BIOLOGY
OF REPRODUCTION
40,
Circadian
988-993
(1989)
Uterine Activity in the Pregnant
Rhesus
Do Prostaglandins
Play a Role?1
CHARLES
A. DUCSAY2
Division
and
CAHLEEN
of Perinazal
Department
of
Macaque:
M. McNUTF
Biology
Physiology
School
of Medicine
Loma
Linda
University
Loma
Linda,
California
92350
ABSTRACT
Eight rhesus macaques
between
127 and 132 days of gestation
had catheters
implanted
into maternal
femoral
vessels
and the amniotic fluid cavity and were placed in a vest-and-tether
system for chronic catheter
maintenance.
Uterine
activity was continuously
recorded,
and paired maternal
arterial
blood and amniotic
fluid samples were collected
at 0900 h (AM) and 2100 h (PM) until delivery and analyzed
for prostaglandin
metabolites
(PGFM
and PGEM-11).
A circadian
pattern
in uterine
contractility
was observed,
with peak
activity occurring
between 1900 and 0100 h (p<0.O0l).
No signcanz
AM-PM
d4fferences
were observed in
maternal
plasma PGFM
(240 ± 24 AM vs. 273 ± 35 PM) or PGEM-Jl
(537 ± 41 AM vs. 484
± 34 PM)
or amniotic fluid PGFM
(360 ± 72 AM vs. 287 ± 70 PM) or PGEM-lI
(1626
± 383 AM vs. 1771 ± 431
PM). All values represent mean ± SEM, pg/mi. Additional samples were collected at 3-h intervals for 24 h at
selected times during the study. This more intensive sampling
protocol
also failed to reveal any significant
time trends in maternal
plasma
or amniotic fluid prostaglandins.
Despite
the lack of AM-PM
differences,
amniotic fluid PGFM
and PGEM-li
increased
significantly
as delivery approached
(p<O.Ol).
It appears that
circadian
uterine activity
is not related
to changes in maternal
plasma
or amniotic
fluid prostaglandins.
Although
prostaglandins
are responsible
for the progression
of labor, other factors
may be involved
in the
generation
of uterine
activity
rhythms prior
to the initiation
of labor.
INTRODUCTION
The stimulatory
effects of prostaglandins
on myometi-ia! contractility
are well documented
(Novy and Liggins, 1980; Reimer
and Roberts,
1986). The present
study examines
the relationship
between
plasma
and
amniotic
fluid prostaglandins
and uterine
contractility
in the rhesus macaque
during late gestation.
Specifically, this study was designed
to test the hypothesis
that
prostaglandins
play a role in the generation
of circadian
uterine
activity.
Studies
in the rhesus macaque
have demonstrated
a
circadian
rhythm
in uterine
activity
in late gestation.
Peak activity
occurred
at approximately
1200 h in the
studies of Harbert and Spisso (1980), while other studies (Ducsay
et al., 1983; Taylor et al., 1983) observed
peak activity
between
2100 and 2400 h. In all of these
reports, however,
uterine contractility
increased
in magnitude and gradually
culminated
in labor and delivery.
Similar rhythms
are believed
to occur in women
(Malek, 1952; Tamby
Raja and Hobel,
1983). The causal
mechanisms
of these rhythms
and their potential
role in
the initiation
of labor remain undefined.
Parturition
may
be an extension
or amplification
of these normally
occurring
events. These rhythms,
therefore,
may serve
as a predictable
model for labor.
Accepted
January
Received
November
MATERIALS
Animals
Surgical
METHODS
Procedures
Eight rhesus
macaques
of known
gestational
age
were maintained
in a controlled
environment
with a
12L:12D
cycle with lights on from 0700 to 1900 h.
Surgery
was performed
between
Days 127 and 132 of
gestation
(term = 167 days) by methods
previously
described
(Ducsay
et al., 1983). Briefly, catheters
were
inserted
into the maternal
femoral
artery and vein and
13. 1989.
