or greater (usually 0.5–1.5 sec). The term spindle is used
because the shape of sleep spindle burst is often like that of a
yarn spindle (see Fig. 1–3). If there is uncertainty about
EEG Patterns
whether activity is a burst of alpha activity or a sleep spindle,
Recognition of certain characteristic EEG patterns is essen- one can display a 10-second window (see Fig. 1–3) and actu-
tial for sleep staging.1–6 EEG activity is recorded using a ally count the deflections (waves) per second. Sleep spindles
differential AC amplifier such that the signal recorded is the arise from thalamocortical oscillations. The reticular nucleus
difference in voltage between two inputs (G1 and G2). By of the thalamus is responsible for generating sleep spindles.
EEG convention, if input G1 is negative with respect to G2, A K complex1,2,8,9 is a high-amplitude biphasic wave
this results in an upward deflection. As noted previously, the composed of an initial negative sharp wave (deflection up)
4 Chapter 1 Sleep Stages and Basic Sleep Monitoring

TABLE 1–4
Alternative Electroencephalographic Derivations with Backup Derivations
ALTERNATIVE DERIVATIONS FZ FAILS CZ FAILS OZ FAILS C4 OR M1 FAILS
Fz-Cz Fpz-Cz Fpz-C3 Fz-Cz Fz-Cz
Cz-Oz Cz-Oz C3-Oz Cz-O1 Cz-Oz
C4-M1 C4-M1 C4-M1 C4-M1 C3-M2

TABLE 1–5
Characteristics of Alpha Rhythm and Sleep Spindles
ALPHA RHYTHM SLEEP SPINDLES
• 8–13 Hz. • 11–16 Hz (classically 12–14 Hz).
• Most prominent over the occipital areas. • Duration ≥ 0.5 sec (0.5–1.5 sec).
• Activity increased by eye closure. • Maximal over central areas.
• Activity suppressed by eye opening. • One of the defining characteristics of stage N2.
• Predominate EEG activity in drowsy, eyes closed stage W. • Thalamocortical oscillations (reticular thalamic
• Common in REM sleep (1–2 Hz slower than during stage W or N1). nucleus).
• Can occur with arousals (brief awakenings). • Can be seen in stage N3 sleep.
• 10% of persons do not produce alpha rhythm with eye closure. • Drug spindles (benzodiazepines) may be slightly faster.
EEG = electroencephalographic; REM = rapid eye movement.

TABLE 1–6
Characteristics of K Complex and Slow Wave Activity
K COMPLEX SLOW WAVE ACTIVITY
• High amplitude–biphasic deflection. • Frequency 0.5–2 Hz and > 75 µV peak to peak in the
• A well-delineated negative sharp wave (upward) followed by a frontal derivations.
positive (downward) slow wave. • Used to define stage N3 sleep.
• Stands out from the lower voltage background. • Stage N2 < 20% SWA (<6 sec).
• Duration ≥ 0.5 sec. • Stage N3 ≥ 20% SWA (≥6 sec).
• Characteristic of stage N2 sleep. • SWA is usually transmitted to eye derivations.
• Maximal over frontal areas (frontal > central > occipital).
• K complex–associated arousal requires arousal to start no more
than 1 second after K complex termination.
SWA = slow wave activity.

TABLE 1–7
Characteristics of Vertex Sharp and Saw-Tooth Waves
VERTEX SHARP WAVES SAW-TOOTH WAVES
• Sharply contoured waves • Trains of triangular waves, often serrated
• Duration < 0.5 sec • 2–6 Hz waves
• Maximal over the central region (derivations containing C3, C4, Cz) • Maximal in amplitude in central derivations
and distinguishable from the background activity (higher amplitude). • Often, but not always, preceding a burst of REMs
• Occurs in stage N1 often near transition to stage N2 • Characteristic of stage R but not required for
scoring stage R
REMs = rapid eye movements.

followed by a slow wave (Fig. 1–4). A burst of spindle activity complex. An arousal during sleep stages N1, N2, and N3 is
is often superimposed on a K complex. A K complex stands scored if there is an abrupt shift of EEG frequency including
out from the lower voltage background. K complex activity alpha, theta, and/or frequencies greater than 16 Hz (but not
is greatest in frontal derivations (also central > occipital). A spindles) that lasts at least 3 seconds, with at least 10 seconds
K complex is said to be associated with an arousal if the of stable sleep preceding the change. Arousals are discussed
arousal commences no more than 1 second after the K in more detail in Chapter 3.
 Chapter 1    Sleep Stages and Basic Sleep Monitoring 5

FIGURE 1–3 Alpha rhythm and sleep spindle activity as visualized

Alpha Sleep spindle in 30-second and 10-second windows.

30 second
window
(as viewed)

1 sec

10 second
window
(as viewed)

1 sec

TABLE 1–8
Summary of Important Wave Form Characteristics
ALPHA SLEEP VERTEX SLOW WAVE SAW-TOOTH
RHYTHM SPINDLE K COMPLEX SHARP WAVE ACTIVITY WAVES
Frequency (Hz) 8–13 11–16 N/A N/A 0.5–2 2–6
Amplitude/ Oscillation Spindle- High amplitude Sharp wave High-amplitude Triangular,
shape shaped (usually > 100 µV) broad wave serrated
oscillation Stands out against >75 µV peak to
EEG background peak
Biphasic-negative
sharp wave
followed by positive
component
Duration Variable ≥0.5 >0.5 sec <500 msec 0.5–2 sec Variable
Location of Occipital Central Frontal Central Frontal Central
highest (vertex)
amplitude
Associated Stage W Stage N2 Stage N2 Stage N1 Stage N2 Stage R
sleep stages/ Stage N1 Stage N3 Stage N3 Stage N3
events Stage R
Arousals
EEG = electroencephalogram; N/A = not applicable.

TABLE 1–9 An example of a K complex associated with an arousal is

Electro-oculographic Derivations shown in Figure 1–5. Also note that the K complex is seen
RECOMMENDED ALTERNATE in the EOG derivations E1-M1 and E2-M2 as an in-phase
deflection.
E1-M2 E1-Fpz As noted previously, the frequency of delta activity is less
E2-M2 E2-Fpz than 4 Hz. EEG activity in this range produces relatively
wide duration deflections, often called delta or slow waves
(see Fig. 1–4). However, for sleep staging, the designation
slow wave activity (SWA)2 specifically refers to waves with
a frequency range of 0.5 to 2 Hz (2- to 0.5-sec duration) and
6 Chapter 1 Sleep Stages and Basic Sleep Monitoring

100 µV

1 sec
V

F4-M1

C4-M1

O2-M1

E1-M2

E2-M2

Chin EMG

Alpha waves Sleep spindle K complex Vertex sharp Slow waves Saw-tooth waves
wave

FIGURE 1–4 Important EEG patterns for sleep staging. The grid lines are 1 second apart. V = position of the vertex sharp wave.

75 µV

1 sec
F4-M1

C4-M1

O2-M1

E1-M2

E2-M2

Chin1-Chin2

FIGURE 1–5 A K complex associated with an arousal. An abrupt shift in EEG frequency immediately follows the K complex that lasts greater than 3 seconds. To be
considered associated with a K complex, an arousal must commence no later than 1 second after K complex termination.

a peak-to-peak amplitude of greater than 75 µV in the meeting amplitude criteria) using the AASM scoring manual
frontal derivations (see Fig. 1–4). SWA has the greatest definition2 (frontal derivations) compared with the R&K
amplitude over frontal areas. In the R&K definitions, only definition (using central derivations).
central derivations were utilized. Because slow wave ampli- Vertex sharp waves (see Fig. 1–4) are narrow-duration
tude is higher over the frontal areas, a given epoch of EEG waves (<500 msec according to the AASM scoring manual2)
activity would potentially have greater SWA (longer duration prominent in derivations containing electrodes near the
 Chapter 1    Sleep Stages and Basic Sleep Monitoring 7

Fpz FIGURE 1–6 Recommended, previous, and alternate

E2
ROC eye movement electrode positions. LOC = left outer
canthus; ROC = right outer canthus.
1 cm
1 cm 1 cm

1 cm 1 cm 1 cm
LOC
E1 E2 E1

Recommended Previous Alternate

vertex (Cz, C3, C4). They are often seen near the transition E2
Look Look
between stage N1 and stage N2 sleep. Saw-tooth waves (see left right
Fig. 1–4) occur during REM sleep, although they are not    
always present during this sleep stage. They are triangular
waves of 2 to 6 Hz of highest amplitude in the central deriva- E1
tions. The presence of saw-tooth waves is not required to  
score stage R. However, the presence of saw-tooth waves is E1-M2
very helpful when they occur.
 
EOG MONITORING FOR SLEEP E2-M2
Recording of eye movements is possible because a potential
difference exists across the eyeball with the front/cornea E1 positive to M2, E1 negative to M2,
positive (+) and back/retina negative (–). Eye movements are deflection down deflection up
detected by EOG recording of voltage changes associated E2 negative to M2, E2 positive to M2,
with eye movement. deflection up deflection down
The recommended EOG electrodes in the AASM scoring FIGURE 1–7 Schematic shows deflections in E1-M2 and E2-M2 from eye movements.
manual2 are illustrated in Figure 1–6. E1 and E2 refer to the
left and right eye electrodes, respectively. Previously eye
electrodes were named right outer canthus (ROC) and left In the recommended EOG derivations, eye movements
outer canthus (LOC). For comparison, the positions of the result in out-of-phase deflections. This is because eye move-
ROC and LOC electrodes are also shown. Please note that ments are conjugate, and when both eyes move laterally or
E1 is placed below the LOC and E2 is placed above the ROC, vertically, they both move toward one EOG electrode and
whereas LOC and ROC were placed lateral to the respective away from the other EOG electrode. The polarity of the eye
outer canthus. Because E1 is below and E2 above the eyes, electrodes determines the net voltage difference of the EOG
vertical as well as horizontal movement can be detected. derivations because the electrodes are much closer to the
Alternate eye electrode positions were also recommended eyes than M2. The schematic in Figure 1–7 illustrates eye
for use with alternate eye movement derivations (see movements and the resulting deflections (this assumes that
Fig. 1–6). The AASM scoring manual recommends the both eye derivation tracings have negative polarity upward
EOG derivations E1-M2 and E2-M2 (see Table 1–8). Note which is standard).
that both eye derivations use the right mastoid (M2) as the Note that when the alternate EOG derivations E1-Fpz and
reference electrode. Previous ROC and LOC derivations E2-Fpz are used, both E1 and E2 are 1 cm below and 1 cm lateral
varied between sleep centers, and these electrodes were ref- to the LOC and ROC, respectively. In this scheme, vertical
erenced either to the same mastoid or to the opposite eye movements result in in-phase deflections and lateral eye
mastoid. The AASM scoring manual also specified the alter- movements result in out-of-phase deflections (Fig. 1–8). The
native eye movement derivations (E1-Fpz and E2-Fpz). If these advantages of the alternative EOG derivations are that verti-
eye movement derivations are used, both E1 and E2 are cal deflections tend to produce larger deflections (blinks are
below and lateral to the LOC and ROC, respectively (see more prominent) and one can distinguish vertical (in-phase)
Fig. 1–6). from horizontal (out-of-phase) eye movements. In addition,
When the eyes move toward an electrode, a positive it is easy to remember that downward eye movements result
voltage is recorded (Fig. 1–7). Recall that in EEG recording, in downward deflections in the eye derivations and upward
by polarity convention, if an eye electrode is negative com- eye movements result in upward deflections. Alternatively,
pared with the reference electrode, the signal has an upward the recommended eye derivations make it easier to recognize
deflection. Thus, eye movement (cornea +) toward an elec- artifacts or EEG activity transmitted to the eye derivations
trode referenced to another electrode further away from the because these cause in-phase deflections while eye move-
eyes results in a downward deflection. ments cause out-of-phase deflections (Fig. 1–9).
8 Chapter 1 Sleep Stages and Basic Sleep Monitoring

Fpz Look Look Eye Movement Patterns

right left
Typical eye movement patterns (Table 1–10) include blinks,
   
slow eye movements (SEMs), REMs, and reading eye move-
E2 E1   ments (Fig. 1–10). SEMs are typical of eyes closed drowsy,
wakefulness, and stage N1 sleep. REMs are seen in eyes open
E1-Fpz
wakefulness or stage R sleep. SEMs typically disappear with
the onset of stage N2 sleep. However, patients on selective
E2-Fpz
serotonin reuptake inhibitors (SSRIs) can have eye move-
ments that are a mixture of slow and more rapid activity that
  persists into stage N2.10,11 This pattern is called “Prozac eyes”

TABLE 1–10
Fpz Look Look Eye Movements Pattern Definitions
up down • Eye blinks: Conjugate vertical eye movements at a
    frequency of 0.5–2 Hz present in wakefulness with the
eyes open or closed.
E2 E1   • Reading eye movements: Trains of conjugate eye
E1-Fpz movements consisting of a slow phase followed by a
rapid phase in the opposite direction as the subject reads.
• Slow eye movements: Conjugate, fairly regular,
E2-Fpz sinusoidal eye movements with an initial deflection
lasting > 500 msec.
 
• Rapid eye movements (REMs): Conjugate, irregular,
FIGURE 1–8 Schematic shows deflections in E1-Fpz and E2-Fpz due to horizontal and
sharply peaked eye movements with an initial deflection
vertical eye movements. Note that, using these derivations, vertical eye movements
usually lasting < 500 msec. Whereas rapid eye movements
result in in-phase deflections whereas lateral eye movements result in out-of-phase
are characteristic of stage R sleep, they may also be seen in
deflections. In addition, downward eye movements result in downward deflections.
wakefulness with eyes open (as patients look around the
room)
Adapted from Iber C, Ancoli-Israel S, Chesson A, Quan SF for the American
Academy of Sleep Medicine: The AASM Manual for the Scoring of Sleep and
Associated Events: Rules, Terminology and Technical Specifications, 1st ed.
Westchester, IL: American Academy of Sleep Medicine, 2007.

