![[Feedback]](/icons/banner/feedbackACT.gif){kind=icon}
![[For Subscribers]](/icons/banner/subscriberACT.gif){kind=icon}
![[Archive]](/icons/banner/archiveACT.gif){kind=icon}
![[Search]](/icons/banner/searchACT.gif){kind=icon}
![[Contents]](/icons/banner/tocACT.gif){kind=icon}
|
|
|
|
|
||
Program in International Nutrition, Department of Nutrition, University of California, Davis, CA
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
KEY WORDS: • iron deficiency • anemia • preterm • low birth weight • pregnancy • corticotropin-releasing hormone
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
in this supplement
reviews the strength and plausibility of the evidence that
iron-deficiency anemia or anemia is the cause of adverse birth
outcomes. Although there are numerous reports of an association between
maternal hemoglobin and both lower birth weight and preterm delivery,
Rasmussen concludes that evidence is insufficient to prove that iron
deficiency plays a causal role in poor pregnancy outcome. To some
extent, this lack of evidence is due to the inadequate design of
published studies. The purpose of this paper is to explore the biological mechanisms through which anemia or iron deficiency might cause poor fetal growth and preterm delivery, regardless of whether a causal link has been proven. Because of the complex hormonal and physiological factors that affect pregnancy outcome and the virtual lack of any studies of the effects of iron deficiency on these factors, the biological mechanisms proposed below are hypothetical. This article summarizes current information about the biological mechanisms involved in preterm delivery and intrauterine growth retardation (IUGR)3 and suggests ways in which these may be affected by anemia or iron deficiency. It is intended to provide information that will be useful for planning which hormones and metabolites should be measured in future studies of the role of iron deficiency and iron-deficiency anemia in pregnancy outcome.
The review starts with a summary, based on the Rasmussen review, of what is known about the possible effects of iron deficiency and iron-deficiency anemia on infant size at birth and duration of gestation. The prevalence of IUGR and preterm delivery is described, with a comparison of the risk factors for these conditions. This is followed by a description of the biological mechanisms involved in the normal delivery process, including the central role of corticotropin-releasing hormone (CRH) in determining the duration of gestation and the association of other hormones with fetal growth. Finally, mechanisms are proposed that might underline iron’s effects on these processes, increasing the risk of preterm delivery and low birth weight.
|
Associations between iron-deficiency anemia and pregnancy outcome |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
In her review, Rasmussen summarizes the results of 54 nonintervention
studies that examined associations between hemoglobin or hematocrit and
pregnancy outcome. Of these, 44 analyzed associations among hemoglobin
or hematocrit, birth weight and percentage of low birth weight (
2500
g) deliveries. In 26 of the 44, anemia, lower hemoglobin or hematocrit,
or low ferritin concentrations were significantly associated with a
higher prevalence of low birth weight, whereas in the other reports,
this was not the case. It is not known to what extent preterm delivery
could explain these observed associations between lower birth weight
and anemia because gestational age was assessed in only 10 of the 44
studies. In 5 of these 10 (2 in Papua New Guinea, 1 in India, 1 in Hong
Kong and 1 in the United States), anemia was associated with a shorter
gestation period. In the U.S. study on adolescents of low socioeconomic
status, preterm delivery explained all of the lower birth weight
associated with iron-deficiency anemia when other potential
confounders were controlled for by regression analysis (Scholl et al. 1992
).
In 21 of the 54 nonintervention studies summarized by Rasmussen, gestation duration was not measured, and in two of the studies, the sample was restricted to full-term births. Of the remaining 31 studies that examined an association between either hemoglobin or hematocrit and duration of gestation, 7 found no association, 22 found a positive association, 1 found a negative association and 1 reported a positive association between duration and the intake and serum concentrations of folate. Ferritin concentrations were assessed in only six of these reports and were positively associated with gestation duration in four of them. We conclude that it is plausible that the frequently observed association between maternal hemoglobin and birth weight might be caused by a shorter gestation.
Intervention studies.