21.
and
AND
1988.
1Supported
by NIH Grant HD 22865.
2Reprint requests.
988
PROSTAGLANDINS
AND
CIRCADIAN
advanced
to lie within
the descending
aorta and the
inferior
vena cava, respectively.
Additional
catheters
were placed in the amniotic
fluid cavity. After surgery,
the animals
were placed
in a primate
vest-and-tether
system
(Ducsay
et al., 1988) to which they had been
previously
acclimated.
Uterine
Activity
Recording
and
Analysis
Uterine activity was continuously
monitored
through
a saline-filled
amniotic
fluid catheter
connected
to an
eight-channel
polygraph
(Gould,
Cleveland,
OH) via a
pressure
transducer.
The recorder
was interlaced
with
an IBM PC-AT microcomputer
with an analog-to-digital converter.
The real-time
data acquisition
system
(Dale et al., 1989) has a sampling
rate of 16 Hz and an
exponential
filter with a user-defined
time constant
that
is applied
to the digitized
signal.
The output of this
filter is regarded
as baseline
and the input and output of
this filter are compared
continuously.
A contraction
is
recorded
each time input minus output remains
greater
than a specified
threshold
(typically
2 to 4 mmHg) for a
minimum
duration
without
exceeding
a preset,
maximum duration.
The software
utilizes pattern recognition
algorithms
for describing
uterine contractions.
This system is effective
in eliminating
activity
generated
by
random
animal movement
without
a loss of sensitivity.
Contraction
amplitude,
duration,
time to peak amplitude, and area of each contraction
were displayed
and
immediately
written
to disk. Uterine
contractility
data
were analyzed
as previously
described
in detail (Ducsay
et al., 1983). The total contraction
area for each hour of
the day, defined
as the hourly contraction
area (HCA),
was computed
and used to quantify
uterine activity and
to evaluate
its rhythmicity.
This parameter
is the most
complete
measure of contractile
activity since it encompasses
frequency,
amplitude
and duration
of uterine
contractions.
The mean HCA (MHCA)
was defined
as
the sum of the HCAs divided
by the total number
of
hours of monitoring
for each animal. The HCA:MHCA
ratio was computed
for each hourly interval
during a
24-h cycle. Data from successive
days recorded
from
each animal were pooled,
and the mean HCA:MHCA
was calculated.
Sanple
Collection
Beginning
arterial blood
simultaneously
and
Prostaglandin
Assays
two to three days after surgery,
maternal
and amniotic
fluid samples were collected
every other day at 0900 and 2100 h.
UTERINE
ACTIVITY
989
Additional
samples
were collected
at 3-h intervals
for
24 h at selected
times throughout
the course
of the
study. Samples
were collected
in syringes
containing
indomethacin
ethylenediaminetetraacetate,
(EDTA)
and
centrifuged
at 4’C. Maternal
arterial plasma and amniotic fluid were stored at -70’C
until assayed.
Maternal
erythrocytes
were washed
in saline, resuspended,
then
returned
to maternal
circulation.
Due to their stability
when compared
to their parent
compounds,
the metabolites
of prostaglandin
F2u
(13,14-dihydro-15-keto
PGF2;
PGFM)
and prostaglandin
E2 (11 -deoxy-1 3,14-dihydro-1
5-keto
11 16Ecycloprostagland.in
E2; PGEM-Il)
were measured.
The
PGFM
antiserum
was obtained
from Dr. Lawrence
Levine,
Brandeis
University
and the PGEM-I1
antiserum was a generous
gift of the UpJohn
Company,
Kalamazoo,
MI. The antisera were used in conjunction
with the appropriate
tritiated
compounds
(Amersham
Corp., Arlington
Heights,
IL, and New England
Nuclear, Boston,
MA) according
to methods
previously
described
(Novy et al., 1987).