F4-M1

C4-M1

O2-M1

E1-M1

E2-M2

Chin EMG

K complex REM

FIGURE 1–9 Using the recommended electro-oculographic (EOG) derivations, the K complex results in deflections that are in phase and the rapid eye
movement (REM) results in out-of-phase deflections. The vertical lines are 1 second apart. EMG = electromyography.
 Chapter 1    Sleep Stages and Basic Sleep Monitoring 9

100 µV
1 sec

Slow eye E1-M2

movements
(SEMs) E2-M2

Rapid eye E1-M2

movements
(REMs) E2-M2

Reading eye
E1-M2
movements
E2-M2

E1-M2
Blinks
E2-M2

FIGURE 1–10 Eye movement patterns. The grid lines are 1 second apart.

but can occur with any of the SSRIs (see Chapter 4). Reading THREE ELECTRODES ARE RECOMMENDED
eye movements are due to a slow scan of the written page TO RECORD THE CHIN EMG
(left to right) followed by a rapid return to the left. This
results in a slowly increasing downward deflection in E2-M2
followed by a rapid upward deflection. In E1-M2, there is a
slow upward deflection followed by a rapid downward
deflection (see Fig. 1–10). Chin1

Chin (Submental) EMG Monitoring Chin2 Chin3

The monitoring of chin EMG activity is an essential element
only for identifying stage R (REM sleep). In stage R, the chin
EMG is relatively reduced: the amplitude is equal to or lower Electrode 1. Midline 1 cm above interior edge of mandible
than the lowest EMG amplitude in NREM sleep.
The placement of EMG electrodes recommended by the Electrode 2. 2 cm below inferior edge of mandible and
2 cm right of the midline
AASM scoring manual is illustrated in Figure 1–11. The
scoring manual defines the positions of the electrodes but Electrode 3. 2 cm below inferior edge of mandible and
2 cm left of the midline
does not assign them names.
For convenience, labels are assigned in Figure 1–11. Standard chin EMG derivations Chin2 - Chin1 or Chin3 - Chin1
The standard chin derivation consists of either of the elec-
FIGURE 1–11 Submental (chin) EMG electrode positions. The terms Chin1, Chin2, and
trodes below the mandible referred to the electrode above Chin3 are not specified in the American Academy of Sleep Medicine (AASM) scoring
the mandible. That is chin2–chin1 or chin3–chin1. The elec- manual but are added for convenience. The standard derivation is either of the electrodes
trode not used in the displayed derivation is placed as a below the mandible referred to the electrode above the mandible.
backup.
If the EMG derivation sensitivity (gain) is adjusted high
enough to show some activity in NREM sleep, a drop in
activity may be seen on transition to REM sleep. However, identifying ONLY stage R. The reduction in the chin EMG
the EMG can fall to the REM level before the onset of REM amplitude during REM sleep is a reflection of the general-
sleep. Depending on the gain, a reduction in the EMG ized skeletal-muscle hypotonia present in this sleep stage. In
amplitude from wakefulness to sleep and often a further the tracings in Figure 1–12, there is a fall in chin EMG
reduction on transition from stage N1 to stage N3 may be amplitude (A) just before saw-tooth waves (B) and the REMs
seen. However, chin EMG activity is a requirement for (C) occur.
10 Chapter 1 Sleep Stages and Basic Sleep Monitoring

B
F4-M1

C4-M1

O2-M1

E1-M2
C
E2-M2

Chin EMG
A

FIGURE 1–12 A 30-second tracing shows a reduction in the chin EMG on transition to stage R sleep (A). Note saw-tooth waves (B) and REMs (C).

CLINICAL REVIEW QUESTIONS: B. Minimum EEG amplitude peak to peak > 75 µV in

frontal derivations, frequency 0.5–2 Hz
1. The standard EEG montage for sleep recording is F4-M1,
C4-M1, and O2-M1. If electrode C4 fails, which of the fol- C. Minimum EEG amplitude peak to peak > 75 µV in
lowing montages should be used? frontal derivations, frequency < 4 Hz
A. F4-M1, C3-M2, O2-M1 D. Minimum EEG amplitude peak to peak > 50 µV in
frontal derivations, frequency < 4 Hz
B. F4-M1, C3-M1, O2-M1
C. F3-M2, C3-M2, O1-M2 5. On right lateral gaze, which of the following deflections
D. F4-M2, C3-M2, O2-M2 are noted in the recommended EOG derivations?
A. E1-M2 Deflection up E2-M2 Deflection down
2. Alpha rhythm is characterized by which of the
B. E1-M2 Deflection down E2-M2 Deflection up
following?
C. E1-M2 Deflection up E2-M2 Deflection up
A. 8–13 Hz, attenuated by eye opening, most prominent
in occipital derivations D. E1-M2 Deflection down E2-M2 Deflection down
B. 11–16 Hz, attenuated by eye closure, most prominent 6. Which of the following is true about SEMs (using the
in occipital derivations recommended eye derivations)?
C. 8–13 Hz, attenuated by eye opening, most prominent A. Can occur during stage W or N1, are sinusoidal out-
in frontal derivations of-phase eye movements.
D. 11–16 Hz, attenuated by eye opening, most promi- B. Can occur during stage W only, are sinusoidal out-of-
nent in central derivations phase eye movements.
3. Sleep spindles are characterized by which of the C. Can occur during stage W or N1, are sinusoidal
following? in-phase movements.
A. 12–14 Hz activity, most prominent in the occipital D. Can occur only during stage W only, are sinusoidal
areas out-of-phase movements.
B. 8–13 Hz, thalamocortical oscillations
C. 11–16 Hz, most prominent in frontal derivations Answers
D. 11–16 Hz, generated by the reticular nucleus of the 1. A. The alternate derivation for C4-M1 is C3-M2. It is
thalamus not necessary to change the other derivations. (See
FAQ for scoring manual V4. http://www.aasmnet.org/
4. SWA for sleep staging is characterized by which of the
FAQs.aspx?cid=29)
following?
A. Minimum EEG amplitude peak to peak > 75 µV in 2. A. Alpha rhythm is 8–13 Hz, attenuated by eye opening,
central derivations, frequency 0.5–2 Hz most prominent in occipital derivations.
 Chapter 1    Sleep Stages and Basic Sleep Monitoring 11

3. D. Sleep spindles have a frequency of 11–16 Hz and rep- of Sleep and Associated Events: Rules, Terminology and Tech-
resent thalamocortical oscillations generated by the retic- nical Specifications, 1st ed. Westchester, IL: American Academy
of Sleep Medicine, 2007.
ular nucleus of the thalamus. Sleep spindles are most
3. International Federation of Societies for Electroencephalogra-
prominent in central derivations. phy and Clinical Neurophysiology: Ten twenty electrode
system. EEG Clin Neurophysiol 1958;10:371–375.
4. B. SWA is characterized by a minimum amplitude peak 4. Williams RL, Karacan I, Hursch CJ: Electroencephalography of
to peak of > 75 µV in the frontal derivations with a fre- Human Sleep: Clinical Applications. New York: John Wiley &
quency of 0.5 to 2 Hz. Sons, 1974.
5. West P, Kryger MH: Sleep and respiration: terminology and
5. A. In the recommended derivations, eye movements methodology. Clin Chest Med 1985;6:691–712.
6. Caraskadon MA, Rechschaffen A: Monitoring and staging
cause out-of-phase deflections. Because the cornea is
human sleep. In Kryger MH, Roth T, Dement WC (eds): Prin-
positive with respect to the retina, a rightward gaze results ciples and Practice of Sleep Medicine. Philadelphia: Elsevier
in E2 being positive with respect to M2 (E2 is closer to the Saunders, 2005, pp. 1359–1377.
cornea) and this results in a downward deflection. With 7. DeGennaro L, Ferrara M: Sleep spindles: an overview. Sleep
a rightward gaze, E1 is negative with respect to M2 (upward Med Rev 2003;7:423–440.
8. McCormick L, Nielsen T, Nicolas A, et al: Topographical dis-
deflection).
tribution of spindles and K complexes in normal subjects. Sleep
1997;20:939–941.
6. A. SEMs can occur during wake (eyes closed drowsy 9. Silber MH, Ancoli-Israel S, Bonnet MH, et al: The visual
wake) or stage N1 and are sinusoidal out-of-phase scoring of sleep in adults. J Clin Sleep Med 2007;15:121–131.
movements. 10. Schenck CH, Mahowlad MW, Kim SW, et al. Prominent eye
movements during NREM sleep and REM sleep behavior dis-
REFERENCES order associated with fluoxetine treatment of obsessive-
compulsive disorder. Sleep 1992;15:226–235.
1. Rechtschaffen A, Kales A (eds): A Manual of Standardized 11. Armitage R, Trivedi M, Rush AJ: Fluoxetine and oculomotor
Terminology, Techniques and Scoring System for Sleep Stages activity during sleep in depressed patients. Neuropsychophar-
of Human Sleep. Los Angeles: Brain Information Service/Brain macology 1995;12:159–165.
Research Institute, UCLA, 1968.
2. Iber C, Ancoli-Israel S, Chesson A, Quan SF for the American
Academy of Sleep Medicine: The AASM Manual for the Scoring
 Chapter 2

The Technology of Sleep

Monitoring: Differential
Amplifiers, Digital
Polysomnography,
and Filters
In sleep monitoring (polysomnography [PSG]), electroen-
Chapter Points cephalographic (EEG), electro-oculographic (EOG), and
• Common mode rejection of unwanted signals by a electromyographic (EMG) activity is recorded by differential
differential AC amplifier depends on having low and AC amplifiers that amplify the difference in voltage between
fairly equal electrode impedances. An electrode two inputs1–3 (Fig. 2–1). Each differential amplifier has two
impedance less than 5 KΩ is desirable (<10 KΩ inputs and a ground. By convention in EEG recording, if
acceptable). input 1 (G1) is negative relative to input 2 (G2), the deflection
• Digital PSG typically uses a combination of AC is upward (negative up polarity).
referential, AC true bipolar, and DC recording. Signals common to both inputs are not amplified
• In referential recording, each electrode is recorded (common mode rejection) (Fig. 2–2). Actually, each of the
in comparison with a common reference electrode. inputs is recorded against the common ground and input 2
Any derivation (combination of differences between is inverted. This allows common signals to cancel each other
electrodes) can be displayed by digital subtraction but differences between input 1 and input 2 to be amplified.
[(C4-Ref ) − (M1-Ref ) = C4-M1] during acquisition or later Use of differential amplifiers permits the recording of very
during review of the study. If the reference electrode is low voltage EEG signals that are superimposed upon larger
faulty, all channels will be affected. DC scalp voltage changes and 60-cycle interference from
• Most digital AC amplifiers record with a wide bandwidth nearby AC power lines. Common mode rejection depends
(“wide open”), for example, a low-frequency filter of 0.03 on the impedance at input 1 and 2 being relatively equal.
and a high-frequency filter of 100 Hz. Each derivation is Otherwise, common signals will produce unequal voltages
then displayed after processing with the desired at the two inputs. Making the intrinsic impedance of the
low- and high-frequency digital filters. The recorded inputs much higher than the impedance of the electrodes
data are not changed by the display filters. This allows minimizes the effect of unequal electrode impedances.
display with different filter settings if desired. However, a poorly conducting electrode (high impedance)
• Digital recording requires appropriate sampling rates will typically result in a large amount of 60-Hz artifact (signal
by the A/D converter depending on the variable being contamination). The ground of each differential AC ampli-
recorded. A suitable high-frequency filter must be fier is connected to the common patient ground (commonly,
used to prevent aliasing distortion. an electrode placed on the forehead). This common ground
• The resolution of the monitor is usually what limits the helps balance the inputs to all the differential amplifiers,
possible resolution of the displayed data rather than thereby improving common mode rejection. The use of
the sampling rate. grounds in EEG recording is discussed at the end of the
chapter.
13
14 Chapter 2    The Technology of Sleep Monitoring

It should be noted that a localized EEG transient (e.g., and abdomen), leg EMG inputs, and electrocardiographic
sharp wave) that is located midway between two electrodes (ECG) inputs.
will produce an equal signal in both sides of the differential In referential recording, multiple electrodes are recorded
AC amplifier that will cancel out (output approximately against a common electrical reference (often a single or two
zero). This cancellation effect will alter the overall EEG signal linked electrodes placed near the vertex). A display of any
amplitude less if electrodes are further apart. Thus, a greater derivation using two referentially recorded electrodes is then
distance between two electrode inputs will increase the obtained by digital subtraction [(electrode A − reference) −
amplitude of the recorded signal (less cancellation). This is (electrode B − reference) = electrode A − electrode B] either
one reason the recommended EEG derivations use contra- during live recording or during review (see Fig. 2–3). The
lateral mastoid references (C4-M1, not C4-M2). digital subtraction for display does NOT change the recorded
data. For example, if the sleep technologist failed to observe
REFERENTIAL AND BIPOLAR RECORDING that the electrode F4 went bad during the recording, the
reviewer can change the viewed frontal derivation to F3-M1
Most digital recording systems use a combintation of refer- or F3-M2 (the recommended alternative) (Fig. 2–4). For this
ential, true bipolar, and DC recording (Table 2–1).2 In true reason, both F3 and F4 are recorded (against the reference
bipolar recording, each amplifier records the difference electrode) even though only F4-M1 may be displayed in the
between two electrodes of interest (A–B, C–D). Before the default montage. Of note, if the reference electrode is faulty,
digital era, paper recording was performed using a selector all referential signals are affected (Fig. 2–5). In Figure 2–5,
panel and dedicated individual differential amplifiers. Using note that the true bipolar channels are not affected by a faulty
this approach, it is possible to change the electrodes (deriva- reference electrode. In most digital PSG systems, the EEG,
tion) that are recorded with a given amplifier (Fig. 2–3). EOG, mastoid, and chin EMG electrodes are recorded refer-
However, changing the derivation once the signal is recorded entially (see Table 2–1). DC recording is used for nasal pres-
(changing from A–B to A–D) is not possible. Today, selector sure, pulse oximetry, and other DC signals such as those
panels are rarely used in digital sleep recording. However,
true bipolar recording is still used for inputs that one would
not desire to change in review—for example, the two inputs TABLE 2–1
of the thermal flow sensor, respiratory effort bands (thorax Types of Recording
Referential EEG: F4, F3, C4, C3, O2, O1, M1, M2
recording EOG: E1, E2, M1, M2
IF G1 IS NEGATIVE TO G2,
DERIVATION G1-G2 THE DEFLECTION IS UPWARD Chin1, Chin2, Chin3
Reference
30 µV
True bipolar ECG, thermal flow, thorax and
Differential amplifier 20 µV
(two inputs abdominal sensors, right and left
each) anterior tibial EMG
G1
DC Nasal pressure, SpO2, positive airway
G2 0 pressure device (flow, leak, pressure),
end-tidal or transcutaneous PCO2
10 µV
Ground ECG = electrocardiography; EEG = electroencephalography;
EMG = electromyography; EOG = electro-oculography; PCO2 = partial
FIGURE 2–1 Differential amplifier. The difference between the two inputs is amplified
pressure of carbon dioxide; SpO2 = pulse oximetry.
(for simplicity, the amplification factor = 1).