Rasmussen identified 17 iron, folate, or iron and folate intervention studies in which the effect of the supplements on size at birth, duration of gestation or both was assessed. The design of these studies was remarkably poor in that false negatives (primarily from treatment of nonanemic women), bias or potential confounders were present in nearly all of them. Duration of gestation was assessed in only six of the studies, and none were conducted in developing countries or with anemic women.
|
Prevalence of preterm delivery and low birth weight |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
The relative contribution of preterm delivery and fetal growth
retardation to low birth weight has been compared by using birth weight
and gestational age data from 25 developing regions and 11 developed
regions of the world (Villar and Belizan 1982
). The
prevalence of low birth weight, IUGR-low birth weight and preterm
low birth weight was 23.6, 17.0 and 6.7%, respectively, for developing
countries and 5.9, 2.6 and 3.3%, respectively, for developed
countries. These prevalences refer only to low-birth-weight infants and
therefore underestimate substantially the total prevalences of
intrauterine growth restriction and preterm delivery. In addition,
especially in developing countries, preterm delivery is often not
diagnosed at all or is diagnosed inaccurately.
Preterm delivery is the main cause of perinatal mortality and morbidity
in industrialized countries, in which the prevalence has not fallen in
the last few decades. In fact, in a large, carefully conducted Canadian
study, the rate of preterm births over the past 20 years increased from
6.6 to 9.8% for <37 wk, 1.7 to 2.3% for <34 wk and 1.0 to 1.2% for
<32 wk (Kramer et al. 1998
). Although half of the
apparent increase in prevalence was due to earlier use of ultrasound
dating, other factors such as preterm induction of labor, use of
cocaine and an increase in the number of pregnant unmarried women have
tended to increase the prevalence.
|
Risk factors for low birth weight and preterm delivery: similarities and differences |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
). In
developing countries, more IUGR is due to low maternal weight and
height (undernutrition during the mother’s development), low pregnancy
weight gain (which is influenced by energy intake during pregnancy) and
maternal infection. Low maternal weight at conception and low weight
gain during pregnancy are independent predictors of poor fetal growth
and IUGR. Low maternal height also plays a role independently of low
body mass index.
In a large sample of Canadian women with gestational time confirmed by
ultrasound, significant predictors of preterm birth (before 37 wk)
included low maternal stature, noncompletion of high school, unmarried
status, smoking, diabetes, urinary tract infection within 2 wk of
delivery, prepregnancy hypertension, severe pregnancy-induced
hypertension, previous history of a preterm birth, low birth weight or
infant death at birth (Kramer et al. 1992
). Anemia was
not mentioned in the analysis.
The statement was made that "the degree of overlap between the sets
of risk factors for preterm delivery and growth retardation (IUGR) is
so great that it is easiest to list the discrepancies" (Yu and Wood 1987
). Of the many risk factors for preterm delivery, a
few (including cervical trauma, diethylstilbestrol exposure, maternal
infections and febrile states) are associated with preterm delivery,
but not with IUGR (Bakketeig 1991
, Institute of
Medicine et al. 1985
). The relative risks of preterm delivery
and IUGR for each risk factor are usually different, however.
The relationship between maternal nutrition and risk of preterm
delivery has not been studied adequately. Most investigators reported
no relationship between maternal nutritional status or interventions
and duration of gestation. However, as observed by Kramer et al. (1992)
, most studies had major design problems, including the
following: assessment of total pregnancy gain, which is affected by
duration, rather than rate of gain or net gain; use of average rate of
gain, which is also affected by length of gestation because rate is
slower in the last trimester; the confounding effect of fetal size
later in gestation; the inclusion of induced deliveries; and poor
assessment of gestational age. Reported date of last menstrual period
is prone to significant error especially at the extremes of gestational
age. The use of ultrasound rather than reported date of the last
menstrual period lowers the estimated prevalence of small for
gestational age by 
30–50% in industrialized countries
(Bakketeig 1991
).
Maternal stress, anxiety and other psychological factors appear to be
more strongly associated with risk of preterm delivery rather than with
risk of IUGR (Hedegaard et al. 1996
, Lobel et al. 1992
, Nordentoft et al. 1996
). One of the more
certain risk factors for preterm delivery, maternal infection, has been
implicated in up to 40% of cases (Kurki et al. 1992
).