The sensitivities
of the assays were 6 and 3 pg for
the PGFM
and PGEM-II
assays,
respectively
(note:
sensitivity
is defined as the smallest
amount of prostaglandin per assay tube that reduces
the number of cpm
of the labeled prostaglandin
bound at zero mass by two
standard
deviations).
Accuracy
of the assays was determined
by recovery
experiments
in which
known
amounts
of each prostaglandin
were added to individual
aliquots of plasma and amniotic
fluid. Linear regression
analysis
of prostaglandin
concentrations
measured
(y)
vs. prostaglandin
concentrations
measured
(x) yielded
the following
regression
equations
and correlation
coefficients:
y = 1.03x + 46.2; r = 0.97 (PGFM),
and y =
0.19x
+
6.90; r = 0.97
(POEM-lI).
The intraassay
coefficients
of variation
were 8.1 and 5.9 for the PGFM
and POEM-Il
assays,
respectively,
and the interassay
coefficients
of variation
were less than 10% for both
assays. The specificities
of the antisera
were adequate
as judged by their low cross-reactivities
with interfering
prostaglandins
(less than 1%).
,
Statistical
Analyses
Uterine
activity
data and changes
in prostaglandin
concentrations
by time of day were analyzed
by twoway ANOVA.
Duncan’s
test was used to determine
which times differed
when the variance
ratio (F) was
significant.
Log transformations
were used to achieve
homogeneity
of variance
when appropriate.
Changes
in
DUCSAY
990
TABLE
1. Maternal
artenal
(MA)
and amniotic
fluid (AF) prostaglandin
AND McNUTf
(P0)
Time of Day
concentrations.
0900h
n
2100h
n
PGFM
MA
AF
240.8
360.8
±
±
24.0
72.8
41
33
273.0
287.5
POEM-Il
MA
AF
537.4
±
41.8
1626.9
48
40
484.1
1771.2
Vals
represent
prost.aglandin
evaluated
mean
± SEM.
35.3
69.7
35
35
± 32.1
± 431.8
38
29
pghnl.
concentration
by two-way
± 383.6
±
±
with
analysis
gestational
age
were
of covariance.
RESULTS
Uterine
activity was recorded
for periods
of 8 to 21
days at 127 to 155 days of gestation.
The mean HCA:
MHCA ratios are illustrated
in Figure 1 and represent
more than 2500 hours of continuous
recordings.
These
data demonstrated
a circadian
pattern with peak activity
occurring
between
1900 and 0100 h (p<0.01),
followed
by a decrease
with a nadir between
0400 and 0900 h.
The significant
increase
in overall
uterine contractility
appeared
to be coincident
with the advent of the dark
phase
of the light cycle,
represented
by the crosshatched
bars in Figure
1.
Maternal
arterial
plasma
concentrations
of PGFM
and POEM-H
failed to reveal differences
between
samples collected
at 0900 vs. 2100 h (Table 1). Likewise,
no AM-PM
differences
were observed
in amniotic
fluid
prostaglandin
concentrations
measured
during the same
time periods
(Table
1).
To further
assess
possible
rhythmicity
in prostaglandin concentrations,
a more frequent
sampling
protocol was employed
at selected
times throughout
the
study. Maternal
arterial plasma and amniotic
fluid samples were collected
at 3-h intervals
over a 24-h period.
Figure
2 illustrates
maternal
arterial
plasma
prostaglandin concentrations.
Neither maternal
plasma PGFM
(closed
circles)
nor POEM-Il
(open circles)
demonstrated a rhythm
when examined
over 24 h (p>0.0S)
(Fig. 2).
Amniotic
fluid concentrations
of PGFM (closed circles) failed to demonstrate
changes
when examined
over 24 h (Fig. 3). In contrast,
POEM-lI
levels (open
circles)
were lowest at 2100 h, with an increase
observed during the late night and early morning
hours
(Fig. 3). These fluctuations,
however,
were not significant (p>O.O5).