DETAILS

DIFFERENTIAL AMPLIFIER

80 µV Common mode rejection Input 1
60 µV
 
Input 1 Output
20 µV
Output 1 Ground 
0 

Ground
Input 2
Input 2 

FIGURE 2–2 Common mode rejection by a differential amplifier (for simplicity, the amplification factor = 1).
 Chapter 2    The Technology of Sleep Monitoring 15

BIPOLAR RECORDING REFERENTIAL RECORDING FIGURE 2–3 The difference between true bipolar
recording and referential recording. In referential
C4 C4-M1 C4
Amplifier 1
recording, each electrode is recorded against a common
C3 C3 reference. Specific derivations are then displayed by
C3-M2 Reference (Ref) digital subtraction (during acquisition or review).
M1 Amplifier 2 M1

M2 M2

Selector box
DISPLAY BIPOLAR VIEWS

C4  M1  (C4  Ref)  (M1  Ref)

C3  M2  (C3  Ref)  (M2  Ref)

F4-M1
F3-M1
C4-M1

O2-M1
E1-M2

E2-M2

Chin EMG

FIGURE 2–4 In the F4-M1, derivation 60-Hz artifact is present. The F4 electrode is at fault because other derivations containing M1 are not affected.
The displayed derivation is then changed to F3-M1, allowing an artifact-free frontal derivation to be viewed. Viewing a derivation using any two of the
referentially recorded electrodes is possible. The recommended derivation when the F4 electrode is faulty is actually F3-M2, but the use of F3-M1 is shown
for illustration. EMG = electromyography.

E1-M2

E2-M2

F4-M1

C4-M1

O2-M1

Chin1-Chin3

ECG
R, L legs

Airflow

Chest

Abdomen
SpO2
A B

FIGURE 2–5 A, The reference electrode is faulty. All referentially recorded electrodes show artifact. The true bipolar channels and DC
channels are not affected. In B, the reference electrode was repaired. SpO2 = pulse oximetry.
16 Chapter 2    The Technology of Sleep Monitoring

1 sec Signal 4 Hz  4/sec TABLE 2–2

Recommended Sampling Rate for Various
Polysomnographic Signals
SAMPLING RATES DESIRABLE (HZ) MINIMAL (HZ)
Sampling rate 8/sec
EEG 500 200
EOG 500 200
8 samples/sec
EMG 500 200
Sampling rate 3/sec
ECG 500 200
Airflow 100 25
Oximetry 25 10
Nasal pressure 100 25
3 samples/sec
Esophageal pressure 100 25
FIGURE 2–6 A signal at 4 Hz is sampled at 8/sec with fair reproduction. However, with
sampling at 3/sec, the signal is distorted and a lower-frequency signal is introduced. Body position 1 1
Snoring 500 200
from the positive airway pressure device (flow, leak, Rib cage/abdominal 100 25
tidal volume, delivered pressure), end-tidal, or partial movements
pressure of carbon dioxide (PCO2) device (end-tidal or ECG = electrocardiography; EEG = electroencephalography;
transcutaneous). EMG = electromyography; EOG = electro-oculography.
From Iber C, Ancoli-Israel S, Chesson A, Quan SF for the American Academy
of Sleep Medicine: The AASM Manual for the Scoring of Sleep and Associated
Sampling Rate Events: Rules, Terminology and Technical Specifications, 1st ed. Westchester,
IL: American Academy of Sleep Medicine, 2007.
Most digital recording systems use analog amplifiers that
produce a continuous signal output. The signal is then
sampled by an analog-to-digital (A/D) conversion board that A/D converter
converts the signal to a digital form that can be stored and Input at
12 bit (122  4096 bits)
manipulated by a computer. The sampling rate must be more head box
than twice the frequencies being recorded to avoid signal 000000000001
5 V (± 2.5 V) 000000000010
distortion (Nyquist theorem).1,2,4 If lower sampling rates are
4000 µV

000000000011
 1250 000000000100
used, the signal can be very distorted and the addition of 000000000101
000000000110
frequencies lower than the original signal sampled may be
introduced (Fig. 2–6). For this reason, signals with a fre-
quency higher than half the sampling rate must be filtered Etc,
out because they can cause aliasing distortion.2,4,5 For Input at
example, if the sampling rate is 200 samples/sec, the ampli- A/D board
fied signal must be processed by a high frequency filter with Resolution 0.97 µV per bit
a cutoff frequency of 100 Hz or lower before being sampled
(A/D converter). The required sampling rate depends on the FIGURE 2–7 Dynamic range and resolution of an analog-to-digital (A/D) converter.
frequency of the signal to be recorded. Slower varying signals
require a lower sampling rate. In Table 2–2, the sampling range. A typical A/D converter might have a dynamic range
rates recommended by the American Academy of Sleep for the amplified signal of 5 V (±2.5 V). Commonly, a set
Medicine (AASM) scoring manual6 are illustrated. Some amplification is applied to all AC signals before A/D conver-
digital PSG systems have the ability to record different signals sion (e.g., a gain of 1250). If one assumes an amplification of
at different sampling rates. Ultimately, the computer program 1250, then the dynamic range (peak to peak) of an A/D
uses only a small portion of the data for the display because converter with an amplified voltage range of 5 V expressed
monitor resolution (in pixels per displayed time duration) is as the unamplified signal would be approximately 4000 µV
usually much less than the sampling rate.5 (4000 µV × 1250 = 5,000,000 µV = 5.0 V). If a 12-bit A/D
A/D conversion is also characterized by the dynamic converter is used, this would result in a resolution of 0.97 µV/
range (the range of voltages accepted by the A/D converter) bit (4000 µV/4096 digital values) (Fig. 2–7).
and the resolution. The dynamic range may be expressed as
the amplified or unamplified signal range. The resolution
Monitor Resolution
depends on the A/D converter as well as the dynamic range.
A 12-bit DC converter produces 212 = 4096 digital values An important limitation on the accuracy of signal recording
(bits) or a 16-bit converter = 65,536 values across the dynamic and display is introduced by the fact that the monitor
 Chapter 2    The Technology of Sleep Monitoring 17

Respiratory event
100%
95%

SpO2 90%
85%
80%
75%

Desaturation
Left
Right
Position Prone
Supine
Sitting

Wake
REM
Stage 1
2
3
Epoch
Time

FIGURE 2–8 An overview of the entire night of the recording shows respiratory events, pulse oximetry (SpO2), desaturation events, body position, and sleep stage
(hypnogram). REM = rapid eye movement.

resolution is usually much lower than the data sampling rate. an all-night condensed view with graphs of the hypnogram
Because the sampling rate used for most digital systems is (representation of sleep stages), arterial oxygen saturation
200 samples/sec or higher, the resolution of the monitor is (SpO2 tracing), continuous positive airway pressure (CPAP)
often the limiting factor in the accuracy of signal display. The levels, respiratory events, and body position (Fig. 2–8). This
minimum monitor resolution recommended by the AASM allows a useful overview of the entire recording. One can
scoring manual is 1600 × 1200. Assuming 1600 pixels hori- usually select a time point (double click) on a given position
zontally, the visual sampling rate for displays of a 30- or in the summary view and be taken to that time point in the
10-second window of data corresponds to a sampling rate of more detailed tracings.
approximately 50 and 150 samples/sec if the entire monitor
display consists of waveforms. Then according to the Nyquist FILTERS (LOW-FREQUENCY, HIGH-
theorem, frequencies of 25 or 75 Hz or greater would be
FREQUENCY, AND NOTCH FILTERS)
prone to aliasing. A monitor-induced aliasing distortion of
data can sometimes be noted if switching from a 30-second Any signal of interest can be contaminated by unwanted
to a 10-second view significantly changes the shape of the low- or high-frequency signals or 50- to 60-Hz artifact (from
activity being visualized.5 nearby AC power lines). Filters allow these components to
be diminished. For example, a low-frequency filter (high-
pass filter) attenuates the amplitude of low-frequency signals.
Time Window for Display
A high-frequency filter (low-pass filter) attenuates the ampli-
During traditional paper-ink recording for sleep, the paper tude of high-frequency signals.1 The amount of signal reduc-
speed was 10 mm/sec, which produced 30-second pages (30- tion due to a given analog or digital filter is given in decibels.
cm-wide paper). A faster speed was used for clinical EEG The amount of signal reduction in decibels (dB) is given by
(30 mm/sec). However, such a fast paper speed would the formula 20 log (voltage-out/voltage-in), where voltage-
produce a very large amount of paper for each sleep study. out and voltage-in are the amplitude of the signal entering
In digital recording, one can choose various time windows and leaving the filter, respectively. A signal reduction of 30%
during either acquisition or review. A 30-second window and 50% (voltage-out/voltage-in ratios of ∼ 0.7 and 0.5,
(equivalent to a paper speed of 10 mm/sec) is used for sleep respectively) corresponds to 3 dB and 6 dB reductions. Dif-
staging and for scoring arousals. Time windows of 60 to 240 ferent filter settings (e.g., 0.3, 1) are named by the “cutoff
seconds may be used to view and score respiratory events frequency,” which is the frequency of the signal that is reduced
and leg movements. Alternatively, viewing data in a by 3 or 6 dB depending on the terminology and the type of
10-second window (equivalent to 30 mm/sec) is the usual filter the manufacturer uses. Therefore, a filter setting of “X
method for clinical EEG recording. This allows better visu- Hz” means that the amplitude of a signal with a frequency
alization of very brief events (sharp waves and spikes) and of X is diminished by 30% or 50% depending on whether the
interictal or epileptiform activity. The 10-second window 3 dB or 6 dB cutoff frequency is used to name the filter.
can also be useful for measuring the frequency of a group of
oscillations or viewing the ECG result. The traditional ECG
Low-Frequency Filter
speed is 25 mm/sec, which is quite close to 30 mm/sec. Some
systems allow split screens with different time windows in A 1-Hz low-frequency filter (3 dB) attenuates a 1-Hz signal
each screen. All digital sleep monitoring systems also provide by 30% (or to 70% of the original signal). Similarly, a 6 dB
18 Chapter 2    The Technology of Sleep Monitoring

filter would attenuate a 1-Hz signal by 50%. Signal strength circuits. In RC circuits, an increase in step voltage produces
of frequencies below 1 Hz would be attenuated even more an abrupt increase in voltage across the resistor, then an
(Figs. 2–9 and 2–10). It is important to realize that frequen- exponential fall in voltage to 1/e (0.37) of the maximum
cies slightly above the low-frequency filter setting of 1 Hz voltage in one time constant (TC). In a simple, low-frequency
will also be attenuated by a 1-Hz low-frequency filter, filter RC circuit, the frequency (fc) at which the output
although to a lesser degree. Figure 2–10 illustrates the effect voltage across the resistor is attenuated to 0.37% of the input
of various low-frequency filters (denoted by their 3 dB cutoff voltage is related to the TC by the formula fc = 1/(2π/TC).
frequency) on low-frequency signals. A range of possible In RC circuits, the TC = RC, where R is the resistance and
low-frequency filter settings (off, 0.01, 0.03, 0.1, 0.3, 1, 3, and C the capacitance of the circuit. Even if digital filters are
10) is commonly provided. used, the relationship between the TC and the 3 dB fre-
Sometimes low-frequency filter settings are specified as a quency is given by Equation 2–1:
time constant rather than as a cutoff frequency (Fig. 2–11).
Traditional analog filters used resistance-capacitance (RC) TC = 1/(2π × filter frequency ) Equation 2–1

FIGURE 2–9 A low-frequency filter setting LOW FILTER (HIGH PASS)

of 1 Hz attenuates a signal of 1 Hz by 50% (6 This assumes a 6 dB filter or filter setting = 1/2 amplitude
dB filter). Signals with lower frequencies are Filter settings Signals 1 to 10 Hz
attenuated more. Signals slightly above 1 Hz 0.1 0.3 1.0 1.0
100 100 are attenuated, but
are also attenuated (but < 50%). Note that
Percent amplitude

Percent amplitude
3 less than 50%
the horizonal axis uses a logarithmic scale and 75 75
Filter setting of 1
the vertical axis is linear. 50 10 50 reduces signal with
frequency of 1 Hz
25 25 by 50%
Filter settings 1.0
0 0 Lower frequency
0.1 0.2 0.5 1 2 5 10 0.1 0.2 0.5 1 2 5 10
attenuated more
Signal frequency Signal frequency than 50%

FIGURE 2–10 Effect of different low-frequency filter (LF)

settings (3 dB filter) given in hertz on slow wave
amplitude. Note that an LF 1 reduces a 1-Hz wave by about  1 Hz
30% (from 100 to 70 µV). An LF 0.3 Hz reduced the 1-Hz
wave only slightly, whereas an LF 3 Hz essentially eliminated
1 Hz activity.
LF off 100

LF 0.3

LF 1 70

LF 3 100 µV

1 second
 Chapter 2    The Technology of Sleep Monitoring 19

FIGURE 2–11 The lower the low-frequency filter

C cutoff frequency (LF), the longer the time constant
Voltage (TC). Here, C is the capacitance and R is the resistance
Vin R 100 µV in a traditional resistance-capacitance (RC) filter. Vin
= input voltage.
LF = 0.1 Hz TC = 1.6

Vin input
voltage LF = 0.3 Hz TC = 0.53

LF = 1 Hz TC = 0.16
Voltage 37%
across R
37% LF = 5 Hz TC = 0.03

TC
1 sec

30 Hz high filter FIGURE 2–12 The effects of a high-frequency filter. A 6 dB 30-Hz filter
0.3 Hz low filter Frequency slightly attenuates a 30-Hz signal by 50%. Signals less than 30 Hz are attenuated
lower than 30 Hz less (low pass). Signals with a frequency higher than 30 Hz are attenuated
100 attenuated slightly much more. The gray shows a frequency range that is attenuated less
Percent amplitude

