Citation
How sleep is different in pregnancy

Material Information

Title:
How sleep is different in pregnancy exploring how sleep patterns and cytokine and endocrine parameters change in the 3rd trimester of pregnancy
Creator:
Okun, Michele Lee
Publication Date:
Language:
English
Physical Description:
xxi, 269 leaves : ; 28 cm

Subjects

Subjects / Keywords:
Pregnant women ( lcsh )
Sleep ( lcsh )
Cytokines ( lcsh )
Endocrine gynecology ( lcsh )
Cytokines ( fast )
Endocrine gynecology ( fast )
Pregnant women ( fast )
Sleep ( fast )
Genre:
bibliography ( marcgt )
theses ( marcgt )
non-fiction ( marcgt )

Notes

Bibliography:
Includes bibliographical references (leaves 235-269).
General Note:
Department of Health and Behavioral Sciences
Statement of Responsibility:
by Michele Lee Okun.

Record Information

Source Institution:
|University of Colorado Denver
Holding Location:
|Auraria Library
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
66899503 ( OCLC )
ocm66899503
Classification:
LD1193.L566 2005d O58 ( lcc )

Full Text
V
HOW SLEEP IS DIFFERENT IN PREGNANCY: EXPLORING HOW SLEEP
PATTERNS AND CYTOKINE AND ENDOCRINE PARAMETERS CHANGE IN
THE 3rd TRIMESTER OF PREGNANCY
by
Michele Lee Okun
B.A., University of California Santa Barbara, 1991
M.A., San Francisco State University, 1998
A thesis submitted to the
University of Colorado at Denver
in partial fulfillment
of the requirements for the degree of
Doctor of Philosophy
Health and Behavioral Sciences
2005


This thesis for the Doctor of Philosophy
degree by
Michele Lee Okun
has been approved
by
Kenneth Wright, Jr.


Okun, Michele Lee (Ph.D., Health and Behavioral Sciences)
How Sleep is Different in Pregnancy: Exploring How Sleep Patterns and Cytokine
and Endocrine Parameters Change in the 3 rd Trimester of Pregnancy
Thesis directed by Associate Professor Mary Coussons-Read
ABSTRACT
There is ample information in the literature about the immunological
consequences of illness, stress, and psychopathology during gestation, but little is
known regarding the potential unfavorable effects of chronic sleep loss/disruption
during pregnancy on maternal health, pregnancy outcomes, and fetal health.
Although women commonly report their sleep to be disrupted during pregnancy, there
is little information addressing the immunological or endocrinological consequences
for pregnant women who experience excessive sleep disruption.
This exploratory study examined how the sleep patterns of both pregnant
women (71) and nonpregnant women (43) are related to and influence various
cytokine and endocrine parameters. Several research questions were asked to
understand this inadequately studied area.
One hundred-fourteen women were recruited though the University of
Colorado. Women completed sleep questionnaires (ESS, PSQI, and Additional Sleep
Questions), provided blood samples, and kept sleep diaries for 2 weeks following the
blood collection. Two subsets of women provided additional data: one subset of
women (13 pregnant and 12 nonpregnant) wore actigraphy watches (Minimitter, OR)
in addition to completing the sleep diaries which provided an objective measure of
the sleep parameters being assessed; another subset gave a second blood sample at the
38-40 week point with the intention of deriving some directionality from the sleep
data. Data were collected at 35-39 weeks of pregnancy. Serum levels of TNF-a, IL-
4, IL-6 and IL-10 were determined via ELISA kits (Biosource Europe); serum levels
of estriol, progesterone and C-Reactive Protein were determined via EIA kits (DSL).
The sleep variables of interest were: SOL, WASO, TIB, TST and SE from both sleep
diaries and actigraphy, in addition to number and length of naps and number of
awakenings from the sleep diary alone.
Significant differences in sleep patterns were identified between the two
groups of subjects. Pregnant women report more dissatisfaction with their sleep
quality, higher daytime dysfunction and appear to require more sleep than the sample
m


of nonpregnant women. Pregnant women experience alterations in sleep and
pregnancy associated cytokine and endocrine variables compared to nonpregnant
women suggesting that sleep may be a moderator in these outcomes.
This abstract accurately represents the content of the candidates thesis. I recommend
its publication.
IV


DEDICATION
An accomplishment such as a doctoral degree is not done alone. In order for
such a feat to be accomplished there must be others who provide support, of all kinds,
throughout the process. This dedication is to acknowledge those few who gave me
the emotional support, the spiritual strength, and the encouraging words whenever I
asked for them, and especially when I didnt.
To my loving and supportive father, Jeff, who continuously told me how
proud he was of me. He subtly boasted about me and my accomplishments to his co-
workers, his friends and to other members of our family. I only knew about this when
I met these individuals face to face. He has always, throughout my entire life, been
my cheerleader and my biggest fan. I am eternally grateful and I love you.
To my beautiful and encouraging sister, Stacy, who like my father bragged
about me quietly. She constantly asked how I was progressing and when I would
finish. She was my outside motivator, something I needed desperately at times. She
made sure I was reminded of all the encouraging words I used to spew at her
throughout her educational career. She was the reminder that I could accomplish
anything I desired. Thank you and I love you!
Finally, to the person who has had the biggest impact on my life during this
process, John. As my best friend you listened to me, more often than not, when
things were rough as well as when they were great. You gave me words of
I


encouragement, reason and support more often than I can remember. Your
willingness to take on the burden of reading my dissertation and making notes that
will always make me smile is something I will never be able to return in kind. The
way you helped me rationalize particular events throughout this process made me
stronger and better able to take on new challenges. There have been so many times
that I wondered how would I make it through? and then your voice came on the line
and we worked it through. You, like my father, are my biggest fan. The
encouragement I received from you I view as a gift. One that I will treasure for the
rest of my life. You are my rock, and I love you!


ACKNOWLEDGMENTS
A project such as this is never done alone. Many people assisted me greatly
throughout my journey. I would like to acknowledge these people here with much
gratitude.
To my committee, Lauren Clark, Loma G. Moore, Bob Ballard and Ken
Wright Jr.: I knew I selected a group of tremendous people to help me through the
process of achieving my doctorate, but each of you brought something unique and
insightful to my life. I will always remember this group and how you helped me
grow.
I want to thank my chair, Mary Coussons-Read, who challenged me to
become a better researcher and person. She let me learn and explore things that may
not have been possible in any other persons lab. She encouraged me at the right
times and criticized me when necessary. These are traits that make a good mentor.
Thank you for bringing me back into the world of PNI and into womens health.
To the two ladies who helped me in times of need- Jenna Lenhart and Jennifer
Karsten. You both gave of your time to allow me to have blood draws possible and
asked for nothing in return. You both are true angels.
I am eternally grateful to Sue Purcell for her selfless desire to help me with
the daunting task of formatting the dissertation. Sue, you have been a true friend and
savior. You are an incredible person, not only for helping but for giving of your time
so unabashedly. I do not know if or when I would have finished this project if it were


not for you and your assistance. I am eternally grateful for your friendship and
support.
Finally to my friends who gave me support and words of encouragement
during these four years. You were my surrogate family and I am thankful for you all.
I will always reflect that A kind word goes a long way.


CONTENTS
Figures................................................................xv
Tables..............................................................xviii
List of Abbreviations.................................................xix
CHAPTER
1. INTRODUCTION.........................................................1
Research Question, Purpose, and Aims............................4
Significance of Study...........................................5
2. BACKGROUND, THEORY, AND LITERATURE REVIEW............................6
Sleep Physiology................................................6
Why is Sleep So Important?.....................................10
Restorative Function of Sleep...........................10
To Regulate Brain Excitability..........................12
To Provide a Stress Free Period.........................13
Costs of Sleep Loss.....................................15
Immunity.......................................................16
The Normal Immune Response An Introduction............16
Neural-Immune System Interactions.......................23


Sleep and Immune Function........................................ 24
The Role of Neural, Immune, and Endocrine Substances
in Sleep....................................................24
The Multifaceted Association Between Sleep and
Inflammation................................................32
Sleep Deprivation and Immunity: A Key to Understanding
Sleep..............................................................35
Voluntary/Experimental Sleep Disruption.....................35
Insomnia and Immune Functioning.............................41
Pregnancy..........................................................42
Endocrinology of Pregnancy..................................43
Physiology of Pregnancy.....................................44
Immunology of Pregnancy.....................................45
Pregnancy and Sleep................................................47
Understanding Insomnia......................................47
Restorative Function of Sleep During Pregnancy..............50
Pregnancy and Alterations in Sleep..........................51
Pregnancy and Sleep Disorders: Is Pregnancy a Risk
Factor for RLS?.............................................59
Understanding Sleep in Pregnancy via Animal Models..........62
Mechanisms of Sleep Disruption in Pregnancy.................63
Pregnancy and Immunity.............................................65
Pregnancy and Alterations in Immune Function................65


Consequences of the Inflammatory Response During
Pregnancy...................................................76
Endocrine Participation During Pregnancy....................79
Disordered Sleep and Immune Function in Pregnancy.................82
3. RESEARCH DESIGN AND METHODS.............................................89
Overview..........................................................89
Hypothesis........................................................90
Research Questions................................................91
Sampling Design...................................................92
Determining Sample Size.....................................93
Protection of Human Subjects................................95
Participants................................................95
Procedure...................................................95
Instruments.................................................96
Variables..................................................109
Methodology for Blood Collection...........................113
Data Analysis....................................................121
Research Question 1. How do sleep patterns differ
between pregnant and nonpregnant women?....................121
Research Question 2. How does sleep disruption affect
cytokine and endocrine parameters in pregnant and
nonpregnant women?.........................................122
Research Question 3. How are sleep parameters related
to cytokine and endocrine levels?..........................123


Research Question 4. How does excessive sleep
disruption affect cytokine and endocrine parameters?........123
Research Question 5. Do cytokine and endocrine
parameters tested at the 35-37 week point versus the 37-
39 week mark differ for pregnant women, and is sleep
disruption a contributor to this change?....................123
Research Question 6. What percentage of 3rd trimester
women report symptoms of RLS and how are these
womens immunological profiles different from those of
3rd trimester women without symptoms of RLS?...............124
4. RESULTS.................................................................125
Final Sample Demographics.........................................125
Research Aim #1: Sleep Patterns in Pregnant and
Nonpregnant Women.................................................127
Research Question 1. How do sleep patterns differ
between pregnant and nonpregnant women?....................127
Research Question 2. How does sleep disruption affect
cytokine and endocrine parameters in pregnant and
nonpregnant women?.........................................139
Research Aim #2: Exploration of How Levels of Sleep
Disruption Influence Immune Profiles During Pregnancy.............143
Research Question 3. How are sleep parameters related
to cytokine and endocrine levels during pregnancy?.........143
Research Question 4. How does excessive sleep
disruption affect cytokine and endocrine parameters?.......156
Research Question 5. Do the immunological and
endocrinological parameters tested at the 35-37 week
point versus the 37-39 week mark differ for pregnant
women, and is sleep disruption a contributor to this
change?....................................................157


Research Question 6. Do the cytokine and endocrine
profiles differ in women who report symptoms of RLS in
the 3rd trimester compared to those who do not?.............158
5. DISCUSSION AND CONCLUSIONS...............................................160
Overview...........................................................160
Research Aim #1: Sleep Patterns in Pregnant and
Nonpregnant Women..................................................163
Research Question 1. How do sleep patterns differ
between pregnant and nonpregnant women?.....................163
Research Question 2. How does sleep disruption affect
cytokine and endocrine parameters in pregnant and
nonpregnant women?..........................................168
Research Aim #2: Exploration of How Levels of Sleep
Disruption Influence Immune Profiles During Pregnancy..............184
Research Question 3. How are sleep parameters related
to cytokine and endocrine levels?...........................184
Research Question 4. How does excessive sleep
disruption affect cytokine and endocrine parameters?........190
Research Question 5. Do the cytokine and endocrine
parameters tested at the 35-37 week mark versus the 37-
39 week mark differ for pregnant women, and is sleep
disruption a significant contributor to this change?........192
Research Question 6. Do the cytokine and endocrine
profiles differ in women who report symptoms of RLS in
the 3rd trimester compared to those who do not?.............196
Strengths/Limitations of the Study.................................197


APPENDIX
A. COLORADO MULTIPLE INSTITUTIONAL
REVIEW BOARD APPROVAL AND SUBJECT
CONSENT FORMS................................208
B. SLEEP DIARY..................................221
C. EPWORTH SLEEPINESS SCALE (ESS)...............223
D. PITTSBURGH SLEEP QUALITY INDEX (PSQI)........225
E. ADDITIONAL SLEEP DISORDERS
QUESTIONNAIRE (ASDQ).........................229
REFERENCES............................................235
xiv


FIGURES
Figure
2-1. EEG Representation of Sleep Stages......................................7
2-2. A Normal Human Hypnogram................................................9
2-3. T-Helper Cells Differentiate Into Thl and Th2 Cells....................19
2-4. Hypothetical Diagram of How Sleep Disruption Can
Influence Immune Parameters in Pregnancy...............................87
4-1. Average Number of Naps (Sleep Diary)..................................129
4-2. Average Length of Nap (Sleep Diary)...................................129
4-3. Sleep Onset Latency (Sleep Diary).....................................130
4-4. Average Number of Awakenings (Sleep Diary)............................130
4-5. Wake After Sleep Onset (Sleep Diary)..................................131
4-6. Time in Bed (Sleep Diary).............................................131
4-7. Total Sleep Time (Sleep Diary)........................................132
4-8. Sleep Efficiency (Sleep Diary)........................................132
4-9. Time in Bed (Actigraphy)..............................................133
4-10. Total Sleep Time (Actigraphy)........................................133
4-11. Sleep Onset Latency (Actigraphy).....................................134
4-12. Sleep Efficiency (Actigraphy)........................................134
4-13. Comparison Between Group and Level of Sleep
Disruption on CRP.....................................................141
xv


4-14. Comparison Between Group and Level of Sleep
Disruption for Progesterone..........................................142
4-15. Higher Serum IL-4 Levels Are Associated with Fewer
Naps per Week........................................................145
4-16. Higher Serum IL-4 Levels Are Associated with
Shorter Naps.........................................................145
4-17. Higher Daytime Dysfunction As Reported on the
PSQI Is Associated with Higher Serum IL-6 Levels....................146
4-18. Higher Progesterone Levels Are Associated with
Fewer Nighttime Awakenings...........................................146
4-19. Poorer Subjective Sleep Quality Is Correlated with
Higher IL-6 Levels...................................................147
4-20. Higher Global PSQI Scores (indicative of being a
poor sleeper) Are Associated with Higher IL-6 Levels................147
4-21. Detection of IL-6 by Microarray.....................................150
4-22. Detection of TNF-a by Microarray....................................151
4-23. IL-lp Produced by PWM-Stimulated Cells Shows a
Negative Correlation with SOL........................................152
4-24. IL-1 [3 Produced by LPS-Stimulated Cells Shows a
Positive Correlation with Average Length of Nap......................152
4-25. IL-6 Produced by LPS-Stimulated Cells Shows a
Positive Correlation with WASO.......................................153
4-26. IL-6 Produced by PWM-Stimulated Cells Shows a
Negative Correlation with SOL........................................154
4-27. TNF-a Produced by PWM-Stimulated Cells Shows a
Positive Correlation with Average Length of Nap......................155
4-28. TNF-a Produced by PWM-Stimulated Cells Shows a
Negative Correlation with SOL........................................155