Thus, there is substantial overlap between the risk factors for preterm delivery and IUGR. Moreover, as will be discussed below, there are strong links in the underlying biological mechanisms associated with these two outcomes.
|
Biological mechanisms involved in the normal delivery process |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
). As a result, there is
relatively little information on factors that affect these mechanisms
and virtually none on nutritional factors. To some extent, the lack of
information is due to the ethical difficulties of doing intervention
trials that could influence delivery in humans. Animal models have been
useful, and the basic mechanism underlying parturition in sheep had
been identified by the mid-1980s. However, the ovine mechanisms and
those in some primates are different from those in humans. Early in pregnancy, the hormone progesterone, secreted by the placenta, keeps the smooth muscle cells of the uterus relaxed and the cervix tightly closed as a result of its influence on the relatively inflexible collagen fibers. Later in pregnancy, the secretion of estrogen by the placenta increases dramatically and parturition begins when the influence of estrogen is greater than that of progesterone.
|
The importance of corticotropin-releasing hormone |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
). Although
increases in plasma concentrations are seen in some women as early as
the second trimester (Goland et al. 1986
), in most
women, levels are low until the third trimester and then rise rapidly
(Emanuel et al. 1994
, Goland et al. 1986
). The mRNA for placental CRH and the release of the
hormone into maternal plasma increase as much as 50-fold during the
third trimester. CRH concentrations in maternal plasma vary up to
50-fold in late pregnancy.
The CRH produced by the placenta is secreted into the fetal circulation
in amounts high enough to stimulate the production of
adrenocorticotropic hormone (ACTH, corticotropin) by the anterior
pituitary of the fetus. This may occur through the binding of CRH with
fetal adrenal CRH receptors. Some CRH secreted by the fetal
hypothalamus may also stimulate fetal ACTH production. The result is an
increased production of cortisol by the fetal adrenal. Fetal cortisol
blocks the inhibitory effect of progesterone on placental CRH
production. Earlier in pregnancy, the high levels of progesterone acted
as a brake on the secretion of placental CRH and fetal cortisol by
competing with cortisol for the glucocorticoid receptors
(Karalis et al. 1996
, Majzoub and Karalis 1999
). As cortisol concentrations rise, the hormone binds to
these placental glucorticoid receptors, for which it has a 2–4 times
greater affinity than does progesterone (Ojasoo et al. 1988
). This positive feedback loop causes additional production
of placental CRH and fetal cortisol.
The increasing concentrations of placental CRH and of ACTH from the fetal pituitary stimulate the fetal adrenal to produce dehydroepiandrosterone sulfate, which is converted to estrogen by the placenta. Changes occur as a result of the increase in estrogen concentrations. The cells of the myometrium synthesize connexin molecules, which move to the cell membrane and connect the cells electrically so that they will contract synchronously during labor; muscle cells in the uterus produce large numbers of oxytocin receptors, necessary for this hormone to cause contraction of the cells during labor; in addition, the synthesis of prostaglandins by placental tissues overlying the cervix is increased, which induces the production of enzymes in the cervix that digest collagen and make the cervix flexible during delivery of the fetus.
Despite the high concentrations of CRH in maternal plasma, maternal
plasma concentrations of ACTH and cortisol remain relatively normal
during pregnancy. This is probably due to CRH-binding protein
(CRH-BP), which neutralizes CRH earlier in pregnancy. The mRNA for
CRH-BP is expressed in the placenta, decidua, myometrium and fetal
membranes during pregnancy and reduces the amount of active CRH in the
maternal circulation (Linton et al. 1988
). However,
during the third trimester, the concentration of CRH-BP declines to
about one third the peak concentration, falling by about 60% during wk
34 and 35 (Chrousos 1999
). This results in an increased
amount of free CRH and explains the substantial increase in CRH and
cortisol in late pregnancy. CRH and CRH-BP form a dimer complex
that may interact with receptors to produce the final biological effect
(Behan et al. 1993
). CHR-BP has also been shown to
reduce the CRH-induced contractile activity of myometrial strips
(Petraglia et al. 1995
). The binding of CRH to
CRH-BP likely also causes the complex to be cleared from the
circulation, explaining the fall in CRH-BP concentrations in late
pregnancy (Woods et al. 1994
).
The role of CRH in preterm labor.
The cause of preterm labor can be physiological, in which a factor required for normal initiation of labor appears too early in gestation, or pathological, in which the initiating factor is abnormal but may induce the normal physiologic mechanisms.
Physiological preterm labor.