.2
Lght
Dark
4
a.
U
r
..
2
1
..
-
800
a.
-
a...
0
2
4
0
0
600
z
U
C
I
z
2=
#{149}
-
400
4
II#{252}
200
I
1100
1300
1500
1700
1900
2100
2300
0100
0300
0500
0700
0900
p<0.0l
p<o.ooi
TIME
OF
-
0-
I
I
I
0900
1200
1500
I
1800
I
I
I
2100
2400
0300
0600
0900
DAY
FiG. I. Mean values for hourly contraction
area (HCA):mean
hourly contraction area (MHCA) ratios for chronically
catheterizcd
rhesus macaques during late gestation.
Lights were on from 0700 to 1900 h (open bars) and off from
1900 to 0700 h (cross-hatched bars). Significant
elevations
in uterine activity
occurred between
1900 and 0100 h (*p<0.Ol).
TIME
OF DAY
FIG. 2. Maternal arterial (MA) plasma prostaglandin
(P0) concentrations
collected over a 24-h period. PGFM levels are represented by the closed circles.
and POEM-il
values are denoted by the open circles. No significant
time trend
in either of the PG compounds
was observed.
PROSTAGLANDINS
I
2500
I
I
I
AND
I
CIRCADIAN
UTERINE
ACTIVITY
991
I
-
I
I
I
I
I
I
I
I
I
I
4
2
0
11100
o
z
0
0
z
2000-
-
800
1500
-
1000
-
600-
-
0
1.!..
(3)
400-
a500200
I
I
I
I
I
0900
1200
1500
1800
2100
TIME
FIG.
3. Amniotic
fluid
(closed circles) and POEM-Il
(3)
0-
[
2400
-
0300
0600
0900
6000
-
5000
-
4000
-
3000
--
©
OF DAY
(AF) concentrations
of prostaglandin
(P0)
(open circles) collected
over 24 h. Although
statically
significant
time trends were observed,
phasic increase in POEM-U concentrations.
there
is a trend
toward
FM
no
=
a bi-
2000-
1000-
Since there
arterial plasma
were
no AM-PM
or amniotic
differences
fluid
in maternal
prostaglandin
concen-
trations,
data from each source
were pooled
and examined
across
gestation.
No changes
in either
maternal
plasma
POFM
or POEM-I!
were observed
as gestation
progressed
fluid
(data
not
shown).
and PGEM-II
PGFM
both
However,
demonstrated
cant (p’czO.Ol) correlations
with gestational
4A and 4B illustrate
significant
increases
fluid
PGFM
and POEM-lI,
respectively,
approached.
The increase
in amniotic
fluid
ins preceded
alterations
in HCA:MHCA
Changes
quency
in total
(not
shown)
contraction
area
followed
and
0
I
>-J
signifi-
age. Figures
in amniotic
as delivery
prostagland(Fig.
4C).
contraction
a similar
I
amniotic
highly
Uterine
activity
data
from
previously
published
Taylor
et al. (1983).
pattern.
examination
fetus,
and
observed
a rhythm
Z
2-
w
I
I
I
I
I
I
I
I
20
18
16
14
12
10
8
6
DAYS
BEFORE
DELIVERY
(p<O.O1).
the
present
by
study
us (Ducsay
Contractile
of hormonal
profiles
amniotic
fluid.
Walsh
in fetal
0
FIG. 4. Relationship
of amniotic
fluid prostaglandin
(P0) FM (A), and
POEM-Il (B) to uterine activity (C) prior to delivery. In Graph.s A and B, significant increase
in concentrations
of both POs was noted
with gestational
age
agree
Ct
al.,
activity
in
the rhesus
macaque
follows
a well-defined
circadian
pattern,
with peak activity
during
the dark phase of the
light cycle.