4- 29. Means for Estriol......................................................158
5- 1. Changes in Progesterone Levels During Pregnancy
(National Sleep Foundation).............................................178
XVII


TABLES
Table
2-1. Summary of Sleep Deprivation Studies..................................40
2- 2. Summary of Sleep and Pregnancy Studies................................59
3- 1. Study Variables and Source of Data..................................112
4- 1. Demographic Variables................................................126
4-2. Sleep Variables Derived from Sleep Diaries...........................127
4-3. Sleep Variables Derived from Actigraphy..............................128
4-4. Data from Subjective Sleep Questionnaires............................135
4-5. Means and SD for Significantly Different Sleep
Parameters............................................................139
xviii


LIST OF ABBREVIATIONS
i
!
I
i
I
AASM American Academy of Sleep Medicine
ACTH Adrenocorticotropic hormone
ADH Antidiuretic hormone
ANOVA Analysis of variance
ASDQ Additional Sleep Disorders Questionnaire
CRH Corticotropin-releasing hormone
CRH Corticotrophin releasing hormone
CRP C-reactive protein
DSD Daily sleep diary
DTH Delayed type hypersensitivity
EEG Electroencephalogram
EIA Enzyme immunosorbent assay
ELISA Enzyme-linked immunosorbent assay
EP Ectopic pregnancy
ESS Epworth Sleepiness Scale
GDM Gestational diabetes mellitus
GH Growth hormone
GHRH Growth hormone releasing hormone
XIX


HPAHPA Hypothalamo-pituitary-adrenal axis
IL Interleukin
LPS Lippopolysaccharide
NK cells Natural Killer cells
NP Nonpregnant
NREM sleep Non-Rapid Eye Movement sleep
OC Oral contraceptives
OSA Obstructive sleep apnea
OSAS Obstructive sleep apnea syndrome
PBMC Peripheral blood mononuclear cell
PLMD Periodic limb movement disorder
PLMS Periodic limb movements in sleep
PNI Psychoneuroimmunology
PSD Partial sleep deprivation
PSG Polysomnography
PSQI Pittsburgh Sleep Quality Index
PWM Pokeweed mitogen
RA Rheumatoid arthritis
REM sleep Rapid eye movement sleep
RLS Restless leg syndrome
XX


RSA Recurrent spontaneous abortions
SCN Superchiasmatic nucleus
SD Sleep deprivation
SE Sleep efficiency
SES Socioeconomic status
SNS Sympathetic nervous system
SO Sleep onset
SOL Sleep onset latency
SWS Slow wave sleep
Th T helper cells
Thl T helper cells that produce IL-2, IFN-y, and lymphotoxin I (pro-inflammatory)
Th2 T helper cells that produce EL-4, -5, -6, 10 and -13 (anti- inflammatory)
TIB Time in bed
TNF-a Tumor necrosis factor alpha
TSD Total sleep deprivation
TST Total sleep time
WASO Wake after sleep onset
WBC White blood cells
I
i
i
i
xxi


CHAPTER 1
INTRODUCTION
The state of pregnancy is accompanied by numerous physical and emotional
changes. Among these are immunological and hormonal changes, alterations in sleep
patterns, and shifts in the perceptions of stress. These three types of changes, either
solo or in concert with one another, can impact the outcome of the pregnancy either
positively or negatively.
A normal pregnancy is one in which the health of the mother and baby are
secure; there are no complications associated with gestation, labor, or delivery. An
aberrant pregnancy is one in which the health of the mother or child is in danger. This
can take many forms but the most noteworthy are preeclampsia, preterm labor, small-
for-gestational-age babies, or maternal gestational diabetes. During pregnancy, it is
critical that the maternal immune system does not reject the child. The literature
suggests that for a normal pregnancy to occur, there must be a maternal shift to a Th2
or an anti-inflammatory cytokine (immune) profile. Aberrant pregnancies may be
characterized by either a failure of this shift to occur or by an over-secretion of Thl or
pro-inflammatory cytokines. This swing to a pro-inflammatory cytokine profile may
be due in part to sleep disruption. Subsequently, it can have consequences for the
mother and/or child if not attended to in the earliest of stages.
1


Numerous factors that influence a pregnancy and its subsequent outcome have
been identified, including illness (Dunn, 1993; Gibbs, Romero, Hillier, Eschenbach,
& Sweet, 1992; McGregor, French, Lawekkin, & Todd, 2000; Pope, Yoshinoya, &
Persellin, 1982; Pope, Yoshinoya, Rutstein, & Persellin, 1983), stress (Carmichael &
Shaw, 2000; Sheehan, 1998; Wadhwa, Sandman, Porto, Dunkel-Schetter, & Garite,
1993), emotional states such as depression and anxiety (Lederman, Lederman, Work,
Jr., & McCann, 1978), and disturbed sleep (Edwards, Blyton, Kesby, Wilcox, &
Sullivan, 2000; Ekholm, Polo, Rauhala, & Ekblad, 1992; Evans et al., 1995; Suzuki et
al., 1993). There is ample information in the literature about the immunological
consequences of illness, stress, and psychopathology during gestation, but little is
known regarding the potential unfavorable effects of chronic sleep loss/disruption
during pregnancy on maternal health, pregnancy outcomes, and fetal health. The
following background and ensuing study will integrate the existing knowledge of the
impact of sleep loss on immunity, sleep loss during pregnancy, and immune changes
during pregnancy to provide a comprehensive understanding of how sleep changes
during pregnancy and how this can affect maternal health and pregnancy outcomes.
Pregnancy is typically associated with changes in immune function, changes
that may be associated with beneficial or harmful outcomes (i.e., uncomplicated birth
or preeclampsia) (Edwards et al., 2000). Changes in immune function during
pregnancy may also exert negative effects on sleep. For example, the shift toward a
2


pronounced Th2 or anti-inflammatory cytokine profile seen in pregnancy would
intimate that Thl or pro-inflammatory cytokines, such as interleukin 1 (IL-1) and
tumor necrosis factor alpha (TNF-a), would be reduced. IL-1 and TNF-a not only
defend against antigens, but they also are important mediators of the sleep process
(Krueger et al., 1998). Hence, one would anticipate a reduction in sleep during
normal pregnancy as a consequence of this decrease in IL-1 and TNF-a. When
considering the complex interactions among sleep, the physiological state of
pregnancy, and the altered immune functions occurring during pregnancy, it is
possible that exacerbated sleep disruption may occur in pregnancy. This could
ultimately skew the pro-inflammatory cytokine profile leading to an inflammatory
state rather than the desired anti-inflammatory state. For this reason, excessive sleep
disruption and/or deprivation may be a potential contributor to poor maternal health
during pregnancy as well as negative pregnancy outcomes.
It is this ever-shifting balance between a Thl and Th2 immune response and
the ability of cytokines to drive or suppress the inflammatory process that is
important for pregnancy. The clinical data showing association between poor
pregnancy outcomes and high pro-inflammatory cytokine levels strongly suggest that
an appropriate shift to a response favoring anti-inflammatory cytokine secretion
appears to be crucial in maintaining normal gestation and delivery. Thl cytokines
relevant to pregnancy include interleukin 1 P (IL-1 p), interleukin 8 (IL-8), tumor
3


necrosis factor alpha (TNF-a), and interleukin 6 (IL-6); elevated levels of IL-6, IL-8,
and TNF-a are associated with premature labor and delivery (Arck et al., 2001;
Coulam, 2000; Zhang, Wang, Zhao, & Kang, 2000). Failure to effectively switch off
aspects of the inflammatory response during pregnancy also has been associated with
the development of preeclampsia (Dekker & Sibai, 1999). Significantly more TNF-a,
IL-6, and IL-ip are produced by lymphocytes from preeclamptic women than those
produced by lymphocytes from women with normal pregnancies (Hayashi et al.,
2005; Saito et al., 1999; Munno et al., 1999). These data show that the balance of the
maternal immune system in pregnancy is critical for successful pregnancy, and that
increased levels of Thl-type or pro-inflammatory cytokines may be related to the
development of serious complications.
Research Question, Purpose, and Aims
The goal of this study is to clarify the role and influences that sleep and
immunity have upon pregnancy. The premise behind this investigation stems from the
fact that the contribution of these physiological events to pregnancy as interrelated
dynamics has not been considered or evaluated. This project will explore the potential
impact that disrupted sleep has on immune and endocrine function during pregnancy.
Currently, there exist little data as to the consequences of this factor. Examining and
exploring sleep patterns throughout pregnancy and its impact on immune functioning
4


will establish an appreciation of how sleep can influence the outcomes of pregnancy.
The present study measures sleep parameters and cytokine and endocrine parameters
in the 3rd trimester of pregnancy to establish that a relationship between these
variables exists. These then provide the groundwork data to test for the existence of
bidirectional relationships among these variables.
Significance of Study
If excessive disturbed sleep negatively impacts immune profiles during
pregnancy, then this information could be used in a productive manner to offset its
negative consequences. By incorporating behavioral intervention techniques currently
used in insomnia treatment and support groups, strategies could be taught that may
lessen the negative effects of sleep disruption. The present study will lead to future
research that will further substantiate these relationships and test the utility of
behavioral interventions for sleep during pregnancy.
5


CHAPTER 2
BACKGROUND, THEORY, AND
LITERATURE REVIEW
The study of sleep and sleep pathology has grown into a large and diverse
research domain. Areas of research extend from the basic science of sleep to the
genetics of sleep to sleep during space travel. Throughout the following sections,
various areas of sleep and sleep research will be discussed. The initial subject will be
that of sleep physiology. It should orient the reader on the basics of sleep and allow
the reader to follow the subsequent topics. A short introduction to how sleep is useful
for tissue restoration will then be followed by some of the consequences of sleep loss.
Sleep Physiology
Sleep has received the attention of philosophers, scientists, psychiatrists, and
artists for centuries. Once thought to be a passive and inactive state, sleep is now
viewed as an active process in which various metabolic processes, tissue restoration,
memory consolidation, and homeostatic balance is maintained (Adam, 1980; Dement,
1994). Sleep consists of two physiological states termed non-rapid eye movement
(NREM) sleep and rapid eye movement (REM) sleep, defined using physiological
measurements including brain activity via the electroencephalogram (EEG), muscle
6


tone via the electromyogram (EMG), and eye movements via the electroculogram
(EOG) (Carskadon & Dement, 1994; Krueger & Majde, 1990; Krueger & Majde,
2003; Stinton & McCarley, 2000).
NREM sleep is subdivided into four Stages (Stages 1, 2, 3 and 4) that are
defined by amplitude, frequency, and morphology of the EEG. Stages 2, 3 and 4 of
NREM sleep contain synchronous EEG patterns produced by synchronized firing
of large numbers of neurons. As shown in Figure 2-1, these patterns include sleep
spindles, K complexes, and high-voltage, delta frequency, slow waves (Carskadon &
Dement, 1994).
Awake
Stage 1
Stage 2
Stage 3 \rrXT*1' A fl
Stage 4
REM
Figure 2-1. EEG Representation of Sleep Stages
The stages of NREM sleep are comparable to a depth of sleep continuum
(Carskadon & Dement, 1994). Stage 1 sleep, thought to be a transitional phase
between wakefulness and true sleep (Stinton & McCarley, 2000), is the first step
7


into sleep and can last for only a few moments. In this Stage, a person can be aroused
easily and it is observed as the transitional Stage throughout the night (Carskadon &
Dement, 1994). Stage 1 is characterized by decreases in alpha activity with the EEG
comprised of mostly low-voltage, mixed-frequency activity, much of it at a rate of 4-
7 Hz. Rapid eye movements are absent, but slow-rolling eye movements appear and
muscle tonus is relatively high or relatively low (Akerstedt, Hume, Minors, &
Waterhouse, 1994; Stinton & McCarley, 2000; Akerstedt & Nilsson, 2003). Stage 1 is
typically followed by Stage 2 sleep, in which bursts of distinctive, rhythmic 12-16 Hz
waveforms called sleep spindles appear. Interspersed with the sleep spindles are
short-duration, high-amplitude K complexes (Stinton & McCarley, 2000; Carskadon
& Dement, 1994). Both of these waveforms appear on a background of 4-7 Hz
activity (Akerstedt et al., 1994; Akerstedt & Nilsson, 2003). Stage 2 initially lasts
about 10-25 minutes with a gradual progression into Stage 3 sleep. Stages 3 and 4
comprise what is considered slow wave sleep (SWS), as indicated by a predominance
of low-frequency (0.5-2Hz), high-amplitude (>75pV) EEG waves referred to as
delta waves (Carskadon & Dement, 1994; Stinton & McCarley, 2000). The
percentage of delta waves determines if it is Stage 3 or 4. Stage 3 sleep is the shorter
duration cycle of the two stages. Stage 3 is scored if the epoch is 20-49% delta and
stage 4 is scored when there is >50% delta (Carskadon & Dement, 1994).
8


REM sleep is characterized by mixed high-frequency, low-amplitude EEG
waves that more resemble the pattern of active wakefulness in animals such as the
cat, along with bursts of rapid eye movements and diminished muscle tone (Stinton &
McCarley, 2000), while humans exhibit a theta pattern, more similar to Stage 1 or 2
sleep (without spindles or K-complexes) (Rechtschaffen & Kales, 1968). The length
of REM sleep is variable throughout the night. The first episode is often quite brief,
lasting only a few moments, while the final REM period can often last greater than 20
minutes (Carskadon & Dement, 1994). A normal human sleep cycle can be seen in
the hypnogram (see Figure 2-2).
Figure 2-2. A Normal Human Hypnogram
9