Abnormal maternal CRH concentrations are highly associated with risk of
preterm or post-term labor, as would be expected from the preceding
discussion. More than 10 years ago, women in preterm labor were
reported to have high plasma concentrations of CRH compared with
control women at the same stage of gestation (Wolfe et al. 1988
). Subsequently, a large prospective cohort study showed
that women who had an abnormally high CRH concentration early in
pregnancy were highly likely to have a preterm delivery (McLean et al. 1995
). In fact, higher risk of either pre- or
post-term delivery could be predicted even at 16–20 wk of
gestation on the basis of higher or lower concentrations of CRH in
maternal serum, respectively. Regardless of the final concentrations,
CRH concentrations increase exponentially during pregnancy. This
sequence of events has been likened to a biological clock in the
fetoplacental unit and determines the duration of pregnancy from an
early stage (Smith 1999
). Higher concentrations of CRH
during labor also predict a shorter labor duration (McLean et al. 1995
).
Pathological causes of preterm labor. Documented causes of increased placental CRH secretion in humans include hypoxia, inflammatory cytokines, glucocorticoids, stress (including noninflammatory and inflammatory stress), preeclampsia and eclampsia.
Concentrations of CRH in maternal plasma were measured in one study of
abnormal pregnancies (Wolfe et al. 1988
). Levels were
not elevated in maternal diabetes but were significantly higher in
pregnancies with antepartum hemorrhage at 28 wk (but not later), during
the third trimester of twin pregnancies, and in women with
pregnancy-induced hypertension, premature rupture of the membranes
or preterm labor (with higher concentrations from at least as early as
28 wk). In some of the pregnancies, CRH concentrations were elevated 11
wk before any other symptoms appeared.
Normal CRH concentrations do not completely guarantee a normal delivery date because fetal infections and other problems can cause early delivery regardless of CRH, and there is considerable interindividual variability in normal concentrations. However, abnormal CRH concentrations are highly predictive of the duration of gestation.
Other actions of CRH.
As will be explained in more detail below, CRH plays a major role in
both the maternal and fetal stress systems. It has been estimated that
during a stressful event, the amount of CRH released is so high and the
available time for exposure to CRH-BP is so low that maternal CRH
will not be completely quenched by CRH-BP. This means that CRH can
still act as a stress hormone during pregnancy (Linton et al. 1990
). Additional actions of CRH include the following:
stimulation of prostaglandin F2
and
prostaglandin E2 production by fetal membranes;
potentiation of the action of oxytocin and
PGF2, which are uterotonics; and activation of
CRH receptors on the smooth muscle of the myometrium. CRH and oxytocin
interact synergistically in the presence of prostaglandins to cause
myometrial contractility. Elevated CRH concentrations also predicted a
significantly lower fetal heart rate reactivity in response to
vibroacoustic stimuli at 31–32 wk of gestation independently of the
length of gestation (Sandman et al. 1999
).
CRH in other species.
An understanding of the differences in CRH metabolism across species is important for the appropriate interpretation of the literature on experimental animals. In sheep, CRH is secreted from the hypothalamus of the fetal brain rather than by the placenta. The CRH induces the fetal pituitary to secrete ACTH and subsequently the fetal adrenal to produce cortisol, as in humans. However, there is no ovine CRH-BP. In primates, as in humans, most of the CRH comes from the placenta rather than the fetal brain, but CRH induces placental estrogen secretion through a different pathway. CRH increases exponentially in the plasma of chimpanzees in the last half of gestation as in humans, but concentrations fall during this time in other primates, including baboons.
|
Hormones that influence fetal growth |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
).
The insulin-like growth factor (IGF) system is probably the most
important with respect to fetal growth, and IGF-1 is the most important
hormone (Gluckman and Harding 1997
). Fetal IGF-1 is
secreted in response to fetal glucose concentrations (Oliver et al. 1993
). IGF directs nutrients to insulin-sensitive
tissues, promoting glycogen and fat storage in muscle, liver and
adipose tissue. Substrate availability is the main determinant of fetal
IGF-1 levels (Gluckman et al. 1983
). In animal models,
maternal starvation quickly lowers fetal IGF-1 concentrations and
subsequently fetal growth (Bassett et al. 1990
).
Maternal IGF-1, IGF-2 and insulin do not cross the placenta but they
may affect placental function; maternal plasma IGF-1 in rodents also
correlates with fetal growth (Mirlesse et al. 1993
).