To our knowledge,
the present
study
is the
first to examine
the relationship
between
circadian
uterine contractility
and alterations
in maternal
plasma
and
amniotic
fluid
prostaglandin
concentrations.
Other
efforts
to elucidate
the factors
involved
in the
generation
of uterine
activity
rhythms
have
centered
around
mother,
3-
fre-
DISCUSSION
with those
1983)
and
©
0
dehydroepiandrosterone
of the rhesus
et al., (1984)
sul-
fate
that
parallels
increases
in
These
data are in agreement
lis et al., 1980;
Seron-Ferre
nocturnal
sone
rhythm
treatment
blocks
uterine
further
suggesting
in fetal
inhibits
activity
adrenal
fetal
rhythms
a relationship
epiandrosterone
sulfate
uterine
contractility.
with earlier
studies
(Chalet al., 1983)
that found
a
steroids.
adrenal
Dexametha-
activity
and
(Ducsay
ct al.,
between
fetal
and nocturnal
uterine
also
1983),
dehydrocontractil-
ity.
The
may
effects
of dexamethasone
be interpreted
to be the
on uterine
result
contractility
of reduced
prosta-
992
DUCSAY
AND
glandin
production
secondary
to reduced
estrogen
synthesis (Novy and Liggins,
1980). However,
dexamethasone
treatment
may
also
inhibit
prostaglandin
production
by stimulating
production
of a circulating
inhibitor
of prostaglandin
synthetase
(Saeed
et al.,
1977; Saeed et al., 1982) or by inducing
synthesis
of a
phospholipase
A2 inhibitor
(Rothhut
et al., 1983;
Errasfa
et a!., 1985).
Regardless
of the exact nature
of dexamethasone
inhibition
of uterine contractility,
prostaglandins
appear
to be the final common
pathway
for stimulation
of
contractile
activity (see Casey and McDonald,
1986, for
review).
However,
conclusive
evidence
that prostaglandins
mediate the signal for the initiation
of labor in
women and nonhuman
primates
is still lacking. This is
attributable
to the fact that identified
hormonal
trends
are unable
to distinguish
whether
increased
prostaglandin
levels before and during labor are the cause or
the result of uterine
contractility.
The present study was designed
to determine
if there
is a relationship
between
maternal
plasma and/or amniotic fluid prostaglandins
and circadian
uterine contractility prior to the initiation
of labor. The prostaglandin
metabolites
PGFM and PGEM-II
were quantified
due to
their relative stability
compared
to their respective
parent compounds.
Maternal
plasma
and amniotic
fluid
samples collected
at 0900 and 2100 h failed to demonstrate any changes
in prostaglandin
concentrations.
One
problem
with this approach
is that alterations
in prostaglandin concentrations
related to change in uterine contractility
may be missed due to infrequent
sampling.
We
attempted
to overcome
this deficiency
with a more
intensive
sampling
protocol.
Results from samples collected
every 3 h throughout
24-h periods
also failed
to demonstrate
changes
in
maternal
plasma
prostaglandin
concentrations
associated with the uterine
activity
rhythm.
We postulated
that if circulating
prostaglandins
were responsible
for
the generation
of the uterine
contractility,
we would
observe
increases
preceding
changes
in contractile
activity. In contrast,
if other factors
were involved
in
stimulation
of the myometrium,
we might
observe
changes
in maternal
plasma
prostaglandins
that followed
the periods
of peak activity.
Peak POEM-Il
concentrations
were reached
at 1800 and 0900 h, suggesting
that we may have observed
a combination
of
the above mentioned
events.
However,
the differences
McNUTF
were not significant.
POFM
concentrations
remained
relatively
constant.
The definitive
data regarding
circulating prostaglandin
concentrations
would come from
samples
obtained
from a uterine vein catheter,
but the
technical
difficulties
of maintaining
such a catheter
make the feasibility
of such studies
unlikely.