Why is Sleep So Important?
Despite various theories, there is no definitive answer yet to the question of
why do we need to sleep. The following sections are intended to shed some light
on this seemingly unanswerable question.
Restorative Function of Sleep
Although sleep is essential for survival, it is only in the past three decades that
the role of sleep in restoring and repairing the human body, including protein
synthesis and mitotic division (Adam & Oswald, 1977), has been identified. The need
for sleep is a basic physiological need and has similar homeostatic properties to
hunger or thirst in that sleep reverses sleepiness (Roth, Roehrs, Carskadon, &
Dement, 1994). There is an intrinsic 24-hour sleep-wake cycle that parallels the
circadian rhythm of the endocrine and immune systems; this cycle governs many of
an individuals behaviors(Roth et al., 1994). Hence, the impact of sleep loss reaches
beyond the immediate realm of just making an individual sleepy; it impacts the entire
biological entity either directly or indirectly, due to the multiple brain regions,
endocrine factors, and various neurotransmitters involved in sleep regulation (Benca
& Quintas, 1997). To illustrate, bodily restitution is thought to occur during SWS as
Growth Hormone (GH) is secreted primarily during these Stages and cortisol
secretion is suppressed (Akerstedt & Nilsson, 2003). Additionally, NREM sleep is
10


thought to conserve energy, cool the body and brain, and promote immune function
(Bryant, Trinder, & Curtis, 2004).
According to Adam & Oswald (Adam & Oswald, 1977; Adam & Oswald,
1983; Adam & Oswald, 1984), sleep is essential for the body to exit a catabolic state
and enter into an anabolic state in which renewal and synthesis of tissues can be
initiated. Many of the endocrine and immune substrates have circadian rhythms that
correspond to the sleep-wake cycle and are essential to optimal health. Cortisol,
considered to stimulate protein catabolism (Adam & Oswald, 1977), is lowest
throughout the sleep period and is secreted to its peak in the early morning hours.
Growth Hormone, an anabolic hormone that increases the synthesis of protein and
mobilizes free fatty acids to provide energy, is secreted primarily during the first few
hours of sleep, specifically during SWS (Adam & Oswald, 1984). The link between
sleep and biological processes initially was indirect: growing children sleep more
than adults; children who were short were poor sleepers; and daily metabolism was
secondarily correlated with REM sleep (Adam & Oswald, 1977). However, recent
data have identified a direct relationship between sleep and biological processes.
Some of the current data address how sleep and sleep loss affect brain energy
homeostasis (Franken & Tafti, 2003), aging and the subsequent circadian phase
shifting observed in the elderly (Tozawa et al., 2003), brain growth factors such as
brain derived neurotrophic factor (BDNF) (Hairston et al., 2004), memory
11


consolidation (Cantero et al., 2003), metabolic function (Bonnet, Berry, & Arand,
1991; Bonnet & Arand, 1996; Bonnet & Arand, 2003; Spiegel, Leproult, & Van
Cauter, 1999), and coronary artery and cerebrovascular diseases (Leineweber,
Kecklund, Janszky, Akerstedt, & Orth-Gomer, 2003; Nilsson, Roost, Engstrom,
Hedblad, & Berglund, 2004; von Kanel & Dimsdale, 2003).
To Regulate Brain Excitability
The question as to the function of sleep is one of the most elusive questions
still to be answered. The theories regarding the function of sleep are as numerous as
the people who study them. Rene Drucker-Colin (Drucker-Colin, 1995) notes five
theories recognized by another researcher (Webb, 1974) in the early 1980s to include
(1) the restorative theory; (2) the protective theory; (3) the energy conservation
theory; (4) the ethological theory; and (5) the instinctive theory. Since then, he notes
numerous others both related and different have been suggested: (1) sleep serves the
purpose of cell maturation during ontogenic development; (2) sleep serves to dispose
energy; (3) sleep serves to control oculo-motor activity; (4) sleep maintains
catecholamine levels; (5) sleep consolidates memory; (6) sleep maintains life; and (7)
sleep cools the brain (Drucker-Colin, 1995). Although these all may have validity,
Drucker-Colin acknowledges that the list could be infinite. He takes a broader
approach and from the General Unifying Approach, believes that we should view
12


sleep as serving multiple functions and try to come up with a unifying concept to
explain how and why sleep encompasses them all (Drucker-Colin, 1995).
His argument, admittedly without much supporting evidence, is as follows.
There are two phases of sleep: NREM (quiet) and REM (active sleep). The oscillating
between the phases allows for the body to reset itself based on activity experienced
throughout the prior day and to permit the organism to function adequately during
waking hours (Drucker-Colin, 1995). The mechanism which accounts for this shifting
back and forth is what he calls the excitostat. The purpose of REM and its
excitable state is to maintain the needs of the brain. Because it cannot sustain this
level of excitability it must switch to SWS for rest. The excitostat determines
how much SWS and REM sleep a person needs based on the previous days activity,
which subsequently accounts for the variability observed in an individuals sleep
patterns.
To Provide a Stress Free Period
Bom and Fehm (2000) discuss and describe how sleep, especially sleep
occurring earlier in the night, can be regarded from a psychoneuroendocrinological
perspective as a stress-free period. The release of adrenocorticotropic hormone
(ACTH) and cortisol by pituitary-adrenal activity follows a circadian rhythm and is
synchronized to the sleep-wake cycle. In typical situations, ACTH/cortisol
13


concentrations are at a minimum in the early hours of sleep coinciding with SWS and
are at their maximum during the second half of the night coinciding with REM sleep.
The authors point out that this pattern of endocrine activity is unique to the state of
sleep and can be modified by various forms of stress (Bom & Fehm, 2000).
They acknowledge that various physical and psychological stressors can
negatively influence the sensitivity of neuroendocrine regulation during sleep, thus
shifting how much and when the nadirs of ACTH/cortisol occur. Data from
Weitzmann and colleagues(Weitzmann, Nogeire, & Perlow, 1974) suggested that
allowing only 3 hours of sleep resulted in reduced plasma cortisol concentrations.
Sleep deprivation and the subsequent consequences resulting from sleep deprivation
will be elaborated upon later. What is important to note here is that the cortisol nadir
concentrations appear to be the essential marker of various states of stress including
strenuous exercise, depression, and normal aging (Bom & Fehm, 2000). Could
pregnancy and the sleep disruption experienced during pregnancy be viewed as a
chronic (albeit definite) stressor that can alter neuroendocrine activity? Or is altered
endocrine activity during pregnancy a causal influence on the sleep disruption
reported? Support for the notion that sleep disruption is a stressor comes from data
showing that IL-1 and TNF-a were found to be significantly higher in blood samples
taken during a night of total sleep deprivation as compared to a regular night of sleep
(Haack, Pollmacher, & Mullington, 2004; Shearer et al., 2001; Uthgenannt,
14


Schoolmann, Pietrowsky, Fehm, & Born, 1995; Bom, Lange, Hansen, Molle, &
Fehm, 1997; Vgontzas et al., 2002). Events that simulate an immune challenge by
stimulating the release of pro-inflammatory cytokines (including IL-1 and TNF-a) are
considered stressors. Since total sleep deprivation showed increased pro-
inflammatory cytokine levels, it can be argued that sleep deprivation is a stressor.
However, the distinction between total sleep deprivation and partial sleep disruption
and what that leads to immunologically and endocrinologically during pregnancy has
yet to be determined.
Costs of Sleep Loss
Sleep disorders, such as insomnia, obstructive sleep apnea, narcolepsy and
restless legs syndrome, are one of the most common health concerns, affecting over
one-third of all Americans (Drobnich, 2005; Rosa & Bonnet, 1993; Rosa, 1995; Rosa,
2005; Poceta, Loube, Kellgren, Bizik, & Mitler, 1999; Society for Neuroscience,
2002). Often, sleep disorders are the primary or secondary source of morbidity and
mortality, such as is the case with cardiovascular disease and mental dysfunction. It is
accepted that there is bi-directional communication between the brain and the host
defense system (Benca & Quintas, 1997), which is why sleep loss is believed to
exacerbate various disease states via various neurochemical pathways. If an
individual chronically goes without enough sleep, degradation in mental function is
15


easily observed (Mitler, 1996; Society for Neuroscience, 2002). In addition to the
detrimental effects on health, the expense to the economy due to lost productivity
resulting from any sleep disorder reaches into the hundreds of billions of dollars per
year (Dement & Mitler, 1993; Mitler, Hajdukovic, Shafor, Hahn, & Kripke, 1987;
Mitler et al., 1988; Poceta et al., 1999; Society for Neuroscience, 2002). The effects
of sleep loss extend far beyond just feeling fatigued after a poor nights sleep. How it
specifically impacts pregnancy will be addressed in this proposal.
Immunity
The Normal Immune Response An Introduction
The immune system is the bodys defense mechanism against a plethora of
potentially pathogenic microorganisms that we encounter on a daily basis. It also is
the regulatory body against tumor growth. The process of maintaining a healthy
physiological state is a continual and ever-challenging task. There are several
subsystems that work in concert to provide protection from antigens. The innate
immune system, beginning with the skin, provides a physical barrier to pathogens.
Those agents that pass through this barrier encounter a non-specific inflammatory
response carried out by immune cells, such as neutrophils, and initiate a primary
immune response via the antigen-presenting cells, such as dendritic cells (Paul, 2003).
16


This system does not require previous exposure or memory of exposure to the antigen
for it to begin working (Hucklebridge & Clow, 2002). It is a standardized, consistent
response that works without the availability of memory, i.e., it will not remember
being re-exposed to an antigen (Elgert, 1996). The innate immune system is our first
line of defense.
However, the primary immune response, otherwise referred to as acquired
immunity, is a much more sophisticated and reactionary response. The characteristics
that define acquired immunity are: (1) specificity, (2) inducibility, (3) diversity, (4)
memory, (5) distinguishing self from non-self, and (6) self-limiting (Elgert, 1996).
The primary response is carried out by various immune cells, including
monocytes/macrophages, lymphocytes (T & B), natural killer (NK) cells, neutrophils
and mast cells. These cells are found in peripheral organized tissues such as the
spleen and lymph nodes. Most of the lymphocytes and macrophages circulate through
the lymph and the blood, scanning and searching for antigens. Once the antigen-
specific lymphocyte encounters an antigen, a cascade of events takes place: antibody
production begins and the expansion and differentiation of antigen-specific helper and
effector T lymphocytes occur (Hucklebridge & Clow, 2002; Paul, 2003). Normally,
the combination of the innate immune response and the primary response is sufficient
to eradicate or control the foreign antigen, although exceptions do exist (Paul, 2003).
17


Of interest to this study are T lymphocytes and the chemical substances
(cytokines) they produce once the immune system is activated. The terminology
discussed in this paper regarding this topic may at times appear contradictory or
confusing. For instance, there are references to pro- and anti-inflammatory cytokines,
as well as to Thl and Th2 cytokines. Although these terms usually refer to the same
concept, newer research and literature has broadened the knowledge of only a decade
ago. When there are references to the terms pro- or anti-inflammatory cytokines,
it means that the authors identified and used these terms in their research. Likewise,
if the term Thl or Th2 is used, then that is what the authors used in their papers.
Typically, the former are in older papers while the latter are from more recent work,
although it may depend on the field of research from which the reference came. For
example, the area of allergy and immunity has focused more on the Thl vs. Th2
profile rather than the pro- vs. anti-inflammatory model. It appears that the Thl vs.
Th2 paradigm is clearer and more specific in understanding how the immune system
may respond to various situations. For the sake of clarity to the literature cited, the
original terms are used in this paper.
As illustrated in Figure 2-3, T helper cells (Th) differentiate into two types
depending on the type of priming they receive: Thl cells produce IL-2, IFN-y, and
lymphotoxin; Th2 cells secrete IL-3, -4, -5, -6, -9, -10, and -13. The focus of Thl
cells is to induce a cellular immune response, while the focus of Th2 cells is to assist
18


B cells in becoming effective antibody secretors (Hucklebridge & Clow, 2002; Paul,
2003). These two arms of the immune system appear to oppose the action of the
other. Physiological stressors, such as pregnancy or sleep disruption, may shift this
balance in an inappropriate way that could prolong or encourage development of
infections.
DC11 his
myebic!
dike
DC2
lymphoid
Hie
IFNy. 1L-2, LI a.
CM-CSF
+other rrtokiitts
IL-3,114 IL-5. IL-6
IL-9, IHO, IM3
+ oiler cytokines
(A ntibody-niediatedl
Bendtzen 1999
Figure 2-3. T-Helper Cells Differentiate Into Thl and Th2 Cells
The cytokines under study fall into two categories: Thl (IL-6 and TNF-a) and
Th2 (IL-4 and IL-10). IL-6 has varying roles depending on whether it is in the
periphery or within the CNS. In the periphery it is a pro-inflammatory cytokine
19


because it mediates innate immunity through inflammatory reactions (Elgert, 1996).
Among its actions is its ability to induce the acute phase response, including inducing
pyrogenic and soporific effects, facilitating neuroendocrine communication,
stimulating the hypothalamo-pituitary-adrenal (HPA) axis and causing ACTH release,
as well as lymphocyte activation and antibody differentiation (Wilson, Finch, &
Cohen, 2002). It has also been shown to have an endocrine role. The production of
IL-6 is suppressed by glucocorticoids and estrogens (Papanicolaou & Vgontzas,
2000). Postmenopausal women undergoing estrogen replacement therapy have
reduced levels of IL-6, indicating a diminished risk of cardiovascular mortality
(Papanicolaou & Vgontzas, 2000). C-reactive protein, a marker of generalized
inflammation, is synthesized and secreted by the liver in response to IL-6 (Meier-
Ewert et al., 2001). Vgontzas and colleagues (Vgontzas et al., 2000) found elevated
IL-6 correlated with sleep disturbance in individuals with sleep apnea. Hence,
elevated levels of IL-6 due to sleep disturbance may be a factor in illness
susceptibility during pregnancy. More recently, it has emerged as a reporter cytokine
for intraamniotic infection (Amtzen, Kjollesdal, Halgunset, Vatten, & Austgulen,
1998). It is produced by many different cells in response to IL-1 and TNF-a (Elgert,
1996). It is also involved in stress modulation, and in its anti-inflammatory role it
inhibits the synthesis of TNF-a and induces expression of IL-1 receptor antagonist
20


and the TNF-a soluble receptor (Wilson et al., 2002). Hence, the levels of these three
pro-inflammatory cytokines are interdependent on one another.
TNF-a is produced by activated macrophages and, to a lesser extent, by
antigen-activated Thl cells, activated NK cells, and activated mast cells (Elgert,
1996). As mentioned, TNF-a is considered a pro-inflammatory cytokine because its
major local effect is to initiate an inflammatory response. In low concentrations, it
can increase expression of adhesion molecules on endothelial cells, increase
neutrophil adhesion, and stimulate macrophages to produce D-l, 11-6 and TNF-a itself
(Elgert, 1996). It also increases vascular diameter leading to increased local blood
flow and systemically, it is a potent pyrogen.
The Th2 (or anti-inflammatory) cytokine IL-4 is involved in the activation,
proliferation, and differentiation of B cells. It inhibits macrophage activation, thus
impeding the production of TNF, IL-1, and IL-6 by macrophages, making it
important in determining the dominant role of Thl versus Th2 profile during
pregnancy. IL-10, the other cytokine under study, is also involved in the suppression
of macrophage, T-cell, and NK cell effector functions. It is a known Thl suppressor
by its ability to inhibit the proliferation and secretion of Thl lymphocytes (Schafer-
Somi, 2003). IL-10 has immunosuppressive properties in that it appears to suppress
rejection of grafts after organ transplantations. This may have relevance to pregnancy.
Since it is anti-inflammatory, elevated levels of IL-10 may be necessary for
21


implantation of the embryo as well as maintenance throughout pregnancy. Studies
have shown that lowered levels of IL-10 at term were associated with higher
preeclampsia evidence of a heightened Thl response (Hennessy, Pilmore, Simmons,
& Painter, 1999).
C-reactive protein (CRP), although not a cytokine, is the most prominent
serum marker of the acute-phase response (Meier-Ewert et al., 2001). It is
synthesized and secreted in the liver in response to cytokines and is an important
serum marker for inflammation (Shamsuzzaman et al., 2002). The role of CRP in the
inflammatory process is still not well defined, but large increases occur during the
acute-phase response. There is also evidence that CRP is present at low levels in
asymptomatic people, which may reflect baseline activity of circulating cytokines
(Meier-Ewert et al., 2001). It has been shown to be elevated in individuals with sleep
apnea and to be an independent marker of morbidity and mortality in patients with
unstable angina (Papanicolaou & Vgontzas, 2000). In individuals with obstructive
sleep apnea (OS A), CRP levels have been reported to be significantly higher than in
subjects without OSA (Shamsuzzaman et al., 2002). Not only is this noted as a
marker of inflammation, but it is suggestive of increased cardiovascular risk.
22