75 than 0.70 by the combination of a 0.3-Hz low-frequency filter and a 30-Hz

30 Hz signal attenuated
high-frequency filter (HF). This is often referred to as the bandwidth
50 by 50% by a 6 dB 30 Hz HF
(or bandpass of the filters). Note that the horizontal axis of the plot
25 uses a logarithmic scale and the vertical axis is linear.
Frequency
0 higher than
0.1 0.2 0.5 1 2 5 10 50 100 200 30 Hz attenuated
0.3 30 even more
Signal frequency (Hz)

For example, a 0.3-Hz low-frequency filter has a TC of Using the combination of a low-frequency and a high-
approximately 0.53 second. Of note, the actual TC after a frequency filter, a range of frequencies is amplified. Alterna-
step increase in voltage may vary depending on the high- tively, if digital filters are applied to raw digital data, a range
frequency filter setting as well. The lower the cutoff fre- of frequencies is displayed. The range of signal displayed or
quency, the longer the time constant (see Fig. 2–11). If amplified is called the bandwidth.
amplifiers are calibrated by step (square wave) voltage
change, the actual TC can be noted from the time it takes 60-Hz or Notch Filters
for the deflection to return to 0.37 of the maximum Most amplifiers (digital PSG systems) provide optional notch
deflection. filters to significantly attenuate a narrow range of frequency
associated with power line signal contamination (e.g., 50 or
60 Hz). The notch filter can be added or removed. If the
notch filter is turned on, it is applied to the signal in addition
High-Frequency Filters
to the low-frequency and high-frequency filters. The routine
A 35-Hz high-frequency filter attenuates a signal of 35 Hz use of a notch filter is usually not recommended. The sudden
by 50% (6 dB filter), and frequencies above 35 Hz would appearance of increased 60-Hz activity in a derivation is a
be attenuated more. In addition, frequencies slightly below clue that one or more electrodes is faulty. However, as previ-
the high-frequency filter setting will also be slightly attenu- ously mentioned, use of a high-frequency filter of 35 Hz
ated. Figure 2–12 illustrates the effects of a 6 dB 30 Hz (commonly used for EEG and EOG derivations) already
filter. A range of high-frequency filter settings is typically substantially attenuates a 60-Hz signal (much the same as
provided (off, 3, 15, 35, 70, and 100 Hz). Note that using a turning on the 60-Hz filter)
30-Hz high-frequency filter (see Fig. 2–12) significantly Turning on and off the 60-Hz (notch) filter can be useful
attenuates 60-Hz signals. Therefore, the addition of a 60-Hz in determining the degree of signal contamination by 60-Hz
notch filter adds little if a 30- to 35-Hz filter is already interference. If turning off the notch filter dramatically
being used. increases signal amplitude, this suggests considerable 60-Hz
20 Chapter 2    The Technology of Sleep Monitoring

FIGURE 2–13 If signal amplitude

increases significantly after the 60-Hz filter 60-Hz notch filter 60-Hz notch filter 60-Hz notch filter
is turned off, this is evidence that significant 100 V ON OFF OFF
HF 100 Hz HF 100 Hz HF 35 Hz
60-Hz activity is contaminating the signal.
Turning on and off the 60-Hz filter will have
less effect when the high-frequency filter
setting is 35 Hz because much of the 60-Hz
activity has already been filtered out. Chin1-Chin2

signal contamination (Fig. 2–13). One would expect switch- TABLE 2–3
ing on and off the 60-Hz filter to have more effect if a high- Recommended Filter Settings
frequency filter of 100 Hz was used (commonly used for
LOW FREQUENCY HIGH FREQUENCY
EMG derivations) compared with a high-frequency filter of
35 Hz (EEG and EOG derivations). For this reason, 60-Hz EEG 0.3 Hz 35 Hz
contamination is most frequently visualized in the chin and EOG 0.3 Hz 35 Hz
leg EMG derivations even if the 60-Hz filter is turned on.
Artifacts including 60-Hz artifact are discussed in more EMG 10 Hz 100 Hz
detail in Chapter 4. ECG 0.3 Hz 70 Hz
Respiration 0.1 Hz 15 Hz
AMPLIFIER FILTER SETTINGS FOR Snoring 10 Hz 100 Hz
DIGITAL SLEEP RECORDING ECG = electrocardiography; EEG = electroencephalography;
EMG = electromyography; EOG = electro-oculography.
Sleep recording with traditional dedicated bipolar AC ampli- From Iber C, Ancoli-Israel S, Chesson A, Quan SF for the American Academy
fiers used RC circuits (RC filters) as filters and the recorded of Sleep Medicine: The AASM Manual for the Scoring of Sleep and Associated
(on paper or computer) signal was filtered at the current Events: Rules, Terminology and Technical Specifications, 1st ed. Westchester,
IL: American Academy of Sleep Medicine, 2007.
amplif ier settings. Today, most amplifiers used for digital
recording (referential and true bipolar) record signals “wide
open,” that is, with default low-frequency filter (0.03–0.1)
and a high-frequency filter setting usually at or less than half
Clinical Example of the Effects of Filter Settings
the sampling rate (e.g., 100-Hz for a sampling rate of 200/
sec). Thus, “raw” signals are actually recorded over a wide As discussed in Chapter 7, monitoring nasal pressure pro-
frequency range or bandwidth (between default low and vides a more accurate estimate of airflow than thermal
high frequencies) but are viewed (displayed) after applica- sensors. During upper airway narrowing, the nasal pressure
tion of selected digital low-frequency and high-frequency signal shows a flattening (flow plateau) during inspiration.
filters. The digital filters alter the displayed signal but NOT Some sleep centers record nasal pressure with an AC ampli-
the recorded data. This allows multiple choices of filters if fier instead of acquiring the signal in the DC mode. However,
desired by the technologist or reviewer. The filter settings a low-frequency filter setting of 0.03 or less (or a long TC) is
recommended by the AASM scoring manual6,7 are shown in ideal to allow demonstration of a flow plateau in the nasal
Table 2–3. The filter settings are selected to include the fre- pressure signal (Fig. 2–14). To accurately record or display a
quencies of interest in sleep monitoring. For example, to very slowly varying signal, a sufficiently low cutoff frequency
detect slow waves and eye movements but avoid the effect of must be used for the low filter.
scalp DC voltage changes (very low frequency), a low- If the nasal pressure signal is unfiltered, vibration during
frequency filter of 0.3 Hz is selected. Setting the low- snoring is often visible. However, the ability to see snoring
frequency filter of the EEG or EOG channels higher would (high-frequency vibration) depends on the high-frequency
reduce slow wave and eye movement amplitude. For EMG filter settings. Use of a fairly low high-frequency filter setting
and ECG channels, a low-frequency filter of 10-Hz is used, will reduce high-frequency signals such as noted in the nasal
because the relevant activity is of a much higher frequency. pressure tracing during snoring (Fig. 2–15). Ideally, one
For EEG and EOG monitoring, selection of a 35-Hz high- would use a high-frequency filter setting of 70 to 100 Hz.
frequency filter removes unwanted higher frequencies but
attenuates the characteristic EEG patterns such as sleep spin-
DIGITIAL PSG SYSTEM OVERVIEW
dles (11–16 Hz) to a lesser degree. In contrast, the EMG
frequences of interest are much higher and a high-frequency The typical digital PSG system includes a headbox in which
filter of 100 Hz is usually selected. individual electrodes are attached to an amplifier. An
 Chapter 2    The Technology of Sleep Monitoring 21

FIGURE 2–14 Effects of different low-frequency filter settings

Low filter on the nasal pressure signal. To visualize the plateau in the nasal
settings: pressure signal, either a DC recording or an AC recording with a
DC very small low-frequency filter setting is needed (either 0.01 or
0.03 is ideal).
0.01

0.03

0.1

0.3

Nasal pressure high filter 3 Hz

Nasal
pressure

Snore

Nasal pressure high filter off

Nasal
pressure

Snore

Snoring as vibration in nasal pressure signal

Effect of high (low pass) filter

FIGURE 2–15 Effect of a high-frequency filter of 3 Hz on a high-frequency signal (snoring).

accessory box for DC channel inputs or dedicated input jacks derivations and processes the data with the selected digital
on the amplifier are also usually available. The amplifier is low-frequency and high-frequency filters. A display sensitiv-
then connected to the A/D converter. Today, the A/D con- ity is also chosen to determine the upper and lower limits
verter is often contained within the amplifier that sits at the of data to be displayed in the channel width (digital gain).
patient’s bedside. The digitized signal can then be sent over The changes in the display (specific derivations, digital
ethernet cables to the computer or sent in the wireless mode filters, digital gain) do not change the raw data that are
to a computer, which then records the digital data. This recorded by the computer. The entire process is summarized
arrangement avoids the difficulties that occur when an in Figure 2–16.
analog signal is sent over a long distance (60 Hz contamina-
tion or loss of signal strength). A schematic of a typical
Montages for Digital Recording
system is shown in Figure 2–16. A typical PSG amplifier
often has a fixed gain and default low- and high-frequency Digital systems allow the user to specify a number of user-
filter settings (e.g., 0.1 and 100 Hz). AC signals are recorded defined display montages with the ability to select the
over a wide frequency range (bandwidth). The A/D con- number of channels (traces) to be displayed, the derivations
verter samples the signal and raw digital data are stored in for each channel, the order in which the desired derivations
the computer. After the raw data are digitized and stored, are displayed (the inputs for each channel), as well as the
extensive manipulation is possible to produce the desired sensitivity (gain), low- and high-frequency filter settings,
signal display. During acquisition and review, the computer notch filter on or off, and the color of each tracing. A sample
program performs digital subtraction to display the desired montage (Table 2–4) is displayed in Figure 2–17. Typically,
22 Chapter 2    The Technology of Sleep Monitoring

Amplifier
Computer
Fixed
program Data
gain Raw
Electrode A/D display
LF 0.1 data Digital sensitivity
box converter
HF 100 stored Digital filters C4-M1
C4 -Ref (LF, HF)
M1 -Ref

FIGURE 2–16 Schematic of a digital polysomnography (PSG) system. In many systems, the A/D converter is within the same
unit housing the amplifier(s). Modern digital PSG systems usually perform a fixed amplification with default low (LF) and high
(HF) filters permitting a wide frequency range to be amplified. The digitized data are stored on media (hard drive in the
computer). The PSG software then scales the raw data, applies the selected low-frequency and high-frequency filters, and
provides a display (either during acquisition or at review).

TABLE 2–4
Montage 1: Diagnostic Adult
CHANNEL INPUT SENSITIVITY (P-P) LF HF NOTCH
(TRACING) TYPE INPUT 1 INPUT 2 µV UNLESS DC (HZ) (HZ) FILTER
1 Ref F4 M1 150 0.3 35 Off
2 Ref C4 M1 150 0.3 35 Off
3 Ref O2 M1 150 0.3 35 Off
4 Ref F3 M2 150 0.3 35 Off
5 Ref C3 M2 150 0.3 35 Off
6 Ref O1 M2 150 0.3 35 Off
7 Ref E1 M2 150 0.3 35 Off
8 Ref E2 M2 150 0.3 35 Off
9 Ref Chin1 Chin2 150 10 100 Off
10 BP ECG1 ECG2 1500 0.3 70 Off
11 DC Nasal pressure N/A −1 to +1 V †
DC 100 Off
*
12 BP NOTF—input 1 Input 2 750 0.1 15 Off
*
13 BP Snore—input 1 Input 2 750 10 100 Off
*
14 BP Thorax—input 2 Input 2 1500 0.1 15 Off
*
15 BP Abdomen—input 1 Input 2 1500 0.1 15 Off
16 DC SpO2 DC 0–1 V DC N/A Off
17 BP RAT—input 1 Input 2 150 10 100 Off
18 BP LAT—input 1 Input 2 150 10 100 Off
*The sensitivity settings for bipolar channels depend on the output range for a particular device.
†
Varies with transducer type.
BP = dedicated bipolar inputs (input-1, input-2); ECG = electrocardiography; HF = high frequency filter setting; LAT = left anterior tibial; LF = low frequency filter
setting; N/A = not applicable; NOTF = nasal-oral thermal flow sensor; P-P = peak to peak; RAT = right anterior tibial; Ref = referential input; SpO2 = pulse oximetry.

one montage is adapted for a diagnostic study and another gain), low-frequency and high-frequency filters, channel
for a positive-pressure titration. During review or acquisi- width, and inversion of signal. Default settings for each
tion, each individual channel may be altered if so desired or channel can be specified, so they do not have to be individu-
an entirely different montage may be displayed. ally set for each recording. Figure 2–17 illustrates typical
channel controls. Recall that changes in channel settings do
not change the recorded (and digitally stored) data.
Channel Settings/Montages
In sleep recording using paper, the EEG was usually
Each channel (tracing) display can be changed by the viewer recorded at a sensitivity of 50 µV/cm in adults. In children,
with respect to the inputs, sensitivity (sometimes called a lower sensitivity (100 µV/cm) was used because of the very
 Chapter 2    The Technology of Sleep Monitoring 23

high amplitude EEG activity. The term “gain” rather than ultimate size of the channel width depends on the way the
sensitivity was also used. However, this implies an amplifica- computer program scales the signal for display. Some pro-
tion of signal. In digital recording, amplification actually grams have an option to allow signals either to overlap or to
occurs before the signal is digitized. The size of the display be cropped if they exceed the given channel width.
of a given signal is varied by the computer program that
scales the display based on the available channel width and Impedance Checking and Referential
the voltage limits or sensitivity. For example, if a channel
Display View
width of 100 pixels represents 100 µV peak to peak, a signal
of 50-µV peak to peak would vary between the 25th and the Traditionally, after electrodes were applied to the patient’s
75th pixel. The default digital displays for EEG often use 100 head, the impedance of each electrode was checked by plug-
or 150 µV peak to peak per channel width (200 for children). ging the electrodes into an impedance box that allowed com-
Figure 2–18 shows two methods of adjusting the display parison of any electrode referred to the ground electrode or
(gain/voltage per division or peak-to-peak sensitivity). The a combination of all the other electrodes. Most digital
systems can measure impedance on line using a signal from
the amplifier. The values can then be stored with other digital
Channel 5 data for later review. The AASM scoring manual recom-
mends a maximum electrode impedance of 5 KΩ (<10 KΩ
Input 1 Input 2 Low freq filter High freq filter
is acceptable). Another useful method of looking at the
C4 M2 0.3 Hz 35 Hz
quality of each individual electrode is to display all of
the unfiltered referentially recorded electrode against the
Notch filter On Off common reference (rather than the digital subtraction of two
Trace width
100 Sensitivity referentially recorded electrodes). Figure 2–19 displays a ref-
erential view with all high-frequency filters set to 100 Hz.
100 µV p-p Electrode impedance is also displayed. One can tell that F4,
Invert
Chin2, and Chin3 electrodes are faulty and should be
FIGURE 2–17 Example of typical controls for each display channel (tracing). Controls changed or fixed. As previously noted, if all tracings on the
allow selection of the derivation, low-frequency and high-frequency filters, notch filter referential view are bad, this suggests a problem with the
(on or off), and sensitivity. On most digital PSG systems, channel width and trace color reference electrode. However, a faulty reference electrode
can also be selected. p-p = peak to peak. does not affect the true bipolar channels (see Fig. 2–5).