Most newborn IUGR infants have low levels of IGF-1. In a comparison of
15 normal newborns and 30 with IUGR, the IUGR group had significantly
lower concentrations of IGF-1 and higher concentrations of
IGF-binding protein-3 (Cianfarani et al. 1998
). In a
case-control study of 76 full-term deliveries, of which 31 were
IUGR, cord blood concentrations of IGF-1, insulin and
thyroid-stimulating hormone were lower in the IUGR group, but
higher concentrations of growth hormone were present
(Nieto-Díaz et al. 1996
).
Cortisol inhibits longitudinal growth of the sheep fetus in late
gestation (Fowden et al. 1996
) and probably plays a
major role in regulating growth at this time. In sheep, preventing the
cortisol surge by fetal adrenalectomy abolished the normal slowing of
growth at term, and infusing cortisol earlier in gestation lowered the
rate of longitudinal growth to that normally seen in late pregnancy
(Fowden et al. 1996
). Over the entire gestation period,
mean plasma cortisol concentrations accounted for 40–50% of the
variation in fetal crown-rump length increment. Cortisol suppresses
the production of IGF-2 in fetal sheep (Li et al. 1993
).
It also appears to switch fetal cells from proliferation to
differentiation.
Fetal insulin production is also important for normal protein synthesis
in the fetus. IUGR fetuses have small Islets of Langerhans and lower
insulin concentrations. Some members of the growth hormone–prolactin
gene family are produced only by the placenta. These include placental
growth hormone, placental lactogens and prolactin-related proteins.
A placental growth hormone appears in maternal circulation by early in
the second trimester and its concentration rises until delivery. Plasma
from the mothers of IUGR infants have reportedly lower concentrations
of this hormone (Mirlesse et al. 1993
), although the
hormone does not appear in fetal circulation. Human placental lactogen
is detectable in maternal circulation by 6 wk of pregnancy and
increases up to term. It stimulates nitrogen retention and insulin
secretion and is lipolytic. Although it might maintain fetal glucose
during maternal starvation, its importance for fetal growth is not
clear. However, human placental lactogen may stimulate IGF production
by the fetus (Schocknecht et al. 1992
).
|
Mechanisms that might underlie iron’s effects on preterm delivery and fetal growth |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
).
By 18 wk of pregnancy, placental volume may already be inversely
correlated with maternal hemoglobin and serum ferritin concentrations,
even in industrialized countries. On the basis of this observation,
associations between maternal hemoglobin and iron status and factors
known to affect placental size, i.e., human chorionic gonadotropin and
human placental lactogen, were assessed during the first trimester of
pregnancy in 175 women who were consulting about pregnancy termination.
The average duration of gestation was 68 d. Maternal hemoglobin
was significantly negatively correlated with the levels of human
chorionic gonadotropin and human placental lactogen across the normal
hemoglobin range. Although serum ferritin concentrations were low in
21% of the women, there was no correlation with the hormone
concentrations. The authors suggest that the oxygen content of maternal
blood may have had an important influence on the development of the
placenta and subsequently on human chorionic gonadotropin and human
placental lactogen concentrations. There is no reason to believe that
increased human chorionic gonadotropin or human placental lactogen
concentration would have an adverse effect on pregnancy outcome. Because virtually nothing is known about the effects of iron deficiency or anemia on the biological mechanisms that regulate the duration of gestation and fetal growth, the discussion in this section focuses mainly on other factors and illnesses that are known to influence these mechanisms and that could plausibly explain any effects of anemia or iron deficiency. The postulated biological mechanisms are as follows.
Iron deficiency or anemia may increase the stress hormones, norepinephrine and cortisol.
Iron deficiency increases norepinephrine (NE) concentrations
(Dallman 1986
), as does hypoxia (Gülmezoglu et al. 1996
). Norepinephrine is a strong stimulus for the
release of CRH (Calogero et al. 1988
) and cortisol. The
CRH and locus ceruleus/NE (LC/NE) sympathetic systems respond similarly
to many of the same neurochemicals (Chrousos and Gold 1992
). Iron deficiency and anemia were not mentioned among
these, but this has not been studied. Norepinephrine infusion into
pregnant sheep caused a reduction in fetal protein synthesis and
accretion, indicating adverse effects on fetal growth (Milley 1997
).
There is virtually no information concerning the effect of iron
deficiency or anemia on cortisol secretion. In one rat study of the
effect of an iron-free diet, the iron deficiency caused stress, as
evidenced by a higher serum cortisol concentration (Campos et al. 1998
).
Chronic hypoxia activates the stress response.