Neural-Immune System Interactions
During the last 25 years, acceptance has grown that elements of the nervous
and immune systems communicate bi-directionally. The brain interacts with the
periphery via the sympathetic nervous system (SNS) and the hypothalamo-pituitary-
adrenal axis (HPA). Data extending back over three decades show that both primary
and secondary lymphoid organs are innervated by sympathetic fibers and that CD4+
T lymphocytes express receptors that bind to norepinephrine (Sanders & Kohm,
2002). This supports the notion that activation of the immune system can influence
sympathetic activity and that norepinephrine can modulate the activity of CD4+ T
cells and B cells (Sanders & Kohm, 2002). Hallmark studies by Besedovsky et al.
(cited in Sanders & Kohm, 2002) have shown that soluble factors such as IL-1 are
secreted by immune cells that stimulate the nervous system and can alter
noradrenergic activity.
The work on these soluble factors, known as cytokines, has exploded in
recent years and is pivotal to this study. Cytokines are messengers secreted by
immune cells that regulate the immune system and modulate the nervous system
(Dunn, 2002). How cytokines affect the brain is summarized by Dunn (2002) and are
as follow:
1. Cytokines can be transported into the brain to a limited extent using
selective uptake systems.
23


2. Cytokines can act on brain tissue at sites where the blood-brain barrier is
weak or non-existent.
3. Cytokines may act directly or indirectly on peripheral nerves that send
afferent signals to the brain.
4. Cytokines can act on peripheral tissues inducing the synthesis of
molecules whose ability to penetrate the brain is not limited by the barrier.
A major target appears to be endothelial cells, which bear receptors for
IL-1.
5. Cytokines can be synthesized by immune cells that infiltrate the brain.
It is the final point that is pertinent to the proposed study. Establishing the
relationship that altered immune function stemming from pregnancy or sleep
disruption can vary cytokine production, which can subsequently exacerbate
infections and sleep patterns throughout pregnancy, is a crucial step.
Sleep and Immune Function
The Role of Neural. Immune, and Endocrine Substances in
Sleep
Despite the accordant thinking that too little is known about sleep and its
processes, an important consideration is that a better understanding of how the brain
works will only come once it is learned what sleep itself does for the brain (Krueger
24


& Kamovsky, 1995). The understanding of how neural, immune, and endocrine
processes work together to modulate sleep will provide important information in this
context. The alterations in sleep during infection have been described in detail
elsewhere (Krueger & Kamovsky, 1987; Opp & Imeri, 1999; Obal, Jr. & Krueger,
2001). In general, many cytokines are involved in physiological sleep regulation. The
expressions of some cytokines are greatly amplified by microbial challenge. This
excess cytokine production during infection induces sleep responses. The excessive
sleep and wakefulness that occur at different times during the course of the infectious
process may result from dynamic changes in various cytokines that occur during the
host's response to infectious challenge. However, in addition to the exacerbation of
sleep during infection, sleep is often disturbed during periods of stress. In addition,
data support the implication that sleep loss encourages vulnerability to infection.
Additionally, the Psychoneuroimmunology (PNI) literature asserts that stress is
correlated with immunosuppression via the HPA axis (Glaser et al., 1987;
Vingerhoets & Assies, 1991; Dunn, 1993) and sleep is often disturbed during periods
of stress.
Several neural-immune substances have been identified to mediate the sleep
response. Typically the function of these substrates can be portrayed both directly and
indirectly either via the HPA axis or the autonomic nervous system (Chang & Opp,
2001). One of these factors, corticotropin-releasing hormone (CRH), is thought to
25


contribute to the regulation of spontaneous awakening in addition to the role that
CRH plays in the bodys response to stressors (Chang & Opp, 2001). CRH is
considered an important component to the sleep-wake cycle because it is the major
hypothalamic releasing hormone of the HPA axis and the ANS and it provides the
organism with a response system that is capable of reacting to stressors of multiple
modalities and on different time scales (Bom & Fehm, 2000; Chang & Opp, 2001),
p. 446). In other words, stressors that have no tissue insult associated with them, i.e.
psychological stressors, are mediated by CRH. An additional relationship has been
proposed by Chang & Opp (2001) that suggests that the cytokine IL-1 acts as a
mediator of sleep responses to CRH release via corticosterone release. When
circulating corticosterone levels were experimentally reduced in rats, IL-1 mRNA
levels were increased several hours later, which was then followed by reduced
waking and increased SWS (Chang & Opp, 2001). Although this relationship remains
opaque, it appears that even in the absence of infection, various stressors can affect
immune and sleep regulation.
The relationship between sleep, the HPA axis, and stress is simplified by the
concept that sleep is important for the recovery from stress. During stressful periods,
the HPA axis secretes an excess of pituitary-adrenal hormones that mediate
susceptibility to disease. If pituitary-adrenal activity is diminished during sleep, the
potential for recovery is increased because, from a psychoendocrinological
26


perspective, early sleep can be regarded as the only period during the day that is free
of stress (Bom & Fehm, 2000). However, HP A activity is only halted during the first
few hours of sleep, during SWS (Bom & Fehm, 2000). Viewing sleep as a stress-free
period implores one to acknowledge that sleep disruption may be a stressor. For acute
stressors, the immune response, via the HPA axis, may be enhanced. In a sleep
paradigm, this is delineated by the observance of increased monocytes and cytokine
levels in the blood of individuals who were sleep deprived for one night (Bom &
Fehm, 2000). However, chronic sleep disruption, often observed in the aged and
depressed, exhibits disinhibited HPA activity and subsequent dysregulation and
reduced cytokine production (Bom & Fehm, 2000). Discriminating between acute
and chronic sleep disruption may ultimately provide answers to many of inconsistent
results thus far reported.
Endocrine substances are important to both the sleep-wake cycle and the
immune system. Melatonin, a hormone produced by the pineal gland, is circadian
rhythm driven by the superchiasmatic nucleus (SCN) of the hypothalamus. Although
the precise role of melatonin in human physiology is poorly understood (Kunz,
Mahlberg, Muller, Tilmann, & Bes, 2004), it nevertheless is recognized as a primary
influence on the internal clock, in that it has phase shifting ability and is thought to
assist with the synchronization of circadian rhythms that have been disrupted, i.e.
jetlag (Kunz et al., 2004). It is also thought to stimulate various components of the
27


immune system, such as enhancing humoral and cellular immune responses (Haimov,
Shochat, & Lavie, 1997; Maestroni, 1993). Specifically, melatonin has been
suggested as regulating IL-4 production in bone marrow T-helper cells (Maestroni,
Hertens, Galli, Conti, & Pedrinis, 1996) as well as interacting with IL-2 in the
modulation of host antitumor immune responses (Lissoni et al., 1992). Lastly,
evidence relating the stimulation of IL-2 secretion during the night with melatonin
suggests that the IL-2 blood concentration could depend in part on the action of
melatonin (Lissoni, Rovelli, Brivio, Brivio, & Fumagalli, 1998).
Progesterone, one of the sex steroid hormones, acts as an immunosuppressive
agent on lymphocytes and influences the onset of SWS (Moldofsky, 1995). During
the menstrual cycle, women experience fluctuations in progesterone levels depending
on which phase they are in luteal (high progesterone) or follicular (low
progesterone). Driver and Baker (1998) reported that healthy young women had
more stage 2 sleep, higher spindle frequency activity, and less rapid-eye movement
(REM) sleep when progesterone predominates in the luteal phase. However, they
also reported that sleep regulatory mechanisms, indicated by the onset to sleep, slow-
wave sleep (SWS), and slow-wave activity, appear to be unaffected by menstrual
phase in women with normal cycles. However, they did find that women with
premenstrual mood symptoms have more Stage 2 sleep and seemingly less SWS and
REM sleep that the normal counterparts (Driver & Baker, 1998). In an attempt to
28


understand progesterones contribution to sleep, Moldofsky et al. (1995) evaluated
immune and endocrine parameters in healthy nonpregnant women during high and
low progesterone phases of the menstrual cycle. Results indicate that onset to SWS is
delayed and the amount of SWS is reduced during the high progesterone phase
(Moldofsky, 1995). In addition to the relationship between SWS and high
progesterone levels, there were also differences in sleep-related NK activity and
progesterone levels: high progesterone levels corresponded to a decline in nocturnal
NK cell activity (Moldofsky, 1995).
IL-1 is a cytokine of particular interest when discussing sleep and immunity
because it is deeply involved in both the regulation of the immune system and
physiological sleep (Covelli et al., 1992; Hohagen et al., 1993). In the historic search
for a sleep-promoting substance, IL-1 turned out to be the frontrunner for this title. It
was shown that muramyl peptides induced production of lymphocyte-activating
factor, also referred to as endogenous pyrogen, and now called Interleukin 1 (IL-1)
(Krueger & Kamovsky, 1995). At that point in time, the process of sleep was thought
to be a function solely of the brain (Krueger & Kamovsky, 1995) and, up until now,
IL-I was presumed to be solely an immune substance. Years later, IL-1 was shown by
Fontana (Fontana, 1982) to be a product of astrocytes; hence IL-1 was a brain product
as well as an immune product (Krueger & Kamovsky, 1995). Thus the current
hypothesis is that IL-1 operates both within the CNS and the periphery creating an
29


interrelationship between the immune and endocrine systems and the sleep-wake
cycle (Moldofsky, 1994). IL-1 activates the secretion of CRH, but it also activates the
HPA axis at all levels, which subsequently increases sleep (Krueger & Kamovsky,
1995). Although the complete function of IL-1 is still under investigation, the role it
plays in sleep and sleep inhibition has initiated great interest in this cytokine.
Since this discovery, a plethora of research exploring the possibility that
various cytokines may be sleep promoting as well as sleep inhibiting has occurred.
Returning to the Moldofsky work (1995), IL-l-like activity showed peaks at midday
(l-3pm) and during the nighttime hours (11-midnight), which the author maintains
corresponded to subjective times of sleepiness. In a study designed to evaluate the
role of B-endorphins and IL-1 [3 during sleep, Covelli and colleagues (1992)
discovered significant differences in IL-1 (3 production in subjects who slept and those
who could not. The two subjects who were unable to fall asleep had no IL-1 [3
secretion during the study. Hence, the authors conclude that normal sleep is
associated with nocturnal rises in IL-1 (3, while disturbed sleep inversely affects IL-1 (3
secretion (Covelli et al., 1992). Supporting data shortly followed showing that IL-lp
incubated in the presence of endotoxin was at a maximum around the time of sleep
onset and in the first few hours of sleep, with a decline as the sleep cycle progressed
(Hohagen et al., 1993). There is also evidence of sleep-inhibiting cytokines. Anti-
30


inflammatory cytokines that inhibit production of IL-1, such as IL-4, IL-10 and IL-
13, inhibit spontaneous sleep (Krueger & Majde, 2003).
The temporal relationship of Growth Hormone (GH) and Growth Hormone
Releasing Hormone (GHRH) with SWS has led to increased attention to their overall
role in the sleep-wake process. GH is thought to be completely sleep-dependent
because whether sleep is delayed, interrupted, or impending, sleep onset will elicit a
pulse in GH secretion (Van Cauter, 1990). The somatotropic axis is now considered a
potent stimulator of sleep (Obal, Jr. & Krueger, 2001). The function of the
somatotropic axis is to regulate anabolism and tissue growth by producing various
hormones, including GH. It also produces two neurohormones: GHRH, which
stimulates GH synthesis and release, and somatostatin, which inhibits GH synthesis
and release (Obal, Jr. & Krueger, 2001). Of note is the sensitive feedback loop this
system incorporates. Both somatostatin and GH, in addition to IGF-1, inhibit GHRH
production. Hence, diminished production of GH, usually corresponding to a decrease
in SWS, would signal an increase of GHRH to subsequently elevate production of
GH.
The majority of the data supporting the role of GH and GHRH in sleep stems
from disordered sleep observed in aging populations (Perras, Marshall, Kohler, Bom,
& Fehm, 1999). Studies assessing how GH and GHRH can affect sleep provide a
clear indication of the importance of these hormones. In a study assessing an
31


intranasal injection of GHRH on endocrine function and sleep in young and old men,
SWS-associated GH elevation was reduced while REM and SWS were enhanced
regardless of age (Perras et al., 1999). It appears that exogenous introduction of
GHRH mimicked endogenous GHRH via the negative feedback loop and
subsequently exerted effects on sleep parameters (Perras et al., 1999).
The Multifaceted Association Between Sleep and
Inflammation
While sleep clearly plays a role in the homeostatic regulation of the
sympathetic nervous system (SNS) (Irwin, Thompson, Miller, Gillin, & Ziegler,
1999) and the immune system (Dinges, Douglas, Hamarman, Zaugg, & Kapoor,
1995; Krueger & Kamovsky, 1995), there is also convincing evidence that not only
does sleep influence mediators of inflammation, including various cytokines and
polymorphonuclear cells, but that inflammatory conditions and mediators modify
sleep as well. Cytokines can amplify their own production via autocrine induction
once the inflammatory cascade has been stimulated, resulting in an upregulation of
the inflammatory process (Wilson et al., 2002). Extensive data by Pollmachers group
(Gudewill et al., 1992; Pollmacher et al., 1993; Pollmacher, Mullington, Korth, &
Hinze-Selch, 1995; Haack, Schuld, Kraus, & Pollmacher, 2001; Mullington, Hinze-
Selch, & Pollmacher, 2001) demonstrate how an injection of endotoxin can elicit
activation of the immune system, similar to a natural bacterial infection, and can
32


suppress REM sleep regardless of the time of day, with the effect being dose-
dependent.
The amount and intensity of NREM sleep that occurs in the course of host
response to infection or during an inflammatory response is thought to indicate an
immune-supportive function for NREM sleep (Mullington et al., 2000). During the
acute phase response, cytokines such as IL-1 and TNF induce somnolence (Krueger
et al., 1998). When the immune system is activated, macrophages and glial cells
produce IL-1, while neurons and astrocytes produce TNF (Krueger & Toth, 1994).
IL-1 and TNF, in turn, induce fever and/or SWS (Krueger & Kamovsky, 1995). A
consistent sleep-wake cycle ensures that white blood cells (WBC) are more capable
of producing IL-1 or TNF, which can induce sleep in times of illness (Krueger et al.,
1998; Moldofsky, 1995) Lastly, a more recent marker used to assess generalized
inflammation is C-reactive protein (CRP). It is made in the liver and its synthesis is
controlled by IL-6 and TNF-a (Meier-Ewert et al., 2001; Yokoe et al., 2003). This
marker is being utilized to predict subsequent myocardial infarction, stroke, and other
inflammatory diseases (Akerstedt & Nilsson, 2003). It also has been identified as a
marker of potential cardiovascular disease in patients with obstructive sleep apnea
(Shamsuzzaman et al., 2002).
Many studies of various inflammatory conditions have assessed subjects for
sleep changes, both subjectively and objectively, and for symptomatology changes
i
33


following modifications in sleep activity. Children undergoing adenoidectomy and/or
tonsillectomy were subjected to pre- and post-operative polysomnography tests (PSG)
(Jain & Sahni, 2002). There was a significant improvement in PSG scores following
surgery for all patients. Evidence that sleep may influence alterations in pro-
inflammatory cytokines levels in patients with allergic rhinitis stems from
information that symptoms appeared to be worse in the first few hours of awakening
(Krouse, Davis, & Krouse, 2002). Additionally, increases in various interleukins,
including IL-lp and IL-5, have been noted in patients with seasonal allergic rhinitis
during the pollen season (Krouse et al., 2002). Recent data showed elevated levels of
IL-lp, IL-4, and IL-10 (deemed proallergic) in allergic subjects and a correlation with
increased latency to REM sleep and shortened periods of REM sleep. Additionally, in
the non-allergic subjects, higher levels of IL-lra, IL-2 and IL-12 (deemed allergy
inhibitory) were found along with a significant increase in latency to sleep onset
compared to the allergic subjects (Krouse et al., 2002). In patients with rheumatoid
arthritis (RA), an autoimmune disease characterized by inflammation (Bourguignon,
Labyak, & Taibi, 2003), sleep has been shown to mediate symptom severity. In
patients given short-term hypnotic therapy, not only was sleep and daytime sleepiness
improved, but morning joint stiffness in these subjects was ameliorated (Walsh,
Muehlbach, Lauter, Hilliker, & Schweitzer, 1996). These data suggest that various
cytokines that mediate inflammatory responses may be influenced by sleep and these
34