1 second

One division

100 µV/division
50 µV/division

Channel width

100 µV peak to peak 200 µV peak to peak

FIGURE 2–18 Two methods of adjusting the sensitivity (digital gain) are shown. A 100-Hz peak-to-peak signal of 10 Hz is shown. The top panels
specify a voltage per division value. A larger channel width is needed to display a given signal if the actual division size remains constant when voltage per
division decreases. The bottom panel illustrates a method by which the peak-to-peak voltage of the entire channel width is specified. The actual
channel width will depend on the way the computer program scales the display.
24 Chapter 2    The Technology of Sleep Monitoring

Impedance

O1-REF 6.4 KΩ
O2-REF 6.3
C3-REF 5.6
C4-REF 2.5
E1-REF 3.5
E2-REF 4.4
M1-REF 4.8
M2-REF 4.3
Chin1-REF 3.8
Chin2-REF 19.2
Chin3-REF 10.3
F3-REF 6.7

F4-REF 30.2

FIGURE 2–19 A “referential display” with each electrode displayed against the common reference electrode. The electrode impedance is also displayed. One
can see that the Chin2, Chin3, and F4 electrodes should be replaced or repaired.

FIGURE 2–20 The three types of grounds AMPLIFIER SYSTEM

used in a modern PSG amplifier are illustrated
in a simplified schematic. The patient ground Isolated Non-isolated
section section
is separated from the earth ground by an
isolation device or isolated section of the Differential
amplifiers
amplifier. (Courtesy of Marc Paliotta, Grass F4
Technologies.) H = hot; N = neutral. Data
 Isolated power
C4 DC
 supply
Isolation barrier

Reference

Patient
ground Circuit
ground Power
Chassis
ground Earth ground NH

Video-Audio PSG
sec). Simultaneous audio is also usually available and this is
Today, most digital systems allow for the simultaneous very useful for documenting teeth grinding (bruxism),
recording of video and audio signals. Ideally, the video talking during parasomnias, snoring, and other behaviors
should be synchronized with the recorded EEG and other during the recording.
signals. This will allow the reviewer to see patient movement
corresponding exactly to a given time point in the recorded
Grounds
PSG signals. For example, one could note facial twitching
during a particular EEG pattern. Video PSG is an important The terminology is confusing with three different grounds
development and allows the reviewer to confirm the patient being used in modern PSG recording (Fig. 2–20). These
position as well as document unusual behavior (e.g., para- include
somnias) during the night. Video files are often quite large
and are usually compressed (e.g., MPEG4). The size of the 1. The patient ground (iso-ground input on the electrode
file will depend on the quality of the video (10 or 25 frames/ box). This neutral electrode is usually connected to the
 Chapter 2    The Technology of Sleep Monitoring 25

forehead. It is used to balance the inputs of all the differ- C. Decreases the amplitude of a 0.1-Hz signal more than
ential amplifiers (essential for common mode rejection). a 0.3-Hz signal.
2. Chassis ground (container ground). Because a metal D. A and B.
chassis is rarely used today, this would be the amplifier E. A and C.
circuit ground (or the ground of the nonisolated portion
of the amplifier). 6. If a sampling rate of 400 samples/sec is used, what is the
3. Earth ground. In the three-wire power line (three-prong highest frequency cutoff for the high filter that can be
plug) AC input, the three wires are designated “hot (H),” used and still avoid significant aliasing distortion?
“neutral (N),” and “earth” ground. Most amplifiers use an A. 400 Hz.
isolated medical-grade power supply that outputs low- B. 200 Hz.
level DC voltage to power the amplifiers. C. 100 Hz.
In modern systems, the patient ground is never directly D. 50 Hz.
connected to the earth ground. A current-limiting device
7. Signals X, Y, and Z are recorded against a reference
or isolation device is always placed between the patient
(referential) and W1 and W2 are acquired by bipolar
ground and the earth ground. A common method is to use
recording (W1-W2) using digital PSG. Which of the
optical isolation in which the signal is transmitted by light
following is NOT true?
within a small element of the circuit. Figure 2–20 illustrates
one method in which part of the amplifier is isolated from A. The derivation X-Y can be displayed.
the chassis and earth grounds. B. The derivation W1-X can be displayed.
C. If all derivations containing X, Y, and Z show artifact,
CLINICAL REVIEW QUESTIONS the reference electrode is probably faulty.
1. What (low, high) filter settings are recommended for D. The filter settings of the displayed derivation W1-W2
PSG recording (display) of EEG and EOG derivations? can be changed.
A. 0.5, 70 Hz. 8. What is the minimum recommended sampling rate to
B. 0.3, 35 Hz. record the oximetry signal?
C. 0.5, 70 Hz. A. 50 samples/sec
D. 0.5, 35 Hz. B. 25 samples/sec.
C. 10 samples/sec.
2. What are the recommended (low, high) filter settings for
D. 100 samples/sec.
display of EMG?
A. 1 Hz, 70 Hz. 9. What is the recommended electrode impedance less
B. 0.3 Hz, 35 Hz. than?
C. 10 Hz, 100 Hz. A. 10 KΩ.
D. 10 Hz, 70 Hz. B. 5 KΩ.
C. 20 KΩ.
3. What is the minimal recommended sampling rate for D. 1 KΩ.
recording EEG, EOG, EMG, and ECG signals?
A. 100 samples/sec. 10. What are the recommended (low, high) filter settings for
display of the ECG?
B. 200 samples/sec.
A. 10, 100 Hz.
C. 400 samples/sec.
B. 0.3, 35 Hz.
D. 500 samples/sec.
C. 10, 70 Hz.
4. What is the minimal recommended sampling rate for D. 0.3, 70 Hz.
airflow, rib cage and abdominal movements, and NP?
11. Using a high filter of 35 Hz for the EEG and EOG display
A. 10 samples/sec. (recording) reduces 60-Hz activity in the displayed
B. 25 samples/sec. signal.
C. 100 samples/sec. A. True
D. 200 samples/sec. B. False
5. Which of the following is true about the effect of a low 12. What are the recommended (low, high) filter settings for
filter setting of 0.3 Hz? display of the thermal airflow or chest and abdominal
A. Has minimal effect on a 10-Hz signal. RIP signals?
B. Does not affect the amplitude of slow wave activity A. 0.3, 35 Hz.
(0.5–2 Hz). B. 0.1, 15 Hz.
26 Chapter 2    The Technology of Sleep Monitoring

C. 0.3, 15 Hz. 8. C. (10 samples/sec is the minimum recommended rate

D. 0.1, 35 Hz. but 25 Hz is desirable).

9. B. 5 KΩ is recommended, <10 KΩ or less is acceptable.

Answers
10. D. Although a high filter of 70 Hz is the official recom-
1. B. mendation, using a high filter setting of 100 Hz may
allow better visualization of pacer spikes.
2. C.
11. True. A high filter with a cutoff setting of X reduces the
3. B. activity of frequency of signals higher than X. When
using a high-frequency filter of 35 Hz, turning on and off
4. B.
the 60-Hz notch filter has less effect on the viewed signal.
5. E. A low filter of X Hz has some effect on signals with
12. B. Using a low filter of 0.1 instead of 0.3 allows more
a slightly higher frequency than X. Therefore, a 0.3-Hz
accurate visualization of a slowly varying signal.
filter does have some effect on a frequency range of 0.5
to 2 Hz. The lower a frequency signal is compared with
X Hz, the more the decrement in amplitude. A 0.3-Hz REFERENCES
filter decreases the amplitude of a 0.1-Hz signal more 1. Tyner FS, Knott JR, Brem Mayer W: Fundamentals of EEG Tech-
than a 0.3-Hz signal. nology. New York: Raven, 1983.
2. Fisch BJ: Spehlman’s EEG Primer. New York: Elsevier, 1991, pp.
6. B. 200 Hz. The Nyquist theorem states that the 39–65.
3. Berry RB: Sleep Medicine Pearls, 2nd ed. Philadelphia: Hanley
minimum sampling frequency to record a signal of X Hz & Belfus, 2003, pp. 67–69.
is 2 X. Conversely, all frequencies greater than X are 4. Epstein C: Digital EEG. Trouble in paradise? J Clin Neurophysiol
undersampled if a sampling rate of 2 X is used and if 2006;23:190–193.
sampled can result in signal distortion. For this reason, 5. Epstein CM: Aliasing in the visual EEG: a potential pitfall of video
a high filter of X must be applied to the signal to dimin- display technology. Clin Neurophysiol 2003;114:1974–1976.
6. Iber C, Ancoli-Israel S, Chesson A, Quan SF for the American
ish those frequencies before the total signal reaches the Academy of Sleep Medicine: The AASM Manual for the Scoring
A/D converter and is sampled. of Sleep and Associated Events: Rules, Terminology and Techni-
cal Specifications, 1st ed. Westchester, IL: American Academy of
7. B. True bipolar recording does not allow one part of the Sleep Medicine, 2007.
derivation (W1, W2) to be displayed against another 7. Silber MH, Ancoli-Israel S, Bonnet MH, et al: The visual scoring
electrode. of sleep in adults. J Clin Sleep Med 2007;15:121–131.
 Chapter 3

Sleep Staging in Adults

The AASM scoring manual uses new nomenclature (Table

Chapter Points 3–1) for the sleep stages, uses frontal and occipital as well as
• The AASM scoring manual uses new nomenclature for central EEG derivations, does not use the term “movement
wake and sleep (stages W, N1, N2, N3, and R). Stage N3 time,” and has no 3-minute rule for stage N2 sleep. In addi-
replaces stages 3 and 4 and stage R replaces stage tion, stages 3 and 4 are combined into stage N3. The succeed-
REM. ing discussion follows the new rules with some minor
• Sleep is scored in sequential 30-second epochs. If adaptations for brevity and clarity. Answers to questions
more than one sleep stage occurs in an epoch, the posed to the AASM Scoring Manual Steering Committee
epoch is scored based on the sleep stage occupying and clarifications of the staging rules are posted in a fre-
the majority of the epoch. quently asked questions (FAQs) document on the internet
• The 3-minute rule for stage N2 is no longer used. (http://www.aasmnet.org/Resources/PDF/FAQsScoring
• The scoring of stage N3 is based on slow wave activity Manual.pdf). The definitions of the EEG and eye movement
in the frontal derivation. patterns used for scoring are discussed in more detail in
• If subjects do not generate alpha rhythm on eye Chapter 1.
closure, the onset of stage N1 is based on the earliest
occurrence of slow eye movements, EEG activity in the
4 to 7 Hz range with slowing of the background SCORING BY EPOCHS
frequency by ≥ 1 Hz compared to wake, or vertex The AASM scoring manual continues the convention of
sharp waves. staging sleep in sequential 30-second epochs. Each epoch is
• The AASM scoring manual provides specific rules for assigned a sleep stage. If two or more stages coexist during
scoring intervening epochs between epochs of a single epoch, the epoch is assigned the stage comprising
definite stage N2 and definite stage R. The scoring the greatest portion of the epoch.
manual does not provide rules for transitions between
stage W and stage R or between stage N1 and stage R.
Based on FAQ V7, begin scoring stage R when an Stage W (Wake)
epoch of definite stage R is present (contains both low During wakefulness, patients make the transition from full
EMG tone and REMs). alertness to the early stages of drowsiness. During eyes open
stage W (Fig. 3–1), the EEG consists of low-amplitude activ-
ity (chiefly beta and alpha frequencies) without the rhyth-
micity of alpha rhythm (8–13 Hz most prominent over
occipital derivations). Often, muscle artifact (high-frequency
From 1968 to 2007, sleep was staged according to the activity) is also present in the EEG. Rapid eye movements
manual by Rechtschaffen and Kales (R&K).1,2 In the R&K (REMs) and eye blinks (vertical movements 0.5–2 Hz) may
manual, only central derivations were used to stage sleep, occur. The submental (chin) electromyogram (EMG) is
the term “movement time” was utilized to characterize usually relatively increased compared with that during sleep.
epochs in which the electroencephalographic (EEG) and The majority of individuals with eyes-closed stage W will
eye movement tracings are obscured by patient movement, demonstrate alpha rhythm most prominent in the occipital
and there was a 3-minute rule for the continuation of stage area. Slow eye movements (SEMs) may also be present and
2 (now known as stage N2). The AASM Manual for the the chin EMG activity is relatively high.
Scoring of Sleep and Associated Events3,4 was published in The rules for scoring stage W are listed in Table 3–2. In
2007 (subsequently referred to as the “AASM scoring subjects who generate alpha rhythm, stage W is scored when
manual”). This manual changed the rules of staging sleep more than 50% of the epoch contains alpha rhythm over the
and made recommendations about the methods used to occipital region (see Table 3–2, rule A). SEMs may or may
monitor sleep. not be present during periods when alpha rhythm is present.
27
28 Chapter 3    Sleep Staging in Adults