Low hemoglobin concentrations can cause a state of chronic hypoxia,
which is presumably exacerbated in pregnancy when oxygen demands are
particularly high because of the metabolism of the mother and the
fetus. Although changes in transplacental oxygen transfer are
relatively small when maternal hemoglobin concentrations fall, oxygen
transfer to the fetus is probably reduced in anemic women
(Viteri 1994
). Infant birth weight was directly related
to calculated maternal arterial oxygen content in a study of women
living at high altitude in the United States (Moore et al. 1982b
). Moreover, either maternal or fetal hypoxia could
activate the stress response, described below.
Although there is no information on the effect of low hemoglobin concentrations, some information is available concerning the effects of other situations in which chronic hypoxia affects pregnancy outcome. These include animal models in which hypoxia is caused by reducing blood flow, using a hypobaric chamber, high altitude, smoking and hemoglobinopathies. Fetal hypoxia can also be caused by factors that reduce maternal uteroplacental circulation. For example, preeclampsia with chronic hypertension produces a low partial pressure of oxygen in the fetus as well as reduced glucose and fetal glycogen stores.
Hypoxia-induced stress responses.
Stress may be defined as "a state of threatened homeostasis, which is
reestablished by a complex repertoire of physiologic and behavioral
adaptive responses of the organism" (Chrousos 1998
).
CRH plays a major role in the response to stress as well as being a
major player in fetal development and delivery, and although it has
never been tested directly, it is plausible that iron deficiency
creates a stress response. As stated by Chrousos (1998)
,
"the systems responsible for reproduction, growth and immunity are
directly linked to the stress system, and each is profoundly influenced
by the effectors of the stress response."
In the nonpregnant individual, the stress system is located in both the
central nervous system and the periphery. Its role is to create
physiological changes that redirect energy, oxygen and nutrients to the
central nervous system and stressed body sites, and behavioral changes
such as fight-or-flight responses. The stress system is complex; its
description is beyond the scope of this review. More detailed reviews
are available (Chrousos 1998
, Chrousos et al.
1992
, Stratakis et al. 1995
). Systems of
relevance here include the following: the CRH system, which is found
throughout the brain (particularly in the hypothalamus and the medulla)
and works synergistically with arginine-vasopressin neurons of the
hypothalamus; the LC/NE system in the medulla and pons of the brain,
and their peripheral effectors including the
hypothalamic-pituitary-adrenal axis; the efferent
sympathetic/adrenomedullary system; and components of the
parasympathetic system, which coordinate the stress response.
CRH is the primary regulator of the hypothalamic-pituitary-adrenal
axis, and administration of CRH can produce most of the physiological
and behavioral responses to stress, including stimulation of ACTH
production by the pituitary and glucocorticoid production by the
adrenals. The LC/NE system produces epinephrine and norepinephrine.
There are numerous interactions among the components of the stress
system, including a positive feedback loop between the CRH and
LC-NE/sympathetic systems so that activation of one system causes
activation of the other. As discussed above, iron deficiency causes an
increased production of norepinephrine, and norepinephrine is a strong
stimulus of CRH secretion (Calogero et al. 1988
).
The fetal hypothalamic-pituitary-adrenal axis is highly responsive to stress, which leads to the increased release of glucocorticoids as a central adaptive mechanism. The glucocorticoids cause the catabolism of fat, glycogen and protein, with a subsequent increase in blood glucose concentrations. Over time, chronically elevated glucocorticoid concentrations can, however, lead to impaired tissue growth and muscle atrophy.
The stress system is closely integrated with other parts of the central nervous system that regulate reproduction, growth and immunity, as well as behavior and emotion.
Hypoxia induces maternal and fetal stress, and the release of CRH.
"Hypoxia stimulates CRH production in cultured (placental)
cells ... which may be associated with the increased concentration
of CRH found in the blood of women with toxemia of pregnancy and in the
umbilical circulation of subjects with IUGR and reduced umbilical
artery blood flow." This statement was made in a review by
Smith (1998)
, a pioneer in CRH research; unfortunately,
no references to the research supporting this statement were included.
Creating a chronic or acute hypoxic state is the classical way in which
to study the effects of stress in pregnant animal models. Hypoxia is
induced in various ways, including partial ligation of the uterine
arteries, hypobaric chambers and partial occlusion of the umbilical
artery. In pregnant sheep, acute short-term hypoxia (hours to 2 wk)
causes an increase in ACTH and cortisol within 3 h. Concentrations
stay elevated for at least 7 h, falling to normal after 16 h
(Challis et al. 1986
).