same cytokines may subsequently alter sleep independently. Hence, the relationship
between sleep and the immune system can therefore be considered bi-directional. The
utility of this information lies in the connection between sleep and immunity during
the pregnant state. Since complicated pregnancies are associated with a shift toward a
pro-inflammatory cytokine profile rather than an anti-inflammatory profile, then one
would anticipate reductions in REM and SWS sleep.
Sleep Deprivation and Immunity: A Key to Understanding Sleep
Voluntarv/Experimental Sleep Disruption
Voluntary sleep curtailment has become endemic in our society. Along with
the dangers of increased work and driving accidents, chronic sleep loss has health
risks as well. How sleep loss/deprivation can affect immunity is poorly understood.
Few studies involving humans have been conducted and the animal studies are often
not applicable. Understanding the role of sleep in health and disease has come mainly
from studies utilizing various forms of sleep deprivation and primarily in normal,
healthy controls. The few studies that have been conducted attempt to interpret what
happens physiologically, cognitively, and immunologically to a person deprived of
sleep. In addition, these studies hope to provide answers to the complex question as to
why we sleep. Despite the best intentions, studies of sleep deprivation are plagued by
35


methodological problems (Bonnet, 1980). Critics of results derived from sleep studies
note that randomized controlled, double-blinded studies in humans cannot be
performed, nor are the experimental conditions remotely close to routine or natural
(Bonnet, 1980; Dinges et al., 1994). However, studies that are reported suggest that
sleep deprivation does have an effect on immune function and that continued research
is necessary.
Evaluation of the effects of sleep deprivation upon immune function has
generally focused on disorders of excessive daytime sleepiness in healthy adult men
and depressed populations. For instance, one innovative study found elevated levels
of TNF-a in patients with disorders of excessive daytime sleepiness (e.g., sleep apnea
and narcolepsy) compared to controls (Vgontzas et al., 1997). Experimentally-
induced sleep deprivation has been found to alter the diurnal pattern of cellular and
humoral immune functions (Dinges et al., 1995; Heiser et al., 2000; Moldofsky, Lue,
Davidson, & Gorczynski, 1989) and possibly decrease overall immune function
(Redwine, Hauger, Gillin, & Irwin, 2000) in normal adults. More recently an interest
in determining the influence of sleep and sleep deprivation on vaccination response
has been evaluated. Spiegel, Sheridan, and Van Cauter (2002) assessed the effect of
partial sleep deprivation (PSD) on the bodys ability to mount an antibody response to
an influenza vaccination. They reported that PSD reduced antibody production
following vaccination in a sample of healthy men, and that chronic sleep restriction
36


may impair individuals to effectively respond to an influenza vaccination (Spiegel,
Sheridan, & Van Cauter, 2002). Adding to the notion that sleep improves immune
function, Hepatitis A vaccination response in a sleep versus wake protocol was
studied by Lange and her colleagues (2003). They concluded that sleep, compared to
sleep deprivation, on the night after vaccination improves the antibody response
(Lange, Perras, Fehm, & Bom, 2003). An issue that has not been fully explained is
the impact that acute versus extended sleep loss has on immune function. Prolonged
sleep loss can be seen as a stressor, as well as a consequence of various stressors,
including SES, health and/or environmental concerns. Prolonged sleep loss can also
invoke different neuroimmune regulatory mechanisms than those activated by a
single night of wakefulness (Bom et al., 1997). Hence, it must be recognized that
each study assessing sleep deprivation on immunity is utilizing different
methodology. Consequently, the data may be dissimilar but not untrue.
Early data from Palmblad (Palmblad, Petrini, Wasserman, & Akerstedt, 1979)
purport that sleep deprivation reduces both lymphocyte and granulocyte functions,
which they say may be a consequence of excess energy expended to sustain a
prolonged state of alertness. Similarly, Modolfsky (Moldofsky et al., 1989) found NK
cell activity to be decreased after 40 hours of wakefulness, but unlike Palmblads
findings (1979), response to pokeweed mitogen (PWM), an indicator of B cell
function, was increased. Additionally, circulating monocytes, NK cells, and
37


lymphocytes were all observed to increase during one full night of sustained
wakefulness (Bom et al 1997). The hypothesis put forward was that there might be
an interaction of the acute-suppressing effects of sleep with influences of the
circadian pacemaker on numbers of immune cells (Bom et al., 1997). Also, it was
noted that the frequency of blood sampling used in this study provided information
unavailable in previous studies that assayed only 1-2 time points; sampling occurred
every 3 hours for 51 hours (Bom et al., 1997). In a prolonged sleep loss study, NK
activity was observed to increase after 64-hour sleep deprivation in 20 adults (Dinges
et al., 1994). This observance, however, was eliminated after recovery sleep was
obtained.
It is suggested that immune alterations may be associated with a biological
pressure for sleep (Dinges et al., 1994), and that sleep loss could produce an overall
shift in immune function (Dinges et al., 1995; Irwin et al., 1999) that favors pro-
inflammatory cytokines, such as IL-1 (Krueger & Majde, 1995). Cytokine production
is different between sleep and sleep deprivation, and there exists a circadian rhythm
to the production not only of cytokines, but also of the various immune cells that
produce the cytokines (Bom et al., 1997). This understanding guided Bom and
colleagues (1997) to assess the role of nocturnal sleep on normal immune regulation
in a design to assess acute sleep loss rather than excessive sleep loss. Each of the 10
men served as his own control in a two-part procedure. During the first 51-hour
38


session, the subject slept two consecutive regular sleep-wake cycles. In the second
session, the subject was kept awake for 24 hours and then allowed to sleep normally
during the next 24 hours. They found no alteration in the absolute production of IL-
ip and TNF-a; however, the expected decrease of IL-ip and TNF-a during sleep was
blocked when subjects were kept awake. Hence, there was an increase in the
nocturnal production of both cytokines. They note that when the increase in
monocytes is taken into account, there appears to be no influence of sleep on either
cytokine production (Bom et al., 1997). Other studies, however, have found sleep
deprivation to be associated with a delayed nocturnal release of sleep-associated
cytokines, EL-1, IL-6 and TNF-a, with subsequent recuperation of normal levels on
recovery nights (Moldofsky et al., 1989; Redwine et ah, 2000; Vgontzas et al., 1997).
In a study designed to compare the immune effects of both total sleep deprivation
(TSD) and partial sleep deprivation (PSD), dysregulation of sleep regulatory
cytokines and cytokine receptors was revealed; however, only in total sleep loss were
there significant increases in TNF-aRI and IL-6. This suggests that the benefits of
some sleep, in this case two 2-hour naps per day, could be the prevention of the
negative immune changes observed in the TSD condition (Shearer et al., 2001).
Table 2-1 presents a summary of the sleep deprivation studies cited in the
preceding section.
i
39


Table 2-1. Summary of Sleep Deprivation Studies
Author Sample Size (n) Major Findings
Bom et al 1997 10 All male TSD, then recovery. After wake, nocturnal sleep i cell counts Sleep T IL-2 production by T cells, but not IL-1(3, TNF-a or IL-6
Dinges et al 1994 20 7 f; 13 m 64 h of TSD, recovery 48h. Sleep loss is associated with leukocytosis and T NK cell activity
Heiser et al 2000 10 All male TSD, then recovery. >1 NK cell number and monocytes. No change in leukocytes, lymphocytes, B cells or cortisol.
Irwin et al 1999 17 All male Baseline, then 4h PSD. t in circulating catecholamines. No change in circulating IL-2
Lange et al 2002 18 All male 3 n of either IL-2 or placebo. At highest dose, i circulating lymphocytes; T IL-4. IL-2 may induce shift to Th2 profile
Moldofsky et al 1999 10 All male Baseline, then 40 TSD. Rise in nocturnal response of lymphocytes to PWM delayed by SD.
Palmbald et al 1979 12 All male 64h TSD, then recovery for 1 night. PHA-induced DNA synthesis of blood lymphocytes -l after 48h SD.
Redwine et al 2000 31 All male Baseline, then PSD (4h). In PSD nocturnal secretion of IL-6 was delayed until sleep onset.
Shearer et al 2001 58 All male 2 baseline days, randomly assigned to 4 d of PSD or 4 d TSD. TSD lead to T in plasma sTNFRI and EL-6, but not PSD.
Spiegel et al 2002 25 All male 1 lof 25 Ss 6 nights of PSD (4h), then 7nights of 12h sleep. PSD at time of vaccination i response of ab production to flu.
Vgontzas et al 1997 41 lOf; 31m OSA, narcolepsy, IH and controls. PSG In. TNF-a T in OSA and narcoleptics. IL-6 t in OSA
Numerous immune alterations have been observed in depressed patients,
including over-activated T cells, higher production of cytokines such as IL-lp and IL-
6, and diminished NK activity (Heiser et al., 2000). Sleep deprivation has been
40


suggested as a possible treatment for depression because of its immune-altering
effects (Heiser et al., 2000). A possible hypothesis purported was that since the
biological parameters after sleep deprivation correlate negatively with those in
depression, there could exist a practical relationship between depression and sleep,
and their influences on host defense, particularly with regard to cortisol (Heiser et al.,
2000). This same hypothesis, that disrupted sleep alters immune function, may be
applicable to the state of pregnancy, and may account for some of the immunological
changes reported in poor pregnancy outcomes, such as preeclampsia and preterm
birth (Dekker & Sibai, 1999; Sacks, Studena, Sargent, & Redman, 1998; Wadhwa,
Sandman, & Garite, 2001).
Insomnia and Immune Functioning
Understanding immunity in chronic clinical insomnia patients should provide
a clearer understanding of the various immune alterations one might anticipate from
prolonged sleep loss. However, there are sparse data on this relationship, as most of
the data have come from studies assessing depression (Savard, Laroche, Simard,
Ivers, & Morin, 2003), HIV/ALDS (Reid & Dwyer, 2005), or cancer patients
(Theobald, 2004). The primary issue is that there is no specific definition of insomnia
and it can vary dramatically from person to person. However, despite this limitation,
insomnia does provide a useful model for assessing sleep disruption and immunity,
41


even though there are currently only a couple published studies assessing primary
insomnia and immune function. Savard and colleagues (2003) studied patients with
and without insomnia to determine the effect that sleep disruption has on WBC
subsets, NK cytotoxicity, and 11-1(3, IL-2, and IFN-y production. They found that the
patients with insomnia had lower counts of lymphocyte populations but no
differences in WBC count, NK cell activity, or cytokine production. The other study
investigating insomnia and immunity looked at Thl/Th2 balance (Sakami et al.,
2002). This group of investigators reported that men with insufficient sleep or
difficulty initiating sleep had a shift toward Th2 dominance, indicated by higher
levels of IL-4. Other reports (reviews) exploring insomnia and immune functioning
evaluated subjects with secondary insomnia, that is sleep problems stemming from
physical illness or disease (Reid & Dwyer, 2005; Theobald, 2004). These reviews
affirm that the disease states produce an elevated rate of insomnia, but how this
specifically affects immunity in these populations is still unclear. By assessing
immune function in pregnancy, a time when insomnia is experienced, a better
understanding of how immune function is altered may become apparent.
Pregnancy
Pregnancy has ubiquitous endocrinological, physiological, and immunological
changes associated with it. All are important to the health of the mother and the fetus.
42


The breadth of information available on each of these subtopics is beyond the scope
of this paper; however, each will be sufficiently described to enforce its importance to
the proposed study.
Endocrinology of Pregnancy
Once conception occurs, numerous endocrine changes take place on an
almost-daily basis. About 8-10 days after conception, implantation occurs (Taylor,
Lebovic, & Martin-Cadieux, 2001), which involves a complex series of steps
controlled by secretions of the ovary and the blastocyst, including steroids and growth
factors (Porterfield, 2001). The proper development of the placenta is crucial to the
effective communication between the mother and the developing fetus (Taylor et al.,
2001). The bi-directional communication between the mother and the fetal-placental
unit relies on various factors, but of interest to this study are the hormones
progesterone and estriol, an estrogen. Progesterone is the most described and
understood hormone during pregnancy and is thought necessary for the maintenance
of the pregnancy. It is also the primary contributor to whether uterine quiescence is
present throughout gestation (Buster & Carson, 2003). Maternal serum levels increase
as pregnancy progresses and secretion sites of progesterone increase to include some
from the fetal-placenta unit (Buster & Carson, 2003; Porterfield, 2001; Taylor et al.,
2001). In women with progesterone concentrations under 10 ng/ml, abortion occurred
43


in about 80% of the pregnancies (Buster & Carson, 2003). Estriol, a hormone
originating almost exclusively from the placenta (Buster & Carson, 2003), also
increases as the pregnancy moves toward term, but is interestingly minimal in its
production in nonpregnant women (Porterfield, 2001). It plays a significant role in
augmenting uterine blood flow and in the timing of labor stimulation (Buster &
Carson, 2003). Low maternal concentrations have been associated with fetal and
placental development problems including fetal abnormalities and congenital
derangements (Buster & Carson, 2003; Taylor et al., 2001). Thus, evaluating estriol
levels, particularly in the 3rd trimester, is extremely important.
Physiology of Pregnancy
While it is apparent that pregnancy induces easily observed external changes
in a woman, pregnancy also is associated with various internal physical changes.
These changes are often a result of the developing fetus and alterations in endocrine
patterns. Cardiovascular changes comprise the majority of the variations observed.
These include heart rate and stroke volume increases, increased cardiac output, and
increases in vascular volume (Porterfield, 2001). Respiratory changes include a
reduced functional residual capacity (volume of air in the lungs at the end of a quiet
expiration) and a decrease in residual volume (volume remaining at the end of a
maximal expiration). The effects of progesterone on physiological changes are
44


clearest in respiration. Progesterone acts directly on the CNS to lower the set-point
for regulation of respiration by carbon dioxide, hence increasing ventilation
(Porterfield, 2001). Lastly, there are renal changes that come about during pregnancy.
Most notable is water retention, which is caused by increases in the set-points for
antidiuretic hormone (ADH) and thirst (Porterfield, 2001).
Immunology of Pregnancy
The role of the immune system in pregnancy is becoming clearer with time. It
appears immunological factors, cytokines in particular, are extremely important in the
success or failure of implantation, placental development, and timing of labor (Wood,
1997; Chegini & Williams, 2000; Dudley, 1996; Hill, 2000). During the narrow
implantation window, the amount of cytokine expression is correlated with success or
failure of this process (Dudley, 1996). IL-6 has been shown to be weakly expressed
during the proliferative phase, but highly expressed during this implantation window
suggesting a role for IL-6 in preparing tissue for implantation (Chegini & Williams,
2000). Following implantation, placental development is reliant upon a plethora of
factors all working in concert. Increased levels of cytokines/growth factors, such as
IL-1, IL-6, and TNF-a, are observed during placental development and are thought to
derive from the large numbers of macrophages that populate placental stroma.
45