TABLE 3–1 TABLE 3–2

Comparison of R&K and the American Academy of Stage W Rules
Sleep Manual A. Score epochs as stage W when more than 50% of the
R&K* AASM SCORING MANUAL† epoch has alpha rhythm over the occipital region.
1. EOG: Slow eye movements are characteristic of
Stage W Stage W eyes-closed stage W but are not required criteria for
Stages 1, 2, REM Stages N1, N2, R scoring stage W.
2. Chin EMG: The chin EMG amplitude is variable but is
Stages 3, 4 Stage N3
often higher than during sleep.
Central EEG derivations Frontal, central, and
B. Score epochs without visually discernible alpha
occipital derivations
rhythm (or portions of an epoch without alpha
3-minute rule for No time limit on stage N2 rhythm*) as stage W if any of the following are present:
continuation of stage 2 continuation 1. Eye blinks are present at a frequency of 0.5–2 Hz.
after an epoch with K 2. Reading eye movements are present.
complexes or sleep 3. Irregular conjugate REMs are present associated with
spindles normal or high chin muscle tone (in contrast, stage R
has low chin activity).
Movement arousal based on Arousal based on EEG (and
EMG—an increase in the chin EMG for stage R) *Adaptation of AASM scoring manual stage W, rule B.
Notes:
EMG of any channel 1. Score stage W when the majority of an epoch has either alpha rhythm or
accompanied by a change eye movements consistent with wake as defined in B.
in pattern of any 2. Eye movement patterns (see Chapter 1).
additional channel A. Reading eye movements: trains of conjugate eye movements
consisting of a slow phase followed by a rapid phase in the opposite
Movement time when EEG No movement time direction as the subject reads.
and EOG obscured for Major body movement rules B. Blinks: conjugate vertical eye movements present in wakefulness with
eyes open or closed.
more than half the epoch C. REMs: conjugate irregular, sharply peaked eye movements with an
*Rechtschaffen A, Kales A (eds): A Manual of Standardized Terminology initial deflection lasting < 500 msec.
Techniques and Scoring System for Sleep Stages of Human Sleep. Los EMG = electromyogram; EOG = electro-oculogram; REMs = rapid eye
Angeles: Brain Information Service/Brain Research Institute, UCLA, 1968. movements.
†
Iber C, Ancoli-Israel S, Chesson A, Quan SF for the American Academy of Adapted from Iber C, Ancoli-Israel S, Chesson A, Quan SF for the American
Sleep Medicine: The AASM Manual for the Scoring of Sleep and Associated Academy of Sleep Medicine: The AASM Manual for the Scoring of Sleep and
Events: Rules, Terminology and Technical Specifications, 1st ed. Westchester, Associated Events: Rules, Terminology and Technical Specifications, 1st ed.
IL: American Academy of Sleep Medicine, 2007. Westchester, IL: American Academy of Sleep Medicine, 2007.
EEG = electroencephalography; EMG = electromyography; EOG = electro-
oculography; REM = rapid eye movement.

100 µV F -M
4 1

1 sec C4-M1

O2-M1
E1-M2

E2-M2

Chin EMG

Closed eyes Open eyes

FIGURE 3–1 Stage W. A 30-second epoch with eyes closed and then eyes open. More than 50% of the epoch has occipital alpha activity. EMG = electromyogram.

Epochs without visually discernible alpha rhythm (see Table chin EMG tone for REMs is because stage R (REM sleep) is
3–2, rule B) are scored as stage W if any of the following are characterized by REMs and low muscle tone.
present: Eye blinks of a frequency of 0.5 to 2 Hz, reading eye How should one score epochs that contain both portions
movements, or irregular conjugate REMs with normal or with alpha rhythm (but < 50% of the epoch) AND portions
high chin muscle tone. These eye movement patterns are with eye movements consistent with wake? The AASM
characteristic of stage W. The requirement of normal or high scoring manual did not specifically address this situation.
 Chapter 3    Sleep Staging in Adults 29

However, one can modify stage W rule B to apply to the TABLE 3–3
portions of the epoch without alpha rhythm that contain eye Stage N1 Rules
movements consistent with wake (see Table 3–2, rule B). If
A. In subjects who generate alpha rhythm with eye
the portions of the epoch containing alpha rhythm AND the closure, score stage N1 if
portions of the epoch considered to be wake due to eye 1. EEG: Alpha rhythm is attenuated and replaced by
movements add up to more than 15 seconds (majority of the low-amplitude mixed-frequency (4–7 Hz) activity for
epoch), then the epoch is scored as stage W. more than 50% of the epoch (<50% of the epoch has
Of note, approximately 10% of subjects do not generate alpha rhythm).
alpha rhythm on eye closure and a further 10% may generate a. EEG: Vertex sharp waves may be present but are
limited alpha rhythm. In these subjects, the occipital EEG not required for scoring stage N1.
activity is similar during eye opening and eye closure. When 2. EOG: Slow eye movements may be present in N1, but
alpha rhythm is not generated with eye closure, the rules for these are not required for scoring N1.
3. Chin EMG: Variable amplitude, often lower than wake.
scoring stage W and stage N1 are somewhat different and
sleep onset is more difficult to define (Tables 3–3 and 3–4). B. In subjects who do NOT generate alpha rhythm with eye
In patients who do not generate alpha rhythm, epochs satis- closure, score stage N1 commencing with the earliest of
fying rule B in Table 3–2 are scored as stage W. Otherwise, any of the following phenomena:
epochs are scored as stage W if they do NOT meet criteria 1. The EEG shows 4- to 7-Hz activity with slowing of
for stages N1, N2, N3, or R. In contrast to patients generating background frequencies by 1 Hz or greater from those
of stage W.
alpha rhythm, the presence of SEMs is a criterion for scoring
2. Vertex sharp waves.
stage N1 in subjects who do not generate alpha rhythm (see 3. Slow eye movements.
Table 3–4).
Note: Because slow eye movements often commence before attenuation of
In Figure 3–1, a 30-second tracing shows the transition alpha rhythm, sleep latency may be slightly shorter for some individuals who
from eyes-closed stage W to eyes-open stage W. Slightly do not generate alpha rhythm than for those who do.
more than 50% of the epoch contains alpha activity. Alpha EEG = electroencephalogram; EMG = electromyogram;
EOG = electro-oculogram.
activity is attenuated with eye opening and REMs are noted. Adapted from Iber C, Ancoli-Israel S, Chesson A, Quan SF for the American
In Figure 3–2, portions of the epoch contain alpha rhythm Academy of Sleep Medicine: The AASM Manual for the Scoring of Sleep and
and other portions are considered stage W owing to the Associated Events: Rules, Terminology and Technical Specifications, 1st ed.
Westchester, IL: American Academy of Sleep Medicine, 2007.
presence of eye movements consistent with wakefulness

TABLE 3–4
Scoring Stage W and N1
EEG* EOG CHIN EMG
Alpha rhythm Stage W >50% of the epoch with alpha activity SEMs may be present Variable
on eye Eyes closed
closure Stage W Low-amplitude beta and alpha frequencies REMs Normal or high
Eyes open
Blinks Variable
Reading eye movements
Stage N1 >50% of epoch with alpha attenuation SEMs may be present Variable
and replacement with low-amplitude
mixed-frequency EEG
No alpha Stage W Low-amplitude beta and alpha frequencies REMs Normal or high
rhythm on
Eye blinks Variable
eye closure
Reading eye movements
SEMs absent
Stage N1 Vertex sharp wave Variable Variable
Slowing of 4 to 7 Hz, slowing of Variable Variable
frequency ≥ 1 Hz, compared to stage W
Low-amplitude beta and alpha frequencies SEMs appear Variable
Note: Bold text denotes essential features.
*The EEG is assumed not to contain sleep spindles or K complexes not associated with arousal.
EEG = electroencephalogram; EMG = electromyogram; EOG = electro-oculogram; REMs = rapid eye movements; SEMs = slow eye movements.
30 Chapter 3    Sleep Staging in Adults

100 µV F -M
4 1

C4-M1
O2-M1
E1-M2
E2-M2

Chin EMG

FIGURE 3–2 Stage W. A 30-second epoch containing rapid eye movements (REMs), high chin electromyographic (EMG) activity, and some alpha activity. The majority of the
epoch contains either alpha activity or eye movements consistent with wakefulness.

100 µV
F4-M1
1 sec

C4-M1

O2-M1

E1-M2

E2-M2

Chin EMG

FIGURE 3–3 Stage W. A 30-second epoch is shown containing REMs, blinks, relatively high chin electromyographic (EMG) activity, and the absence of discernible alpha activity.

100 µV
F4-M1
1 sec
C4-M1

O2-M1

E1-M2 SEM

E2-M2

Chin EMG

FIGURE 3–4 Stage W with the eyes closed. A 30-second epoch is shown. Note that alpha rhythm is more prominent in the occipital derivation and present for more than 50% of
the epoch. Slow eye movements (SEMs) are present in this epoch. Although characteristic, they are not a criterion for scoring stage W or N1 in patients producing alpha rhythm
with eye closure. The chin electromyography (EMG) has a low amplitude, but this is variable. In some patients, the chin EMG activity is higher in stage W than during sleep.

(see Table 3–2, stage W rule B). The majority of the epoch eyes-closed stage W. Here, greater than 50% of the epoch has
contains either alpha rhythm or eye movements consistent prominent alpha activity. SEMs are also present and the
with wakefulness, and hence, the epoch is scored as stage W. EMG activity is relatively decreased. It should be noted that
In Figure 3–3, the chin EMG activity is relatively high, REMs SEMs can be seen during both eyes-closed stage W and stage
and blinks are present, and there is no discernible alpha N1. If patients produce alpha rhythm with eye closure, SEMs
activity. The epoch is scored as stage W using stage W rule are not part of the criteria to score stage W (although they
B (see Table 3–2). Figure 3–4 illustrates an example of are characteristic during eyes-closed stage W).
 Chapter 3    Sleep Staging in Adults 31

Stage N1 Stage N2
Low-amplitude mixed-frequency (LAMF) activity is defined Stage N2 is characterized by the presence of one or more
as a low-amplitude EEG pattern with predominantly 4- to nonarousal KCs (i.e., KCs NOT associated with an arousal)
7-Hz activity. Stage N1 is characterized by LAMF activity or one or more trains of SSs (Fig. 3–7). Arousal rules are
and the absence of sleep spindles (SSs) and K complexes discussed later in this chapter. During epochs of stage N2,
(KCs) not associated with arousal. SEMs may occur (see eye movements have usually ceased and the chin EMG is
Table 3–3). At the transition from stage N1 to stage N2, variable but usually at a level lower than that during wakeful-
vertex sharp waves may appear. In patients who produce ness. Recall that a KC is said to be associated with an arousal
alpha activity with eye closure (stage N1 rule A, see Table (KC+Ar) if the arousal commences no more than 1 second
3–3), the onset of stage N1 occurs when more than 50% of after the termination of the KC. Also note that the KC activ-
the epoch is marked by alpha attenuation (alpha activity in ity is seen in the recommended electro-oculographic (EOG)
< 50% of the epoch) and replacement with LAMF EEG (Fig. derivations (E1-M2 and E2-M2) as in-phase deflections in
3–5). In individuals who do not produce alpha activity with contrast to REMs (out-of-phase deflections). The rules for
eye closure (Fig. 3–6), the start of stage N1 occurs at the scoring stage N2 are listed in Table 3–5 and summarized in
earliest occurrence of SEMs, a slowing of the EEG by 1 Hz Table 3–6.
or more from that in stage W, or the presence of vertex sharp
waves (stage N1 rule B, see Table 3–3). Table 3–4 displays Start and Continuation of Stage N2
the characteristics of stage W and stage N1 for patients who According to the stage N2 rules (see Table 3–5, rule A), begin
do and do not produce alpha with eye closure. As noted in scoring stage N2 (see Fig. 3–7) when a KC (not associated
the AASM scoring manual, because SEMs may occur before with an arousal) or an SS occurs in the first half of the
alpha attenuation in subjects who have alpha activity with current epoch or the last half of the previous epoch. This
eye closure, the sleep onset may be scored somewhat earlier assumes that the epoch does not meet criteria for stage N3
in patients who do not produce alpha activity with eye (slow wave activity [SWA] present in ≥ 20% of the epoch,
closure. i.e., ≥6 sec). Recall that SWA is defined as EEG activity of

100 µV
F4-M1
1 sec
C4-M1
O2-M1
E1-M2
E2-M2

Chin1-Chin2

A
V
100 µV
F4-M1
1 sec
C4-M1
O2-M1
E1-M2
E2-M2

Chin EMG

FIGURE 3–5 A, Stage N1. A 30-second epoch in which no sleep spindles or K complexes are noted in the electroencephalogram (EEG). Less than 50% of the epoch has alpha rhythm.
The EEG shows alpha attenuation and low-amplitude mixed-frequency (4–7 Hz) activity for more than 50% of the epoch. Slow eye movements are present but not required. The
chin electromyogram (EMG) is often lower than stage W. B, Stage N1. A vertex sharp wave (V) is noted. No sleep spindles or K complexes are seen in the EEG. The chin EMG is relatively
high in this example.
Exploring the Variety of Random
Documents with Different Content
personages in the entourage of the Sultan. … The Porte,
however, protested, with the support of Germany, against the
appointment of Colonel Schaeffer, who appears to have been
suspected of English sympathies. Russia, too, it was said,
objected, insisting that the appointee must be of the Orthodox
Greek faith. Thus, on the question of selecting a
governor-general for Crete, the concert of the powers broke
down as it did at other points during the long crisis. At the
end of November the name of Prince Francis Joseph of
Battenberg was prominently mentioned as a prospective
candidate of favor. The Cretan assembly proposed, unless a
suitable governor were speedily chosen by the powers, to offer
the post to a candidate of its own selection."

Current History, 1897,

pages 865-866.

Months went on, while the Powers still discussed the Cretan
situation and no agreement was reached. In January, 1898, the
Turkish government appointed Edhem Pasha governor of Candia;
but, in the face of the admirals of the blockading squadrons,
who exercised an undefined authority, he seems to have had
practically little power. Presently, a new attempt was made to
select a Christian Governor-general. France and Russia
proposed Prince George of Greece, but Austria and Turkey
opposed. In April, Austria and Germany withdrew from the
blockade and from the "Concert," leaving Great Britain,
Russia, France and Italy to deal with Cretan affairs alone.
The admirals of these Powers, acting under instructions, then
divided the Cretan coast among themselves, each directing the
administration of such government as could be conducted in his
own part. The British admiral had Candia, the capital town,
and there trouble arose which brought the whole Cretan
business to a crisis. He attempted to take possession of the
customs house (September 6), and landed for that purpose a
small force of 60 men. They were attacked by a Turkish mob,
with which they fought desperately for four hours, losing 12
 killed and some 40 wounded, before they could make their
retreat to the shore and regain their ship. At the same time a
general massacre of Christians in the town was begun and some
800 perished before it was stopped. Edhem Pasha, with about
4,000 Turkish troops at his command, was said to have waited
long for the mob to do its work before he interfered.