Frequent
sampling
of amniotic
fluid failed to demonstrate significant
alterations
in prostaglandin
concentrations. However,
a trend toward increases
in POEM-I!
beginning
at 2400 h was observed.
This increase
occurred after the initiation
of increased
uterine activity at
1900 h. Since POEM-I!
is a metabolite
of PGE2, this
might be the result of increased
POE synthesis
that
occurred
hours earlier, or, it might be due to increased
prostaglandin
production
that followed
increased
uterine activity. More frequent
sampling
between
1900 and
2400 h will be necessary
to make this distinction.
Prostaglandins
are still the probable
causative
agent
for the initiation
of labor. Data from the present study
are in agreement
with the earlier
work that demonstrated
a significant
rise in amniotic
fluid
PGFM
(Mitchell
et a!.,
1976)
and
POEM-Il
(Haluska
et
a!.,
1987)
concentrations
as
delivery
approached.
It is interesting
to compare
the uterine activity
pattern in the present
study with our previously
reported
data (Ducsay
et al., 1983). The patterns
are similar, but
on closer inspection
it becomes
apparent
that the activity was initiated
and reached peak levels 2 to 3 h sooner
in the present
study. The principal
difference
between
the two study protocols
is the light cycle. The earlier
study utilized a 16L:8D cycle with lights on from 0700
to 2300 h. In the present
study, lights were on from
0700 to 1900 h. This lighting
regimen
is identical
to
that employed
by Taylor et al. (1983), and the circadian
characteristics
of the uterine activity rhythms
are similar. Earlier
studies by Harbert
and Spisso (1980) utilized a 12L:12D cycle as well. However,
the lights were
on from 0400 to 1600 h, and the uterine activity cycle
in that study was approximately
180’ out of phase with
the data described
above.
Taken
together,
these data suggest
that additional
factors are involved
in the generation
of uterine activity
rhythms.
There
is abundant
evidence
in human
and
nonhuman
primates
that L:D cycles
are the principal
“zeitgabers”
of circadian
rhythms.
Melatonin
concentrations are elevated
during periods
of darkness
and depressed
during
periods
of illumination
in the rhesus
macaque
as well as in a wide range of other species
PROSTAGLANDINS
AND
CIRCADIAN
(Perlow
et a!., 1981). Data from Harbert
and Spisso
(1981) suggest that catecholamines
may influence
contractile patterns
in the rhesus macaque
during gestation.
These factors
are currently
under investigation
in our
laboratory
to establish
their role in the regulation
of
circadian
uterine activity
rhythms
in the pregnant
rhesus macaque.
ACKNOWLEDGMENTS
ford
The authors would like to express their sincere
and Lisa Harvey for their excellent technical
appreciation
assistance.
to Shirley
Mcd-
REFERENCES
Casey
ML, MacDonald
PC, 1986. The initiation of labor in women: regulation
of phospholipid
and of prostaglandin
production.
Semin Perinatol (NY)
10:270-75
Challis JRG, Socol M. Murata Y, Manning FA, Martin CB Jr, 1980. Diurnal
variations
in maternal
and fetal steroids in pregnant rhesus monkeys. Endocrinology
106:1283-88
Dale PS, Ducsay CA, Gilbert RD. Koos BI. Longo LD. Power GO. A microcomputer program for real-time data acquisition
in the perinatal physiology laboratory.
J Dev Physiol (in press)
Ducsay CA, Cook MJ, Navy Mi, 1988. Simplified
vest and tether system for the
maintenance
of chronically
catheterized
pregnant
rhesus monkeys (Macoca mulatto). Lab Anim Sci 38:343-44
Ducsay CA, Cook Mi, Walsh Sw, Novy Mi, 1983. Circadian patterns and dexamethasone-induced
changes
in uterine
activity in pregnant rhesus monkeys. Am I Obstet Gynecol 145:389-96
Errasfa M, Rothhut B, Fradin A, 1985. The presence of lipocortin in human and
embryonic
skin fibroblasts
and its regulation
by anti-inflammatory
steroids. Biochim
Acta 847:246-52
Haluska GJ, Stanczyk
FL, Cook Mi, Novy Mi, 1987. Temporal
changes in
uterine activity and prostaglandin
response to RU486 in rhesus macaques
in late gestation.