Although unclear if they play a role in placental and fetal growth, they are considered
neutral at worst (Wood, 1997).
The knowledge regarding the orchestrated balance of cytokine synthesis and
function during pregnancy is evolving on a regular basis. The construal of a working
system that explains success or failure of pregnancy is exciting. The Thl (pro-
inflammatory) vs. Th2 (anti-inflammatory) working model has been undeniably
helpful in understanding immune functioning, particularly in the murine model
(Dudley, 1996). Although readily applicable to the mouse, it is not as clear-cut in
humans. EL-6, for instance, is listed among the pro-inflammatory cytokines because it
is a mediator of host response to tissue damage; but it has properties that make it
associated with a Th2 (anti-inflammatory) state because it increases IL-4 and IL-10
(Hunt, 2000). Despite the prominence of pro-inflammatory cytokines such as TNF-a,
IL-ip, and IL-6 at the beginning and the end of pregnancy, it is currently accepted
that Th2-type dominance must occur for a successful pregnancy and that this
regulation is overseen by estrogen and progesterone (Hill, 2000; Hunt, 2000).
Progesterone has been shown to subdue TNF-a transcription and promote
lymphocyte production of Th2-type anti-inflammatory cytokines (Hunt, 2000). In
women with recurrent pregnancy loss, peripheral blood mononuclear cells (PBMCs)
were more likely to secrete IFN-y, which has been shown to be toxic to embryo
development in mice, and less likely to secrete IL-4 and IL-10 (Hill, 2000).
46


The contribution that cytokines make in the timing of labor is of great interest
because there is disagreement as to the level of importance that cytokines play in
parturition. Some believe that an infection is the stimulatory agent that augments the
inflammatory response, increasing pro-inflammatory cytokines such as TNF-cx and
IL-1, and subsequently resulting in preterm labor (Challis & Lye, 2003; Dudley,
2000). However, others purport that the increased cytokine production is merely a
result of the labor process and cytokines are only messengers (Dudley, 2000).
Regardless of which theory is correct, the role of cytokines in parturition is
established in that too heavy of a concentration of pro-inflammatory cytokines prior
to reaching full term is associated with preterm delivery (Challis & Lye, 2003;
Dudley, 2000). The essential message is that the interplay between the immune and
endocrine systems is crucial to a healthy and successful parturition from beginning to
end.
Pregnancy and Sleep
Understanding Insomnia
Insomnia is a frequent complaint in the general population. Usually it is a
symptom of life events, stress, or physical disruptions, which are usually transient.
Sometimes the symptoms endure, occurring several times a week and lasting for
47


months or years; insomnia is then described as chronic. Epidemiological studies have
described the prevalence of insomnia without restrictive criteria, such as frequency or
severity, to be approximately 33% worldwide (Ohayon, 2002). This suggests that a
third of the adult population have experienced some difficulty falling asleep,
maintaining sleep, or had early morning awakenings. When criteria are utilized with
these same individuals, the number drops to approximately 10% (Ohayon, 2002;
Ohayon & Roth, 2003; Roth & Roehrs, 2003). Insomnia is therefore not a rare sleep
disorder. According to the International Classification of Sleep Disorders and the
Diagnostic and Statistical Manual of Mental Disorders, 4th Edition (American Sleep
Disorders Association, 1997; American Psychiatric Association, 1994), the definition
of insomnia is as follows: a symptom complex consisting of difficulty falling asleep,
or maintaining sleep, or nonreffeshing sleep, in combination with some daytime
dysfunction.
Although the American Academy of Sleep Medicine (AASM) acknowledges
sleep during pregnancy as an associated sleep disorder, few epidemiological studies
recognize pregnancy as a category that can contribute to insomnia symptoms.
According to the literature, there are six major diagnostic categories that have been
established for chronic insomnia: medical, psychiatric, circadian, behavioral,
pharmacologic, and primary sleep disorder (Roth & Roehrs, 2003). It can be argued
that pregnancy could be a medical condition as it has become medicalized. To
48


illustrate, the process of delivery is often an appointment that is scheduled around the
physicians schedule, not necessarily when the baby is ready to be bom. Nevertheless
pregnancy is not often considered a medical condition. It is a normal physiological
state for women. Also, the literature does not mention pregnancy as a possible
medical condition that could lead to insomnia. Instead, the medical conditions
referred to consist of pulmonary diseases such as asthma or chronic bronchitis or
rheumatic disease, and physical diseases like arthritis, heart disease or back pain
(Ohayon, 2002).
The insomnia that occurs in pregnancy is often a result of many of the
aforementioned categories, such as back pain or RLS (restless leg syndrome). It is
therefore not possible to lump pregnancy into one of these categories. Moreover, all
of the potential contributors are transient in that there is a definite endpoint. The
physical discomfort, the pregnancy-associated disrupted breathing or RLS, the
hormonal fluctuations all should end with delivery of the baby. This is not addressed
by any of the six groupings. The literature indicates that sleep experienced by
pregnant women, particularly in the 3rd trimester, meets the diagnostic criteria for
insomnia yet little data exist on how to address this from a clinical perspective. In all
of the above categories, pharmacological treatments are often the first choice, but this
is not desirable for most pregnant women. Evaluating the use of behavioral sleep
interventions in the pregnant population has been suggested (Santiago, Nolledo,
49


Kinzler, & Santiago, 2001) but has yet to be explored. These interventions would
include improving sleep hygiene; practicing relaxation techniques; adapting stimulus
control techniques such as using the bed only for sleeping and reducing disruptive
cues in the bedroom, such as noises, lights or TV; and getting out of bed when not
sleeping for a prolonged period of time. These techniques could improve the
insomnia experienced by the pregnant woman and reduce her susceptibility to
infection, irritable mood swings, diminished memory function, and depression
(Martin & Ancoli-Israel, 2002).
Restorative Function of Sleep During Pregnancy
There exists a restorative hypothesis regarding sleep during pregnancy. Sleep
is thought to promote physical and mental restoration (Adam, 1980; Adam & Oswald,
1977). According to Richardson (1996), knowledge of the alterations in metabolism
and arousal that occur as a result of pregnancy is an important consideration.
Pregnancy is a high metabolic effort situation and much of the maternal energy is
directed toward the fetal-placental component in a time-dependent manner
(Richardson, 1996). There is a shift in the favoring of the anabolic activity, beginning
with the mother and ending with the fetus. This physiological shift is timed to
facilitate the appropriate growth and maturation of the fetus, as well as to be
protective (Richardson, 1996). Additionally, expectant mothers describe their activity
50


levels as more passive. This may serve to conserve energy at a time of high
expenditure. Augmenting this theory, despite the evidence that sleep disruption does
occur, much of the pre-existing sleep architecture is left fairly intact during the course
of pregnancy, although there have been some studies showing alterations in SWS and
in Stage 1 at various points of gestation (Hertz et al., 1992; Schorr et al., 1998). This
suggests that there may be an adaptive process to the increased wake-after-sleep onset
observed during pregnancy. The mother becomes more adept at acquiring
consolidated, albeit shorter sleep periods, which may be conducive to the pending
newborns requirements.
Pregnancy and Alterations in Sleep
The sleep architecture of pregnant women is markedly different from their
nonpregnant counterparts (Hertz et al., 1992; Lee, Zaffke, & McEnany, 2000b; Lee,
McEnany, & Zaffke, 2000a; Mindell & Jacobson, 2000). From conception through
delivery, sleep is distinctly altered in the pregnant woman. Sleep quality and quantity
changes vary by trimester, while daytime vigilance undergoes inconsistent variations
that vary from woman to woman. Although there is agreement within the professional
sleep community (AASM, American Academy of Sleep Medicine) of a pregnancy-
associated sleep disorder, most of the data are inconsistent and contradictory. The
discrepancies stem primarily from changes in technology and in the understanding of
51


the sleep process that have occurred over time. Early studies were able to suggest
only crude estimations of sleep changes due to limitations of the EEG recording
equipment used (Branchey & Petre-Quadens, 1968; Karacan, Heine, & Agnew H,
1968), while recent studies are able to delineate more accurately between Stages and
analyze data more in depth (spectral analysis) (Brunner et al., 1994). Unfortunately,
much of the published data resulting from subjective reports of sleep (e.g., sleep
diaries or questionnaires) have utilized no objective measures to corroborate the
results. It is only since the incorporation of various types of unobtrusive, objective
recording equipment that subjective reports of sleep have been validated.
In 1968, Branchey and Petre-Quadens conducted one of the earliest studies
objectively assessing sleep architecture in women. With unsophisticated equipment,
57 nocturnal sleep recordings were conducted on 17 pregnant women. The
presentation of data is significantly different than what is seen today, but the overall
results reveal an increase in paradoxical sleep (analogous to REM sleep) during the
8th month of pregnancy with a clear-cut decrease in the final 3-4 weeks of
pregnancy (Branchey & Petre-Quadens, 1968). Another landmark study (Karacan et
al., 1968) investigated sleep patterns during late pregnancy. Although the sample size
was small and the population homogenous (all were middle-class Caucasians), the
concept that sleep was markedly altered during the last trimester of pregnancy was
gaining affirmation from the scientific community. Although the equipment and
52


scoring procedures were rudimentary, several of the parameters shown altered in
current studies were also shown to change in these earlier studies. Sleep onset latency
(SOL), for example, was shown to be considerably longer in women in the last month
of pregnancy compared to controls, and a complete reduction in Stage 4 sleep was
seen in 57% of the subjects. In addition, the number of awakenings was significantly
higher in the last month of pregnancy when compared to nonpregnant controls
(Karacan et al., 1968). These data were corroborated by a later study by the same
author (Karacan, Williams, Hursch, McCaulley, & Heine, 1969). The authors state
that sleep disturbances are greater than they initially reported and can profoundly
affect the womans physiological disposition, perhaps to the point of inducing
disease (p.933).
Recent objective studies have extended, clarified, and contradicted the initial
findings of the 1960s. Driver and Shapiro conducted PSG studies in five women
throughout their pregnancies (Driver & Shapiro, 1992). Although the sample size was
very small, the repeated recordings of the same women assist in validating the results.
In addition to the original studies, increases in awake times were reported, primarily
due to longer time spent awake following an arousal. This suggests that these women
could have experienced sleep maintenance insomnia; however, no pattern suggesting
an insomnia-like sleep pattern emerged. Data also suggest that SWS is increased, with
Stage 4 elevating significantly from 1st trimester to 2nd and 3rd trimesters. REM sleep
53


significantly decreased from 1st trimester to 3rd trimester. This is only partially in
opposition to Branchey & Petre-Quadens results. While Driver and Shapiro recorded
5 women during weeks 28-39, they failed to report which week of pregnancy the
women were actually in at the time of recording. This can significantly confound the
data as a variety of changes have been reported to occur throughout the 3rd trimester
alone. Branchey & Petre-Quadens on the other hand, studied women during weeks
33-36 and 37-40 and reported the number in each week (Branchey & Petre-Quadens,
1968). This gives greater credence to the conclusions made by these authors.
A similarly conducted study by Brunner et al. investigated sleep during the
course of pregnancy in 9 women (Brunner et al., 1994). Data were recorded for 2
consecutive nights during each trimester. Unlike the Driver and Shapiro (1992) study,
data collection periods were smaller, with the 3rd trimester recordings occurring
during weeks 32-35. And unlike Branchey and Petre-Quadens (1968), no data were
collected during the final few weeks of pregnancy. Despite these methodological
differences, results were similar in that there was an increase in wake-after-sleep
onset mostly in the 3rd trimester; however, they showed a decrease in REM sleep over
the course of the pregnancy contrary to some of the Branchey & Petre-Quadens data.
Consistent with the Driver & Shapiro (1992) study, sleep fragmentation was
suggested by the data.
54


The most recent inpatient PSG study of pregnant women was carried out by
Schorr et al. (1998), whereby 4 women were studied once during each trimester and
compared to age-, weight-, and race-matched nonpregnant controls. Considering the
small sample size, significant differences were observed between the two groups on
measures of SWS: pregnant women experienced much less SWS than the control
women. But unlike the previously mentioned studies, no differences in REM sleep
were noted among or between the groups. And unlike the Driver et al. study, there
were no differences seen across pregnancy, only when compared to controls.
A study utilizing ambulatory polysomnography was carried out in 45
nonpregnant women. Thirty-three of these women conceived and were subsequently
studied throughout pregnancy (Lee et al., 2000b). In conjunction with the PSG data,
subjects completed a 7-day sleep diary just prior to and after the home monitoring.
Data show that total sleep time and sleep efficiency decline during pregnancy;
however, there was a difference in one sleep parameter between nulliparous and
multiparous women: less TST during pregnancy was found for the nulliparous
compared to the multiparous
The small number of studies and the vast methodological differences among
the studies described provide inadequate evidence of the changes that occur in sleep
during pregnancy. More objective studies are warranted in order to clarify this
conundrum. At this point it should be clear that fragmentation of sleep and decreases
55


in total sleep time and sleep efficiency are normal occurrences during the gestational
period (Hertz et al., 1992; Lee et al., 2000b; Lee et al., 2000a; Mindell & Jacobson,
2000). Subsequently, sleep disruption, characterized by arousals and awakenings,
becomes more severe as pregnancy progresses. Objective data supporting this have
been sparse and inconclusive, but many subjective and/or semi-objective studies have
likewise confirmed this notion. A caveat of insomnia research is that the perception of
the sleep by the individual is as crucial as the objective data (personal
communication, R. Bootzin). Hence, subjective reports can be useful in detecting
sleep disturbances, particularly during pregnancy. Data from a subjective survey of
sleep disturbance (Schweiger, 1972) reported generalized sleep disruption in 68% of
the sample. Methodologically, this study asked 100 women, who were at least 38
weeks pregnant, retrospective questions about their sleep during pregnancy; a
psychiatric screening test was also administered. Only basic descriptive data were
reported, including reasons for sleep alterations. Despite the fact that the author
acknowledged that this type of data tends to be unreliable, he used no objective
measures to confirm the data. He cites, subjective ideas on sleep were checked
against electroencephalogram (EEG) recordings by Lewis who found that patients
underestimate their total sleep time and overestimate both sleep-onset time and the
number of awakenings (Schweiger, 1972). However rudimentary the methods, the
data still support the fact that sleep is disturbed during pregnancy.
56