This outbreak brought the four Powers to a decisive agreement.

They joined in imperatively demanding the withdrawal of
Turkish troops and officials from the island, and enforced the
demand. Guarantees for the safety of the Mohammedan population in
life and property were given; it was conceded that the
Sultan's suzerainty over Crete should be maintained, and he
was allowed to hold one military post in the island for a sign
of the fact. On those terms the Turkish evacuation of Crete
was carried out in November, and Prince George of Greece was
appointed, not Governor-general, but High Commissioner of the
four Powers, to organize an autonomous government in the
island and administer it for a period of three years. The
appointment was accepted, and Prince George was received with
rejoicing in Crete on the 21st of December. The blockade had
been raised on the 5th, and on the 26th the admirals departed.

During the following two years (1899-1900) there seems to have

been a generally good condition of order restored and
preserved. A constitution was framed by a national assembly,
which conferred the executive authority on Prince George, as
High Commissioner, with responsible councillors, and created a
Chamber of Deputies, elected for the most part by the people,
but containing ten members appointed by the High Commissioner.
Equal rights for all religious beliefs was made a principle of
the constitution.

TURKEY: A. D. 1899 (May-July).

Representation in the Peace Conference at The Hague.
 See (in this volume)
PEACE CONFERENCE.

{550}

TURKEY: A. D. 1899 (October).

Concessions to the Armenians.

In October an irade was published by the Sultan which withdrew

restrictions on the movements of Armenians in the provinces,
except in the case of suspects; granted pardon or commutation
of sentence to a number of Armenian prisoners; ordered payment
of sums due to Armenian government officials who had been
killed or expelled at the time of the massacres; directed
assistance to be given in the repairing and rebuilding of
churches, schools, and monasteries which had been injured or
destroyed, and also gave direction for the building of an
orphanage near Constantinople.

TURKEY: A. D. 1899 (November).

Railway to the Persian Gulf.

A German Bank Syndicate obtained from the Sultan, in November,

1899, a concession for the extension of the Anatolian Railway
from Konieh in Asia Minor, to Basra, or Bassorah, on the
Persian Gulf. The line, which will pass through Bagdad, and
along the valleys of the Tigris and Euphrates, is to be
completed within eight years from the date of the grant. "The
concession is regarded as a startling proof of German
influence in Constantinople, and a defeat both for Russian and
British diplomacy. It is certainly a defeat for the former,
and will greatly increase suspicion at St. Petersburg as to
the ultimate ends of Germany in Turkey; but we suspect that
Indian statesmen will perceive considerable compensations in
the arrangement. Not to mention that all railways which
approach India develop Indian trade, the railway may secure us
a strong ally in Asia. It is not of much use for Russia to be
running a line from the Caspian to Bushire if when she gets
there she finds Britain and Germany allied in the Persian
Gulf, and able by a railway through Gedrosia to Sind to throw
themselves right across her path."

Spectator (London),
December 2, 1899.

"The opposition of the French company owning the

Smyrna-Kassaba road, which extends east as far as Afion
Karahissar, was removed by granting this company 40 per cent
of the shares in the extension, and the local objection was
obviated by a provision in the concession giving the Turkish
Government the right to purchase the line at any time. Few
railroad lines can be of greater prospective importance than
this 2,000 miles of railroad uniting the Persian Gulf with
Europe, forming a rapid transit to and from the East, opening
up large tracts of agricultural country, and paving the way
for German commercial supremacy in Asia Minor and Mesopotamia.
It is not difficult to see how Germany, with preferential
rates for goods on German lines, will be able to control the
chief markets of Asia Minor and invade the East. … Germans
purchased the Constantinople-Ismid Railroad from an English
company and extended it to Angora. They also checkmated the
French and English by extending their line from Eski-Sher to
Konieh, thus preventing extension of both the Smyrna-Afion
Karahissar and the Smyrna-Aidin-Dinair roads. The two great
distributing points—Constantinople and Smyrna—are thus
controlled by Germans, and German goods may enter the interior
of Asia Minor and the great valley of the Tigris and Euphrates
on German-controlled roads at a decided advantage. Germans
have obtained the right to build docks and warehouses at Haida
Pasha, the terminus of the Anatolian railroads; and with
through rates for German goods on German lines, German freight
cars may be sent across the Bosphorus and travel to Mesopotamia
and the confines of India and Persia without change."
 United States Consular Reports,
April, 1900, page 497.

Professor Hilprecht has remarked, in the "Sunday School Times"

that "a new era for Babylonian archeology will begin when the
railroad from Koniah to Baghdad and Bassorah has been
constructed. It will then take about a week from London to the
ruins of Babylon, where, doubtless, a railway station (Hillah)
will be established. At present the traveler needs at the best
six weeks to cover this route. This railroad," says the
Professor, "has now become a certainty."

TURKEY: A. D. 1899-1901.
Impending outbreak in Macedonia.

The state of things in Macedonia, where the people have long

been on the brink of revolt against Turkish rule, excited to
it from Bulgaria and encouraged from Greece, but warned
otherwise by Russia and Austria, is thus described by the
"Economist," in an article quoted in "Littell's Living Age,"
March, 1899: "It is improbable, for reasons stated below, that
Macedonia will rise in insurrection this year [1899], but,
nevertheless, there is great danger in that quarter, which is
evidently disturbing both Vienna and St. Petersburg, and
exciting apprehensions in Constantinople. The Austrian and
Russian Foreign Offices are both issuing intimations that if a
revolt occurs Turkey will be allowed to put it down by Turkish
methods, and the Sultan is raising more troops, sending
Asiatic levies to Macedonia, and despatching some of his
ablest officers to control the hill districts. Severe warnings
have also been sent both to Belgrade and Sofia, and the Greeks
are warned that if their active party moves the Government of
Athens will not again be saved by Europe from the worst
consequences. All these symptoms imply that there is grave
fear, among those who watch Macedonia, that the patience of
her sorely oppressed people has given way, and that they have
resolved to risk everything rather than remain longer under
 the rule of Pashas from whom no man's life and no woman's
honor is safe for twelve hours together. It is known,
moreover, that the course of events in Crete and the
appearance of the Tsar's Rescript have greatly stirred the
population. The former is held by them to show that if a
Christian population in Turkey will risk massacre, Europe will
not allow them to be exterminated, while the latter has made
submission more difficult by putting an end to hope for the
next five years. …

"Turkish subjects must be driven to despair before they will

rise against the Turks, and if they can even hope to be left
alone, the Macedonians will wait, rather than encounter so
dreadful a risk. They have, it is true, the example of the
Cretans to encourage them, but their country is not an island,
and they have the fate both of the Armenians and the
Thessalians to warn them that on the mainland the Turks cannot
be resisted by half-drilled forces.

"It seems almost a truism to say that Europe is foolish to

allow such a source of danger as Macedonia presents to
continue without a cure; but there is something to be said on
the other side. The Powers sincerely desire peace, and the
Macedonian magazine cannot be flooded without a war, if it be
only a war between Russia and the Sultan.
{551}
Nobody knows to what such a war would lead, or in what
condition Eastern Europe might emerge from it. Moreover,
however much the Macedonians may excite the sympathies of
philanthropists, they have done a good deal to alienate those
of politicians. They decline to be either Austrian or Russian.
They asked for years to be aided by Greece, and when Greece
declared war on Turkey they refused to rise behind Edhem
Pasha, whom they could have cut away from his supports. They
now ask aid from Bulgaria, but they are most unwilling to
submit to Sofia, and so make of Bulgaria a fairly strong
State. They wish, they say, to make of Macedonia a
Principality, but if it were so made the Slavo-Macedonians
would begin fighting the Græco-Macedonians, until both had
been nearly ruined. They must join one party or the other if
they wish to be free, and stick to the one they join, and
fight for it with a coherence which they have never yet
displayed."

On the 7th of January, 1901, a correspondent of the "London

Times" wrote on the same subject from Vienna, as follows:

"The situation in Macedonia, as described in trustworthy

accounts coming from different directions, testifies to the
increasing danger of trouble. Things have gone so far that an
outbreak may occur this year. In diplomatic circles it is
considered impossible that in any case it can be delayed for
longer than a twelvemonth. In Constantinople, Athens, and all
the capitals of the Balkan States the eventuality of a
Macedonian rising has been expected for several years past,
and in more than one instance preparations have been made
accordingly. To what extent the Macedonia committees have
received official patronage in Bulgaria is now of secondary
interest. The mischief has been done, and the agitation in
Macedonia is at present beyond the control of the Bulgarian
authorities, even if they wished to keep it in check, which is
not certain. All that can be said with confidence is that last
summer Austria and Russia made a vigorous and successful effort
to put an end to the almost open encouragement extended to the
Macedonian committees at Sofia, which was within an ace of
involving the Principality in a war with Rumania. The
Austro-Russian 'entente' [an understanding or agreement
between Russia and Austria, in 1897, to act together in
keeping peace in the Balkan peninsula] has, in fact, done
excellent service wherever diplomatic pressure can be brought
to bear. But, unfortunately, that does not include Macedonia.
If the revolutionary element in that province of the Ottoman
Empire sets at defiance the imposing Turkish forces
concentrated on the spot, it is not likely that it will be
 influenced by what is probably regarded as the remote
contingency of the direct armed intervention of Austria and
Russia. All the warnings and scoldings in the world will not
suffice to preserve peace in Macedonia.

"It is difficult to say what foundation there may be for the

statement that the Sultan himself seeks to take advantage of
the disturbed condition of Macedonia for purposes of his own.
It is alleged that he wishes to prevent any change in the
existing regime in Crete by exciting the apprehension in
Athens and elsewhere that an attempt to modify the status quo
in that island would cause a massacre of the Hellenic
population in Macedonia. This view of the case finds
expression in the following extracts from a letter addressed
to the 'Roumanie,' one of the leading organs of
Bukharest:—'The thoroughly bad policy pursued by the Sublime
Porte in Macedonia, which consists in allowing that unhappy
province to remain a prey to Bulgarian agitators so as in case
of need to terrify diplomacy by the spectre of a revolution,
has contributed to open the eyes of the Powers. On the other
hand, the irresistible attraction exercised by the Kingdom of
Greece, not only on the Cretans themselves, but also on all
the rayahs of the Ottoman Empire, is an indisputable fact.'"

See, also (in this volume),

BALKAN AND DANUBIAN STATES.

TURKEY: A. D. 1900.
The Zionist movement of the Jews to colonize Palestine.

See (in this volume)

JEWS: A. D. 1897-1901.

TURKEY: A. D. 1901.
The Cretan question.

The provisional arrangement of government for Crete,

administered by Prince George, of Greece, as High Commissioner
for the Powers, expires by limitation in December, 1901. What
shall then be done with the island is a question that was
referred, by the several Powers of the Concert, in the early
part of the year, to their ambassadors at Rome, in conference
with the Italian Minister of Foreign Affairs. The
administration of Prince George appears to have been quite
remarkably satisfactory to all concerned, and its continuation
was evidently desired, as much by the Cretans as by the
protecting Powers; but the former sought to have it placed on
a basis of permanency, in some form that would be practically
tantamount to the long craved annexation to Greece. Prince
George naturally looks in the same direction, and he is said
to have made it known that he would decline to hold his post
provisionally beyond the term of three years for which he
accepted it in 1898. The ambassadorial conference at Rome
decided, however, that the time has not come for a permanent
settlement of the Cretan question, and that the provisional
arrangement for its government must be renewed. A Press
despatch from Athens, on the 22d of March, 1901, announced the
decision and indicated the circumstances of the situation, as
follows:

"The Cretan Assembly meets at the end of next month, and its
probable attitude towards the question of union with Greece is
already the subject of speculation here. The decision of the
conference of Ambassadors at Rome is embodied in a memorandum
which has been handed to Prince George by the Consuls at
Canea, while a copy of the document has been unofficially
presented to King George 'à titre d'information.' The
Ambassadors express their opinion that any manifestation on
the part of the Cretans in favour of union with Greece would
be inopportune at the present moment, and they propose a
prolongation of the present provisional system of government
without assigning any definite term to the High Commissioner's
mandate.
 "Whether Prince George, who is an enthusiastic advocate of
union with Greece, will accept the new arrangement
unconditionally remains to be seen. Meanwhile the islanders
are occupied with preparations for the elections.

{552}

"It appears that at a recent sitting of the Prince's Council

one of the most prominent of Cretan politicians advocated the
institution of an autonomous Principality on the lines already
laid down by the existing Constitution. The proposal provoked
a violent outburst on the part of the Athenian Press, which
denounces its author as a traitor to the cause of Hellenism.
The opinion apparently prevails here that the establishment of
a Principality would finally preclude the union of the island
with Greece."

TURKEY: A. D. 1901.
Order regulating the visit of Jews to Palestine.

See (in this volume)

JEWS: A. D. 1901.

----------TURKEY: End--------

TWAIN, Mark:
Description of scenes in the Austrian Reichsrath.

See (in this volume)

AUSTRIA-HUNGARY: A. D. 1897 (OCTOBER-DECEMBER).

U.

UCHALI, Treaty of.

See (in this volume)

ITALY: A. D. 1895-1896.
UGANDA: A. D. 1894.
Creation of the Protectorate.

See (in this volume)

BRITISH EAST AFRICA PROTECTORATE: A. D. 1895-1897.

UGANDA: A. D. 1897-1898.
Native insurrection and mutiny of Sudanese troops.