Am I Obstet Gynecol 157:1487-95
Harbert GM Jr. Spisso KR, 1980. Biorhythms
of the primate
uterus
(Macaca
mulatta) during labor and delivery.
Am I Obstet
Gynecol
138:686-96
Harbert GM ii, Spisso KR, 1981. Effect of adrenergic
blockade
on dynamics
of
UTERINE
ACTIVITY
993
the pregnant uterus (Macaca
rnulaua).
Am I Obstet Gynecol 139:767-80
J, 1952. The manifestation
of biological rhythms in delivery.
Gynaecologica 133:363-78
Mitchell MD, Patrick JE, Robinson
IS, Thorburn GD, Challis JRG. 1976. Prostaglandins
in the plasma and amniotic fluid of pregnant rhesus monkeys
and after intrauterine
death. I Endocrinol
71:67-76
Novy Mi, Ducsay CA. Stanczyk
FL. 1987. Plasma concentrations
of prostaglandins F2 (POFM) and POE2 (POEM-lI) metabolites
after transabdominal and transvaginal
cerciage. Am I Obstet Gynecol
156:1543-52
Novy Mi, Liggins
GC, 1980. Role of prostaglandins,
prosiacyclin,
and thromboxanes in the physiologic
control of the uterus and in panurition.
Semin
Perinatol 4?45-66
Perlow Mi, Reppert
SM, Boyar RM. Klein DC, 1981. Daily rhythms in cortisol
and melatonin
in primate cerebrospinal
fluid. Neuroendocrinology
32:
193-96
Reimer RK, Roberts RM, 1986. Activation of uterine smooth muscle contraction implications
for eicosanoid
action and interactions.
Semin Perinatol
10:276-87
Rothhut B, Russo-Marc
F, Wood 1, 1983. Further characterization
of the glucoconicoid-induced
antiphospholipase
protein”renocortin”.
Biochem Biophys Res Commun
117:878-84
Saced SA. McDonald-Gibson
WI, Cuthbert
I, Copas JL, Schneider C, Gardines
P1, Butt NM, Collier HOl, 1977. Endogenous
inhibitor of prostaglandin
synthetase.
Nature (Lond) 270:32-36
Saeed SA, Strickland
D, Young DM, Mitchell MD, 1982. Inhibition
of prostaglandin synthesis by human amniotic fluid: acute reduction
in inhibitory
activity of aniniotic fluid obtained
during
labor. .1 Cm Endocrmol
Metab
55:801-03
Seron-Ferre
M, Taylor NF, Rotten D. Koritnik DR. Jaffe RB. 1983. Changes in
fetal rhesus monkey plasma dehydroepiandrosterone
sulfate: relationship
to gestational
age, adrenal weight and preterm delivery. I Clin Endocrinol
Metal, 57:1173-78
Tamby Raja RL, Hobel Ci, 1983. Characterization
of 24 h uterine activity in the
second half of the human pregnancy.
Proceedings
of the 30th Annual
Meeting
of the Society
for Gynecologic
Investigation,
March 17-20,
Washington,
DC: p. 280. (Absir. 516)
Taylor NF, Martin MC, Nathanielsz
PW, Seron-Ferre
M, 1983. The fetus determines circadian
oscillation
of myomethal
electromyographic
activity
in
the pregnant rhesus monkey. Am J Obstet Gynecol
146:557-67
Walsh SW, Ducsay CA, Novy Mi, 1984. Circadian
hormonal
interactions
among the mother, fetus and amniotic
fluid. Am I Obstet Gynecol
150:
745-53
Malek