In a similar study (Suzuki, Dennerstein, Greenwood, Armstrong, & Satohisa,
1994), 192 pregnant Japanese women in various stages of pregnancy provided
subjective reports of their sleep patterns. Results indicate that 88% of the sample
reported disturbed sleep from their normal experience, with most of the disturbance
occurring in the 3rd trimester. Even though no significant differences were observed,
the data resemble other studies indicating a worsening of sleep in the 1st trimester, a
move toward normalization in the 2nd trimester, and insomnia-like sleep occurring in
the 3rd trimester. Since the primary sleep disturbance occurs in the 3rd trimester, the
understanding of the sleep-wake cycle during late pregnancy was the goal of
Shinkoda et al. (Shinkoda, Matsumoto, & Park, 1999). They had 4 women complete
sleep logs and wear actigraphy watches for a 20-week period several weeks prior to
delivery and 12 weeks postpartum. Data suggest poorer sleep efficiency and longer
time spent awake after sleep onset as the 3rd trimester progresses, but nothing was
statistically significant. Taking into account the small sample size, power may have
been extremely limited.
To assess both expectant parents sleep perceptions during the 3rd trimester,
Elek et al. (Elek, Hudson, & Fleck, 1997) used instruments consisting of fatigue
scales, activity diaries, and actigraphy administered for 4 days once each month. The
only data providing significant results came from the visual analogue fatigue scales -
expectant mothers reported increasing levels of fatigue during the 3rd trimester, which
57


is consistent with the previously addressed studies. Despite the methodological
concerns, the notion that perceptions of fatigue can be an important factor to consider
during pregnancy should be taken seriously, as individualized interventions could be
implemented.
Although not considered methodologically ideal, purely subjective studies
often provide a beginning point to further inquiry. It is often a cost-effective means to
ascertain pilot data for future study. Two such studies assessed the primary reasons
for sleep disturbance during pregnancy. The need to urinate was the main complaint
found in a secondary analysis by Baratte-Beebe & Lee (Baratte-Beebe & Lee, 1999)
of women in childbearing years before and during pregnancy. This increase is
primarily a result of bladder compression by the uterus, primarily during the 1st
trimester, and the descending fetus pressing on the bladder during the 3rd trimester
(Baratte-Beebe & Lee, 1999). The second most common complaint was nighttime
awakenings. Mindell & Jacobson (Mindell & Jacobson, 2000) questioned 127 women
at one of four points during pregnancy about their sleep habits and sleep disturbances.
Consistent with other studies, nighttime awakening was the most common sleep
disturbance. In support of an insomnia-like sleep disturbance, a significant number of
women in this sample reported difficulty staying and/or falling asleep as well as
waking too early. Also, over 40% reported thoughts as the cause for these
problems, another indication of an insomnia-like disorder. Based on these subjective
58


reports, the next step would be to understand why nighttime awakenings are increased
physiologically.
Table 2-2. Summary of Sleep and Pregnancy Studies
Author Sample Size (n) Major Findings
Branchey & Petre-Quadens 1968 17 1-3 nights during 10-40 weeks of pregnancy. T in Paradoxical (REM) sleep
Brunner et al 1994 9 PSG 2n each trimester. T in WASO X in REM. 4 in power density during the course of pregnancy
Driver & Shapiro 1992 5 PSG throughout pregnancy. No X in stage 4. SWS T later in pregnancy. REM X late in pregnancy compared to controls.
Hertz et al 1992 12 P 10 NP 3rd trimester In. P women T WASO, X SE. % REM X and % Stage l.T
Karacan et al 1968 7 EEG and EOG in late pregnancy for 3n. t SOL and awakenings, shorter sleep time and i Stage 4
Karacan et al, 1969 13 P 13 NP EEG and EOG in late pregnancy. SOL is t, sleep is shorter and X in S 4 sleep
Lee et al 2000 45 Ambulatory PSG 2n. Women served as their own controls. T in TST and awakenings; X deep sleep. SE X than prepregnancy
Mindell & Jacobson 2000 127 Evaluated at one of 4 points during pregnancy via questionnaire, t awakenings, difficulty falling asleep. T in TST and naps by end of pregnancy
Schorr et al 1998 4 P 4 NP P women had 1 PSG each trimester. P X in SWS, T alpha-wave intrusion. X % of TST
Pregnancy and Sleep Disorders: Is Pregnancy a Risk Factor
for RLS?
Restless legs syndrome (RLS) is acknowledged by the AASM and is
classified as a sleep disorder in the International Classification of Sleep Disorders
59


(1990). The symptoms associated with RLS involve unpleasant leg sensations that are
most often bilateral and often described as a creeping or crawling sensation; motor
restlessness and relief with movement; worsening or exclusive presence of symptoms
at rest; and symptoms worsening at night (Lee, Zaffke, & Baratte-Beebe, 2001;
Phillips et al., 2000; Pien & Schwab, 2004; Zucconi & Ferini-Strambi, 2004). There
are two forms of RLS: primary and secondary. The primary form occurs in about 60-
80% of all cases and the cause is unknown. The secondary form or RLS-associated
conditions include neuropathies, Rheumatoid arthritis and pregnancy (Zucconi &
Ferini-Strambi, 2004). Mussio-Foumier and Rawak were probably the first to note the
secondary form in pregnancy in 1940, as described by Manconi (Manconi et al.,
2004).
The epidemiological data indicate a range anywhere from 2.5 to 15% for RLS
in the general population (Hening et al., 2004; Zucconi & Ferini-Strambi, 2004). RLS
increases in frequency as the sample ages and occurs more frequently in women
(Zucconi & Ferini-Strambi, 2004). The diagnosis is based primarily on clinical
criteria; however, hypotheses regarding the cause of RLS, particularly in pregnancy,
have ranged from psycho-hormonal to psychomotor-behavioral to iron metabolism
(Manconi et al., 2004). The data describing the frequency of RLS during pregnancy
come from four dated epidemiological studies that used varying definitions of RLS
60


(Manconi et al., 2004). It was not until 1995 that a standard definition for RLS was
adopted and incorporated into practice.
Despite this early inconsistency in the prevalence rates, it is reasonable to
presume that pregnant women are at a 2-3 fold risk of developing RLS compared to
nonpregnant women, particularly in the 3rd trimester (Manconi et al., 2004; Suzuki et
al., 2003). The current thinking for the strong correlation between pregnancy and
RLS is based on three hypotheses as reported by Manconi (2004): (1) hormonal,
specifically with reference to prolactin, progesterone, and estrogen levels; (2)
psychomotor behavioral, with a focus on sleep habits and anxiety during the 3rd
trimester; and (3) metabolic, ascribing RLS to a reduction in folate or iron blood
levels.
RLS is infrequently recognized by physicians attending to pregnant women
(Pien & Schwab, 2004); however, the few studies that have assessed RLS in
pregnancy suggest that rates of occurrence are consistent and higher than reported in
the general population. Suzuki et al. (2003) reported a 19.9% prevalence rate for
pregnant women in Japan; Lee et al. (2001) reported a 23% prevalence rate for
women in their 3rd trimester; and Hedman et al. (2002) reported a 22% prevalence
rate in the 3rd trimester. Based on the previously mentioned hypotheses and the high
prevalence rates, Manconi and colleagues (2004) state that pregnancy may be a risk
factor for RLS.
61


Understanding Sleep in Pregnancy via Animal Models
A few studies using animal models to assess sleep during pregnancy have
been reported. As early as 1970, information was being gathered on the sleep patterns
of animals during pregnancy due to the acknowledgement that sleep was significantly
disturbed in humans (Branchey & Branchey, 1970). By continuously monitoring
sleep in controlled environments before delivery and into postpartum, clearer
information about sleep and wake patterns was acquired. Analysis revealed that as
early as 15 hours before delivery, sleep patterns in rats were modified, reflecting
increased wake time and a marked decrease in REM sleep (Branchey & Branchey,
1970). Although no hormonal data were collected from these animals, the authors
contend that the observed sleep changes are most likely due to dramatic hormonal
changes that occur near or at term. A protocol by Kimura et al. hoped to further
elucidate the sleep-related mechanisms occurring during pregnancy (Kimura, Zhang,
& Inoue, 1996). Two groups of female rats (pregnant and nonpregnant) were
implanted with EEG recording equipment and measured for approximately 3 weeks.
Results indicate that significant increases in sleep, primarily NREM sleep, occurred
in the pregnant rats. There were also similar fluctuations throughout the pregnancy
similar to that in humans more NREM sleep during the early and late periods than
during the midperiod of pregnancy (Kimura et al., 1996). REM sleep was also shown
to increase over the course of gestation compared to the nonpregnant rats, but this
62


was attributed primarily to an increase in the number of episodes rather than time for
each episode. The authors heed caution in comparing animal studies to human
studies, even though sleep has been shown to be disturbed in pregnancy in both
animal and human models.
Mechanisms of Sleep Disruption in Pregnancy
Alterations in sleep patterns during normal pregnancy are typically described
separately for each of the three trimesters. Across the gestational period, women
initially (first trimester) are more fatigued and take more naps, but they also
experience a decrease in total sleep time, longer sleep latency (defined as the time is
takes a person to fall asleep once lights are turned off), and reduced sleep efficiency
(defined as the total amount of time a person slept divided by the total amount of time
spent in bed) during the nocturnal sleep period (Karacan & Williams, 1970; Lee et al.,
2000b; Mindell & Jacobson, 2000; Manber, Colrain, & Lee, 2001; Santiago et al.,
2001). During the second trimester, sleep patterns resemble those that are similar to
patterns observed prior to pregnancy. For instance, the increase in WASO and
number of awakenings observed in the 1st trimester wanes in the 2nd trimester,
subsequently increasing again in the 3rd trimester. However, this dampening of sleep
disruption never reaches baseline patterns (Lee et al., 2000b). It is during the 3rd
trimester that the majority of sleep disturbances occur, with an increase in
63


spontaneous nighttime awakenings compounding physiological sleep disturbances
(Mauri, 1990). All the aforementioned studies that have noted the subjective reasons
for nighttime arousal are in agreement as to the nature of the causes. The documented
contributors to sleep disturbance throughout pregnancy range from nausea, vomiting,
backache and urinary frequency to fetal movements; as well as heartburn to shortness
of breath, leg cramps, itching, and nightmares (Baratte-Beebe & Lee, 1999; Driver &
Shapiro, 1992; Mindell & Jacobson, 2000; Schweiger, 1972). These contributors are
not the sole reason for the awakenings. There are additional biological influences that
may contribute to sleep disturbance during pregnancy.
Estrogen and progesterone are altered during pregnancy; they are also
potential contributors to the differences in sleep patterns observed between the
pregnant and nonpregnant state. Data for the influence of hormones on sleep patterns
can be documented since the early 1960s. Heuser, Kales & Jacobson found that
progesterone did not decrease REM time in a pilot study of 5 young adult
subjects(Heuser, Kales, & Jacobson, 1968), and Hartmann (1966) suggested that
hormonal changes in the menstrual cycle might produce an elevated need for
Dreaming Sleep known today as REM sleep. The data that Branchey and Petre-
Quadens (1968) analyzed led them to conclude that the changes in paradoxical sleep
observed in their subjects could be attributed to the sexual hormones secreted during
pregnancy (p.457). Recent work intimates that estrogen enhances the total time spent
64


in REM and reduces the latency period prior to REM sleep (Manber et al., 2001).
However, the primary source for data on estrogen and its effects on sleep come from
pre- and post-menopausal studies. Women who experience menopause have greater
sleep disruption and mood changes (Baker, Simpson, & Dawson, 1997), but no
studies measured estrogen levels. Women who received hormone replacement
therapy in a double-blind, placebo-controlled study showed improvement in sleep
parameters, but they were nonsignificant compared to the placebo group (Saletu-
Zyhlarz et al., 2003). Information regarding progesterone, on the other hand, is
clearer. Progesterone is secreted in high amounts by the placenta (Lee et al., 2000b)
and increases non-REM (NREM) sleep, shortens sleep latency, and reduces
wakefulness after sleep onset (Manber et al., 2001; Santiago et al., 2001). Moldofsky
(1995) reported that at times of elevated progesterone in normal controls, there is a
delay in onset to SWS and a decrease in Stage 4 sleep, which is associated with a
reduced duration in the decline of NK activity.
Pregnancy and Immunity
Pregnancy and Alterations in Immune Function
Pregnancy is characterized by a plethora of immunologic change and these
changes are often referred to as the immunological paradox of pregnancy because
65


even with the paternal antigens present, the fetus is immunologically tolerated by the
mother and often brought to term (Clark et al., 1996; Coussons-Read et al., 2002;
Driver & Shapiro, 1992; Hegde, Ranpura, D'Souza, & Raghavan, 2001; Piccinni,
Maggi, & Romagnani, 2000a; Saito et al., 1999; Santiago et al., 2001; Wadhwa,
Culhane, Rauh, & Barve, 2001; Clark et al., 1996). Some researchers purport that in
order to maintain the viability of the fetus, a chronic immunosuppressive state must
take place (Hegde et al., 2001; Piccinni et al., 2000a; Saito et al., 1999; Santiago et
al., 2001; Wadhwa et al., 2001). Some have even stated the maternal-fetal
relationship is unique and represents a more step-by-step, programmed interactive
symbiosis, rather than a simple host/tumor or host/allograft relationship (Chaouat et
al., 2002). Data indicate that immediately following insemination, the female
reproductive tract undergoes inflammatory changes in response to semen (Hegde et
al., 2001). These inflammatory changes are due to the high levels of
immunoregulatory molecules called prostaglandins, which are a potent
immunosuppressor contained in seminal fluid (Hegde et al., 2001). Similarly, the
uterus is exposed to antigens from the fetus once conception has occurred and the
maternal-fetal compartment is able to distinguish between self and non-self. Because
of this remarkable ability, other scientists contend that the preponderance of the
changes occur at the maternal-fetal interface, not throughout the entire body (Luppi et
al., 2002a; Piccinni et al., 2000a). Regardless of the degree of immunosuppression,
66


including a shift from a pro-inflammatory to an anti-inflammatory cytokine profile
along with secretion of appropriate anti-abortive chemicals such as TGF-0 and IL-10,
it is accepted that various immune modifications must take place in order to achieve a
successful pregnancy.
The role of Thl and Th2 cell populations in human pregnancy has taken
center stage in the last decade, influencing the interpretation and understanding of the
human immune system and providing a blueprint for insight into the complexities of
pregnancy. The observation that there are two major subsets of CD4+ T helper cells,
Thl and Th2, and that they have different cytokine production patterns and roles in
the immune response, has been a significant advancement (Raghupathy, 2001). Thl
cells secrete IFNy, TNF(3,1L-2, and TNFa, which activate macrophages and cell-
mediated reactions; they are also involved in cytotoxic and delayed-type
hypersensitivity reactions. On the other hand, Th2-type cytokines, IL-4, IL-5, IL-6,
IL-10 and IL-13, are primarily concerned with antibody production (Marzi et al.,
1996; Raghupathy, 2001). Of particular interest in the study of pregnancy is that Thl
and Th2 cells are mutually inhibitory to each other IL-10 inhibits the development
of Thl cells, whereas IFNy prevents activation of Th2 cells (Raghupathy, 2001). It
has been suggested that a shift to a Th2 cytokine-secreting cell population at the
maternal-fetal interface is necessary for maintenance of a successful pregnancy
(Luppi, Haluszczak, Trucco, & DeLoia, 2002b; Marzi et al., 1996; Piccinni, Scaletti,
67