A train of serious troubles in the Uganda Protectorate began

in May, 1897, with an insurrection of some of the chiefs,
instigated by the king, Mwanga, who was restive under British
control. The revolt was suppressed after some sharp fighting,
especially at Kiango, on the 24th of July, and King Mwanga
escaped into German territory. In August he was formally
deposed by a council of chiefs, and his infant son, Chua, was
elected king in his place, under a regency of three of the
chiefs. But a more serious trouble followed, from the mutiny
of a part of the Sudanese troops which had been serving in
Uganda. These troops were being sent to join an expedition,
under Major Macdonald, for the exploration of the districts
adjacent to the Italian sphere of influence, and were not
permitted to take their women with them. This seems to have
been their chief grievance. They also complained of being
overworked, underpaid, insufficiently fed, and commanded by
young officers who would not listen to their complaints. They
seized Fort Lubas, on the frontier between Uganda and Usoga,
made prisoners of several of their officers, whom they finally
murdered, and held the fort against repeated attacks until early
in January, 1898, when they made their escape. They were
pursued and attacked (February 24) at Kabagambe, on Lake
Kioja, where they had built a fort. Many were killed, the
remainder much scattered. A considerable party got away to the
eastern side of the Nile and continued to give trouble there
throughout the year.
Meantime, the deposed king, Mwanga, had escaped from the
Germans and effected a new rising among his late subjects; and
another deposed king, Kabarega, of Unyoro, had also
reappeared, to make trouble in that region. After the
suppression of the Sudanese mutiny these risings were
overcome, with the help of some 1,100 troops brought from
India for the emergency. In March, there was news of
Kabarega's death, and the British Acting Commissioner and
Consul General issued the following proclamation:

"Whereas Kabarega, the deposed King of Unyoro, is reported to

have deceased, and whereas the present disordered state of
affairs in that country has proved that, for the maintenance
of good government and good-will, it is expedient to provide
for the succession to the kingdom of a member of the Royal
House, it is hereby publicly proclaimed that Karukala, son of
Kabarega, is now appointed King of Unyoro, under the
protection of Her Britannic Majesty. The Kingdom of Unyoro
comprises the provinces of—Busindi, Shifalu, Magungu, Kibero,
Bugoma, Bugahiaobeire. This appointment is in accordance with
the general conditions by which countries in British African
Protectorates are guided and regulated, and it secures to the
Kingdom of Unyoro all the advantages which accrue from its
being an integral part of such a Protectorate. The local
government of the country will be administered, under the
guidance of Her Majesty's Representative, by a Council of
Regency of either two or three Chiefs, to be appointed by Her
Majesty's Commissioner. This Council of Regency will, subject
to the approval of Her Majesty's Commissioner, select and
appoint the Katikiro and the other Chiefs of the first rank
required in accordance with local custom. These Chiefs, on
their appointment being confirmed, will select and appoint in
full Council the lesser grade Chiefs, until the system of
local administration is complete."

Great Britain, Parliamentary Publications

(Papers by Command: Africa, Number 7, 1898, page 42).
UGANDA RAILWAY, The.

On the 30th of April, 1900, the British Parliament voted

£1,930,000 for the completion of the railway under
construction from Mombasa, on the Indian Ocean, to Lake
Victoria-Nyanza, officially known as the Mombasa-Victoria
Railway. Previous expenditure had been about £3,000,000. On
the 30th of October it was reported that rails were laid down
to the 452d mile from Mombasa, and that advance gangs were
working about 40 miles beyond that point.

UITLANDERS.

See (in this volume)

SOUTH AFRICA (THE TRANSVAAL): A. D. 1885-1890, and after.

UNGAVA, The district of.

See (in this volume)

CANADA: A. D. 1895.

UNITED CHRISTIAN PARTY.

See (in this volume)

UNITED STATES OF AMERICA: A. D. 1900 (MAY-NOVEMBER).

UNITED IRISH LEAGUE, The.

See (in this volume)

IRELAND: A. D. 1900-1901.

UNITED STATES OF BRAZIL.

See (in this volume)

BRAZIL.
{553}

----------UNITED STATES OF AMERICA: Start--------

UNITED STATES OF AMERICA: A. D. 1868-1885.

Cuban questions in controversy with Spain.

See (in this volume)

CUBA: A. D. 1868-1885.

UNITED STATES OF AMERICA: A. D. 1894.

Legislation to promote the reclamation of arid lands.

The following measure of legislation to promote the

reclamation of arid lands was carried through Congress as an
amendment to the appropriation bill for Sundry Civil
Expenditures, and became law August 18, 1894:

"Section 4.
That to aid the public land States in the reclamation of the
desert lands therein, and the settlement, cultivation, and
sale thereof in small tracts to actual settlers, the Secretary
of the Interior with the approval of the President, be, and
hereby is, authorized and empowered, upon proper application
of the State to contract and agree, from time to time, with
each of the States in which there may be situated desert lands
as defined by the Act entitled 'An Act to provide for the sale
of desert land in certain States and Territories,' approved
March 3d, 1877, and the Act amendatory thereof, approved March
3d, 1891, binding the United States to donate, grant and
patent to the State free of cost for surveyor price such
desert lands, not exceeding one million acres in each State,
as the State may cause to be irrigated, reclaimed, occupied,
and not less than twenty acres of each one hundred and
sixty-acre tract cultivated by actual settlers, within ten
years next after the passage of this Act, as thoroughly as is
required of citizens who may enter under the said desert land
law.

"Before the application of any State is allowed or any

contract or agreement is executed or any segregation of any of
the land from the public domain is ordered by the Secretary of
the Interior, the State shall file a map of the said land
proposed to be irrigated which shall exhibit a plan showing
the mode of the contemplated irrigation and which plan shall
be sufficient to thoroughly irrigate and reclaim said land and
prepare it to raise ordinary agricultural crops and shall also
show the source of the water to be used for irrigation and
reclamation, and the Secretary of the Interior may make
necessary regulations for the reservation of the lands applied
for by the States to date from the date of the filing of the map
and plan of irrigation, but such reservation shall be of no
force whatever if such map and plan of irrigation shall not be
approved. That any State contracting under this section is
hereby authorized to make all necessary contracts to cause the
said lands to be reclaimed, and to induce their settlement and
cultivation in accordance with and subject to the provisions
of this section; but the State shall not be authorized to
lease any of said lands or to use or dispose of the same in
any way whatever, except to secure their reclamation,
cultivation and settlement.

"As fast as any State may furnish satisfactory proof according

to such rules and regulations as may be prescribed by the
Secretary of the Interior, that any of said lands are
irrigated, reclaimed and occupied by actual settlers, patents
shall be issued to the State or its assigns for said land so
reclaimed and settled: Provided, That said States shall not
sell or dispose of more than one hundred and sixty acres of
said land to any one person, and any surplus of money derived
by any State from the sale of said lands in excess of the cost
of their reclamation, shall be held as a trust fund for and be
applied to the reclamation of other desert lands in such
State. That to enable the Secretary of the Interior to examine
 any of the lands that may be selected under the provisions of
this section, there is hereby appropriated out of any moneys
in the Treasury, not otherwise appropriated, one thousand
dollars."

Acts, 53d Congress, 2d Session, chapter 301.

UNITED STATES OF AMERICA: A. D. 1895.

Re-survey of Mexican boundary.

See (in this volume)

MEXICO: A. D. 1892-1895.

UNITED STATES OF AMERICA: A. D. 1895 (January-February).

The monetary situation.
Contract for replenishing the gold reserve in the Treasury.

The alarming situation of the Treasury of the United States at

the beginning of the year 1895 was clearly described by the
President in his special Message to Congress, January 28.

See in volume 5, UNITED STATES OF AMERICA: A. D. 1895).

By the operation of what had been aptly called "the endless

chain" of the greenback currency issues of the government
(paid out with one hand, to be redeemed with the other in
gold, which the declining value of silver brought more and
more into demand) the gold reserve in the Treasury was fast
being exhausted, and the hour was approaching when, without
some effective relief, the obligations of the nation would
have to be paid in depreciated silver coin, and its credit
lost. The appeal of the President to Congress had no effect.
The Senate was controlled by a majority of men who desired
precisely the result which he wished to avert. The state of
things in that body was described by Senator Sherman, of the
Committee on Finance, in the following words:
 "The Committee on Finance is utterly helpless to deal with
this vast question. We are quite divided upon it. We are not
allowed to propose a measure to this Senate which all can
approve of, unless there is attached to it a provision for
free coinage of silver."

The attitude of the House was different, but almost equally

hostile to the President's views. Its Republican majority was
not favorable to the aims of the free silver parties, but held
that the relief needed for the Treasury was to be sought in a
return to higher import duties, as a means of obtaining
increased revenue. Hence, a bill to carry out the
recommendations of the President was rejected in the House, on
the 7th of February, by a vote of 162 against 135.

On the following day, the Secretary of the Treasury, Mr.

Carlisle, exercising authority which he possessed to sell
certain four per cent. thirty year bonds, contracted with
August Belmont &; Co., who represented the Rothschilds of
London, and with the house of J. P. Morgan & Co., of New York,
on behalf of J. S. Morgan & Co., London, and themselves, for
supplying 3,500,000 ounces of standard gold coin of the United
States, at the rate of 817.80441 per ounce, in exchange for
such bonds. It was a condition of the contract that one half
of the coin supplied should be brought from Europe: also that
the contracting syndicate should use its influence to protect
the Treasury against withdrawals of gold.
{554}
At the same time, the Secretary of the Treasury reserved the
right to substitute three per cent. gold bonds, if Congress
would authorize such an issue, to be taken by the syndicate at
par, in place of the four per cents to which his existing
authority was restricted. It was shown that the consequent
saving in interest would be $539,000 per annum, amounting to
$16,174,770 in thirty years; but the proposal was rejected in
the House of Representatives by 167 votes against 120. The
contract was accordingly carried out in its original form,
 with success so far that the withdrawals of gold from the
Treasury dropped for a considerable period to a low point. It
appeared that when this emergency break was put upon the
working of the "endless chain," the sub-treasury in New York
was believed to be within twenty-four hours of a suspension of
gold payments. But the contract was loudly condemned.
nevertheless, by the opponents of the administration.

UNITED STATES OF AMERICA: A. D. 1895 (February).

Renewed insurrection in Cuba.

See (in this volume)

CUBA: A. D. 1895.

UNITED STATES OF AMERICA: A. D. 1895 (April-May).

Decision of the Supreme Court against the constitutionality
of the Income Tax.

Cases testing the constitutionality of the income tax which

Congress had attached to the Tariff Act of 1894, were brought
to a partial decision in the Supreme Court in April, and
finally in May, 1895.

See, in volume 4 of original edition,

or in volume 5 of revised edition,
TARIFF LEGISLATION, UNITED STATES: A. D. 1894.

[Transcriber's note: For this set see, Volume 4,

"TARIFF: (United States): A. D. 1894."]

The cases in question were "Pollock v. Farmers' Loan and

Trust Company," and "Hyde v. Continental Trust Company." On
the first hearing, the illness and absence of one of the
justices, Mr. Jackson, of Tennessee, left but eight members
in attendance, and they divided equally on several points
which were vital to the decision of the question of
constitutionality in the tax. The appellants accordingly
filed a petition for a re-hearing, submitting, among other
reasons, the following: "The question involved in these cases
was as to the constitutionality of the provisions of the
tariff act of August 15, 1894 (sections 27 to 37), purporting
to impose a tax on incomes. The Court has held that the same
are unconstitutional, so far as they purport to impose a tax
upon the rent or income of real estate and income derived
from municipal bonds. It has, however, announced that it was
equally divided in opinion as to the following questions, and
has expressed no opinion in regard to them: (1) Whether the
void provisions invalidate the whole act. (2) Whether, as to
the income from personal property as such, the act is
unconstitutional as laying direct taxes. (3) Whether any part
of the tax, if not considered as a direct tax, is invalid for
want of uniformity.

"The court has reversed the decree of the Circuit Court and
remanded the case, with directions to enter a decree in favor
of complainant in respect only of the voluntary payment of the
tax on the rents and income of defendant's real estate and
that which it holds in trust, and on the income from the
municipal bonds owned or so held by it. While, therefore, the
two points above stated have been decided, there has been no
decision of the remaining questions regarding the
constitutionality of the act, and no judgment has been
announced authoritatively establishing any principle for
interpretation of the statute in those respects."

The re-hearing asked for was granted by the Court on the 6th
of May, when Justice Jackson was able to take his seat on the
bench, after which, on the 20th of May, by the opinion of five
members of the Court against four, the law was pronounced null,
so far as concerned the imposition of a tax on incomes. The
opinion of the majority was delivered by Chief Justice Fuller,
who said, in part:

"The Constitution divided Federal taxation into two great

classes, the class of direct taxes, and the class of duties,
imposts and excises; and prescribed two rules which qualified
the grant of power as to each class. The power to lay direct
taxes apportioned among the several States in proportion to
their representation in the popular branch of Congress, a
representation based on population as ascertained by the
census, was plenary and absolute; but to lay direct taxes
without apportionment was forbidden. The power to lay duties,
imposts, and excises was subject to the qualification that the
imposition must be uniform throughout the United States.

"Our previous decision was confined to the consideration of

the validity of the tax on the income from real estate and on
the income from municipal bonds. … We are now permitted to
broaden the field of inquiry, and to determine to which of the
two great classes a tax upon a person's entire income, whether
derived from rents, or products, or otherwise, of real estate,
or from bonds, stocks, or other forms of personal property,
belongs; and we are unable to conclude that the enforced
subtraction from the yield of all the owner's real or personal
property, in the manner prescribed, is so different from a tax
upon the property itself, that it is not a direct, but an
indirect tax in the meaning of the Constitution.

"The words of the Constitution are to be taken in their

obvious sense, and to have a reasonable construction. In
Gibbons v. Ogden, Mr. Chief Justice Marshall, with his usual
felicity, said: 'As men, whose intentions require no
concealment, generally employ the words which most directly
and aptly express the ideas they intend to convey, the
enlightened patriots who framed our Constitution, and the
people who adopted it must be understood to have employed
words in their natural sense, and to have intended what they
have said.' 9 Wheat. 1, 188. And in Rhode Island v.
Massachusetts, where the question was whether a controversy
between two States over the boundary between them was within
the grant of judicial power, Mr. Justice Baldwin, speaking for
Welcome to our website – the ideal destination for book lovers and
knowledge seekers. With a mission to inspire endlessly, we offer a
vast collection of books, ranging from classic literary works to
specialized publications, self-development books, and children's
literature. Each book is a new journey of discovery, expanding
knowledge and enriching the soul of the reade

Our website is not just a platform for buying books, but a bridge
connecting readers to the timeless values of culture and wisdom. With
an elegant, user-friendly interface and an intelligent search system,
we are committed to providing a quick and convenient shopping
experience. Additionally, our special promotions and home delivery
services ensure that you save time and fully enjoy the joy of reading.

Let us accompany you on the journey of exploring knowledge and

personal growth!

ebookfinal.com