Maggi, & Romagnani, 2000b; Raghupathy, 2001). A large number of studies have
attempted to shed light on this complicated and intricate set of events. Beginning with
a quantitative assessment of cell percentages, Saito et al. (1999) showed decreased
percentages of Thl cells in pregnant women in their 3rd trimester versus nonpregnant
women; higher percentages of Th2 cells in women in their 1st trimester than
nonpregnant women; and the ratios of Thl:Th2 in 2nd and 3rd trimester pregnant
women were significantly lower than in nonpregnant women. Many studies have
concluded that in normal pregnancies, secretions of IL-4 and IL-10, which are Th2
cell-synthesized interleukins (cytokines), are increased over the secretion of ENTF-y,
which is a Thl cell-synthesized interleukin (Ho et al., 2001; Marzi et al., 1996;
Raghupathy, 2001). These two cytokines, IL-4 and IL-10, have been shown to inhibit
the development and function of Thl cells and macrophages, thus preventing the
allograft rejection (Piccinni et al., 2000a). According to Raghupathy (2001), evidence
in mice and humans indicates that humoral responses are potentiated during
pregnancy, but that Delayed Type Hypersensitivity (DTH), natural killer (NK)
activity, responses to intracellular infections, and the course of cell-mediated
autoimmune disorders are ameliorated (p.220). It is this reduction in Thl-type
reactivity and augmentation of Th2-type immunity that is observed in pregnancy.
Changes in peripheral lymphocytes have been noted in several studies
(Gennaro, Fehder, Nuamah, Campbell, & Douglas, 1997; Hidaka et al., 1991; Luppi
68


i
et al., 2002b; Veenstra van Nieuwenhoven et al., 2002; Wadhwa et al., 2001;
Watanabe et al., 1997) but the discord among the data remains an issue. One study
showed a decrease in both T and B cells during the entire pregnancy, in particular
CD4+; while NK cells, specifically CD16+CD57-, increased in the 1st trimester and
decreased in the 3rd trimester (Watanabe et al., 1997). A similar study by Hidaka et al.
(1991) showed increased NK activity during the 1st trimester compared to
nonpregnant controls, with a gradual decline as the pregnancy progressed. Decreased
lymphocyte count and percentages were also noted during each trimester compared to
nonpregnant controls. Others have reported a decreased number of circulating
lymphocytes only during the 3rd trimester. Comparing normal pregnant women at 30
weeks gestation to women in the follicular phase, clear differences in numbers and
percentages in the lymphocytes were observed. Data revealed decreases during
pregnancy in the percentage of IFN-y producing Helper cells, Cytotoxic cells and NK
cells, while only the percentage of IL-2 producing T helper cells was shown to
significantly decrease (Veenstra van Nieuwenhoven et al., 2002). Likewise, Russell &
Miller (1986) reported NK cell activity to be reduced as early as 4 weeks gestation
(Russell & Miller, 1986). While some studies have ascertained data on both pregnant
and nonpregnant women, one must be cautious as these are not repeated measures
studies.
69


Focusing on the proliferative response of lymphocytes and NK cells is thought
to be of greater significance than considering only numbers or percentages. It
provides functional data regarding the ability of the immune system at the time of
sampling. Various studies have begun to provide information, but similar to the
percentage data, inconsistencies are plentiful. Data from Veenstra van Nieuwenhoven
et al. (2002) shed light on the cytokine production of not only lymphocytes, but of
NK cells as well; a decrease in the in vitro proliferative response of lymphocytes to
antigens suggests that sustained immunosuppression is typical in pregnancy (Gennaro
et al., 1997; Luppi et al., 2002a; Wadhwa et al., 2001) and that this sustained
immunosuppression is widely believed to contribute to a successful pregnancy.
However, evidence exists to suggest that this immunosuppression is not inherent
throughout the entire body during pregnancy. Circulating immune cells display
components of a pro-inflammatory state as indicated by activated leukocytes and
various monocytes, including granulocytes, during pregnancy (Luppi et al., 2002a;
Luppi et al., 2002b).
A substantial body of data supports that life-threatening complications of
pregnancy and poor pregnancy outcomes can be associated with an excess of certain
immune-markers. Increased levels of Thl-type cytokines, such as IL-2, TNFa, and
IFN-y, have been observed in spontaneous abortions (Clark et al., 1996), recurrent
miscarriages (MacLean, Wilson, Jenkins, Miller, & Walker, 2002; Raghupathy,
70


2001), and preeclampsia (Dekker & Sibai, 1999), while Th2-type cytokines, including
IL-3, -4 and -10, have been noted to assist in the promotion, establishment, and
completion of successful fetal growth (Clark et al., 1996; Dekker & Sibai, 1999;
Gaunt & Ramin, 2001; Gennaro & Fehder, 1996; Hanson, 2000; Piccinni et al.,
2000a; Piccinni et al., 2000b; Raghupathy, 2001; Sacks et al., 1998; Saito et al., 1999;
Clark et al., 1996). It has been suggested that a predominance of Thl-type immunity
is incompatible with successful pregnancy (Raghupathy, 2001). Data from
Raghupathy and colleagues (2001) strongly suggest that women who have several
recurrent spontaneous abortions (RSA) have higher levels of Thl-type cytokines (IL-
2, IFNy and TNFa) than women who had normal pregnancies. The same studies also
showed that normal pregnancies were associated with higher levels of EL-4, IL-5, DL-
6 and IL-10 than those women with RSA (Raghupathy, 2001); likewise, Marzi et al.
(1996) purport that the absence of a type 2 bias may be associated with, and possibly
a predictor of, pathologic events. Thus, from an immunological point of view,
successful pregnancy is characterized by altered ratios of Thl-type and Th2-type
cytokines in a fashion that is consistent with at least partial immunosuppression and
downregulation of a cell-mediated immune response.
As noted previously, pregnancy is associated with a suppression of the
immune system. In the past decade, increased attention has been paid to the role that
the immune system may play in success or failure of pregnancy, despite the fact that
71


the mechanisms are still quite unclear. About 15 to 20 percent of all pregnancies
result in miscarriage, and the more pregnancy losses a woman experiences, the
greater the risk of losing each subsequent conception (Coulam, 2000). Studies
increasingly are focusing on how immunological problems can result in recurrent
spontaneous abortion or miscarriage. From an alloimmune perspective, there are two
possibilities that could result in RSA: either the mothers immune system does not
recognize the pregnancy or the mother develops an abnormal immunologic response
to the pregnancy (Coulam, 2000). A few hypotheses have been suggested and tested
in order to provide some evidence as to why miscarriages occur. One idea is that
elevated levels of Thl-type cytokines may impair the trophoblast from implanting,
resulting in miscarriage (Hill, Polgar, & Anderson, 1995; Hill et al., 1995). However,
other data suggest TNF-a may be a protector of the feto-placental unit when it is
exposed to teratogenic stress (Torchinsky et al., 2003). These contradictory
hypotheses suggest a more complex interaction and influence of cytokines on
pregnancy when and how the shift occurs toward a Th2 cytokine profile may be
most important. Since there appears to be a role for inflammation in implantation,
pregnancy is not a Th2 phenomenon. Several windows for certain cytokines to be
upregulated or downregulated, accompanied by precise timing and tuning, seem
essential (Chaouat et al., 2002).
72


For women with RSA, there exists in vitro evidence that these women
produce more embryotoxic cytokines, like IFN-y and TFN-(3, which indicates
abnormal immune reactivity (Raghupathy, 1997). The commonly studied cytokines
are IFN-y, TNF-a, and IL-2; however, these cytokines may provide a limited picture.
It may be more likely that a cascade of cytokines, each released at different times, and
their independent action may depend on the time of release and the presence of other
cytokines rather than a balance between Thl and Th2 (Carp, 2004).
Currently, the majority of evidence derives from mouse models, but in vitro
data suggest that these cytokines inhibit early embryo development (i.e., trophoblast
growth and function) and cause abortion in mice (Carp, 2004; Hill et al., 1995). In a
clinical study of women with unexplained recurrent pregnancy loss, women who had
delivered without complication, and a control group of men, data support evidence
that trophoblast cells normally induce a Th2-type cytokine response that ultimately
benefits a successful pregnancy. Levels of IFN-y, TNF-a and TNF-p were all highly
elevated in women with RSA. Interestingly, IL-10 and IL-4 were significantly
reduced in women with RSA, but not in normal women, which suggests the role of
these cytokines in the maintenance of pregnancy, possibly by suppressing the Thl -
type cytokines (Hill, 1996). Another study (Soriano, Hugol, Quang, & Darai, 2003)
assessing the role of cytokines in pregnancy, evaluated women who had ectopic
pregnancy (EP), miscarriage, or normal pregnancy, and found alterations in certain
73


cytokines. Serum IL-6, IL-8 and TNF-a were all significantly higher in the patients
with EP than in patients who had miscarried or had normal pregnancies, while there
were no differences between the miscarriage group and the normal pregnancy group
(Soriano et al., 2003). Research has shown that IL-2R are sensitive to and quantitative
markers of T cell activation and proliferation (MacLean et al., 2002). A recent study
reported elevated levels of IL-2R in pregnant women with a history of recurrent
miscarriage, but not in healthy pregnant women without a history of miscarriage
(MacLean et al., 2002). Another approach to understanding RSA focuses on the role
of NK cells during pregnancy. In mice, the presence of NK cells correlates with fetal
resorption; activation of NK cells in pregnant mice leads to fetal resorption
(Wegmann, Lin, Guilbert, & Mosmann, 1993). NK cells are non-specific circulating
white blood cells that produce tumor necrosis factor (TNF); TNF is considered toxic
to the developing fetus and has been shown to be at higher levels in women who have
RSA (Coulam, 2000). IL-2 and IFN-y promote the activity of NK cells (Wegmann et
al., 1993).
Another outcome that has been considered to be potentially immune-regulated
is that of preterm birth. Experts have begun to view preterm birth as the final result of
many possible causes, including intrauterine infections, hormonal disturbances, fetal
injury, uterine ischemia, and uterine overdistention (Dudley, 1999). Several immune
parameters have been thought to be involved in preterm birth. Evidence exists that
74


inflammatory cytokines, such as IL-lp, TNF-a, IL-6, IL-8 and IL-4, are involved in
infection-associated preterm labor (Dudley, 1999). It is mostly in animal models
where a clearer demonstration has been made that inflammatory cytokines can
mediate early pregnancy loss; however, it appears that cytokines are only part of a
cascade of events and it is an abnormally-regulated maternal immune response, not an
infection, that predisposes towards early pregnancy loss (Dudley, 1999).
Finally, the occurrence of preeclampsia in women during the latter half of
gestation has been identified as a pressing medical concern. The disease process
includes hypertension, altered hematology, placental insufficiency, and edema
(Edwards et al., 2000). It is the leading cause of preterm birth and intrauterine growth
retardation and occurs in approximately 5% of all pregnancies (Dekker & Sibai,
1999). Like RSA, preeclampsia is believed to be a result of an inappropriate
activation of the maternal inflammatory response, including activation of
granulocytes and increased release of TNF-a and IL-6, although the precise etiology
is unclear (Dekker & Sibai, 1999; Sacks et al., 1998). The previously cited study by
Saito et al. (1999) also assessed percentages of Thl, Th2 and ThO cells and Thl:Th2
ratios in women with preeclampsia; they found compelling data to support the idea
that the immunological regulatory mechanism appears to be disrupted in women
with preeclampsia (p.554). In women with preeclampsia, Thl cells were increased,
Th2 cells were decreased, and the Thl:Th2 ratio favored Thl cells dominating. Also,
75


in unstimulated and PHA-stimulated cultured PBMC from preeclamptic women, IL-2
and IFN-y levels were higher than in normal pregnant women, while JL-4 was lower.
And finally, IL-12, which induces the differentiation of ThO cells or Th2 cells to
become Thl cells, is higher in the sera of preeclamptic women (Saito et al., 1999).
Sacks et al. (1998) found changes in peripheral blood leukocytes relative to the
nonpregnant state not only in preeclamptic women, but also in normal pregnancy. The
authors conclude, and note that this has not been documented previously, that
pregnancy itself involves a marked generalized inflammatory response (p.85).
Recent data from Rein and colleagues (2002) support an altered immune regulation in
women with preeclampsia, indicating a shift to Thl-type immunity. They found
elevated IL-2 synthesis in preeclamptic women compared to a control group (Rein et
al., 2002). These data imply that the observed shift from Th2 to Thl may be a marker
for preeclampsia, in addition to being a trigger of the disease.
Consequences of the Inflammatory Response During
Pregnancy
Pregnancy results in numerous alterations in maternal immunity. These
changes, particularly a shift from pro-inflammatory to anti-inflammatory cytokine
profiles, appear essential to the proper development and delivery of the fetus. When
this shift does not occur, whether due to infection, stress, or possibly sleep loss,
preterm labor and delivery are more likely to result. Despite advances in medical
76


technology, preterm birth complicated almost 12% of all pregnancies in 2001 (Peltier,
2003). Hence, this problem is serious and requires attention.
Preterm labor and delivery appear to be a consequence of an early or
premature inflammatory response. Normal parturition involves three physiologically
interdependent processes: (1) restructuring of the cervix to allow it to stretch; (2)
weakening and rupture of membranes; and (3) initiation of rhythmic contractions that
will force the fetus and placenta out (Peltier, 2003). It appears that certain pro-
inflammatory cytokines play a significant role in initiating this normal process. With
an abrogation of progesterone, IL-8, IL-lp, IL-6, and TNF-a production is increased,
which commences the cascade of labor events (Peltier, 2003). It is infection or an
inflammatory response that results in a premature rise in pro-inflammatory cytokine
levels that ultimately begins an early labor process (Peltier, 2003). If infection
accounts for 30% of all preterm labor and delivery (Peltier, 2003), other factors, such
as stress and sleep loss, may account for a significant portion of the remaining
variance.
Similar to preterm birth, preeclampsia affects the newborn and its health.
However, preeclampsia is especially dangerous to the health of the mother and
contributes to mortality in childbirth (Tjoa et al., 2003). Recent studies have
attempted to identify markers that may indicate susceptibility to preterm birth or
preeclampsia prior to the development of the disease (Bertran et al., 2005; Fialova et
i
77


al., 2004; Freeman et al., 2004; Tjoa et al., 2003). C-reactive protein (CRP), a
potential marker, is a sensitive indication of tissue damage and inflammation. It has
been shown that women with preeclampsia (Fialova et al., 2004; Tjoa et al., 2003)
and gestational diabetes mellitus (Qiu, Sorensen, Luthy, & Williams, 2004b) have
elevated levels of CRP. By measuring CRP levels early in pregnancy, data suggest
that an early warning of compromised placental development may be indicated
(Bertran et al., 2005; Kafetzis, Tigani, & Costalos, 2005; Tjoa et al., 2003). However,
given the current research and clinical data available, it is likely too early to suggest
that a simple measure of CRP could forewarn the need to take a preventative
measure.
A growing concern is the development of gestational diabetes mellitus (GDM)
and its subsequent impact on the health and lifestyle of the mother and infant.
Gestational diabetes mellitus afflicts approximately 4% of all pregnancies annually
and may be as high as 14% annually (Qiu et al., 2004b). In order to maintain a proper
provision of maternal nutrients to the feto-placental unit, some insulin resistance is
required (Radaelli, Varastehpour, Catalano, & Hauguel-de Mouzon, 2003). When
women develop GDM, insulin resistance is more severe with accelerated fetal
development as one result. GDM increases the risk of fetal macrosomia and other
neonatal morbidities including hypoglycemia, hypocalcemia, polycythemia, and
jaundice. GDM is associated with an increased frequency of maternal hypertensive
78


disorders and the need for cesarean delivery (Qiu et al., 2004b). Comparable to the
use of CRP in early detection of preeclampsia, increased leukocyte counts are used as
a marker of inflammation that is associated with development of type 2 diabetes. The
possibility of extending this screening test to assess risk for development of GDM is
under study (Wolf et al., 2004). Early data suggest that high leukocyte counts are
associated with inflammation that is involved with development of GDM and future
type 2 diabetes (Wolf et al., 2004).
The data clarify the need to understand and better interpret how inflammation
can influence the progress of pregnancy. With knowledge of how inflammatory
processes become exacerbated, information can be ascertained and disseminated as to
how to reduce the occurrence of inflammation and reduce the susceptibility to
negative pregnancy outcomes.
Endocrine Participation During Pregnancy
The role that hormones play in pregnancy is a complex scenario. Corticotropin
Releasing Hormone (CRH) is normally synthesized by the hypothalamus and is the
primary regulator of the HPA axis. Its final objective in this process is the subsequent
release of glucocorticoids, which in turn influence the immune system. From the
commencement of pregnancy, the placenta, fetal membranes and decidua synthesize
and secrete CRH identical to that of the mother (Florio et al., 2002; Wadhwa,
79