When reduced uterine blood flow was used to lower the fetal arterial
oxygen saturation in sheep (Hooper et al. 1990
), there
was a seven- to eightfold increase in fetal plasma epinephrine after
2 h that normalized by 12 h. Fetal NE, cortisol and
prostaglandin E2 were increased three- to
fourfold after 2 h and remained higher during at least the next
12 h (Ducsay 1998
). Before 126 d of gestation
in a similar hypoxic sheep model (the normal duration of gestation in
the sheep is 150 d), there was no significant effect of hypoxia on
fetal plasma noradrenaline concentrations, but after this,
concentrations increased more than sixfold compared with controls, and
adrenaline increased fourfold (Coulter et al. 1990
). In
a sheep model, fetal hypoxia also reduced the transport of amino acids
to the fetus (Milley 1988
).
Stress is associated with IUGR. CRH is released from the hypothalamus in all individuals in response to stress in addition to being produced by the placenta during pregnancy. During pregnancy, the normal stress signal may be amplified by the placental release of CRH. Placental CRH and hypothalamic CRH are similar in structure. Plasma concentrations are nondetectable in nonpregnant adults.
Is placental CRH also stimulated by chronic fetal stress? To
investigate this question, CRH was measured in the cord blood of IUGR
(at 26–40 wk) and normal infants matched by gestational age
(Goland et al. 1986
). The mean umbilical cord plasma CRH
concentration was 206 ± 26 pmol/L in the IUGR infants and 49
± 17 pmol/L in the appropriate-for-gestational-age infants, a
significant difference. Similarly, mean plasma ACTH was significantly
higher in the IUGR samples (5.7 ± 1.2 vs. 3.3 ± 0.7
pmol/L). Mean cortisol concentrations did not differ, but the mean
dehydroepiandrosterone sulfate level was significantly lower in the
IUGR group (4.8 ± 0.6 vs. 7.7 ± 0.6 µmol/L).
CRH concentrations were significantly correlated with both ACTH and
cortisol concentrations and negatively correlated with
dehydroepiandrosterone sulfate levels. Although cesarean section and
maternal hypertension both increased CRH in the IUGR group (but not the
appropriate-for-gestational-age group), concentrations were still
elevated (to a mean of 131 µmol/L) in the IUGR infants
delivered vaginally. The authors concluded that chronic stress
stimulated CRH production by the placenta and may have affected fetal
pituitary-adrenal function in these high risk pregnancies. The
nature of this stress was not identified, and it is not clear whether
it originated from the mother or the fetus. The negative association
between CRH and fetal androgen secretion was reported in other studies
in which there was intrauterine stress, such as pregnancy hypertension
(Parker et al. 1986
).
In a comparison of IUGR and appropriate-for-gestational-age infants in
Italy, a strong inverse correlation (r = -0.54,
n = 30) was observed between length gain in the first 3
mo of life and cortisol concentrations at birth. In the
appropriate-for-gestational age group, cortisol was inversely
correlated with IGF-1 (n = 15, r = -0.75, P < 0.002) and positively associated with IGF
binding protein-1 (r = 0.52, P < 0.05)
but no associations between cortisol and IGF-factors were seen in
the IUGR group (Cianfarani et al. 1998
).
Maternal stress is associated with preterm delivery.
In a test of the general hypothesis that maternal stress is associated
with elevated maternal plasma concentrations of CRH and a subsequently
higher risk of preterm birth, Hobel et al. (1999b)
measured CRH at 18–20, 28–30 and 35–36 wk in 524 ethnically and
socioeconomically diverse pregnant women from California. Of this
group, 18 women had spontaneous onset of preterm labor. Their samples
were compared with those of 18 control women who delivered at term,
matched by age, previous birth outcome, smoking status and race. The
preterm delivery cases had significantly higher plasma CRH and
adrenocorticotropic hormone at all three gestational periods and higher
cortisol concentrations at 18–20 and 28–30 wk. Maternal stress level
at 18–20 wk, assessed by interview, predicted a significant amount of
variance in CRH at 28–30 wk gestation, with CRH at 18–20 wk
controlled for. The authors concluded that maternal stress and CRH
concentrations may be good markers for risk of preterm birth.
In a prospective study of 8719 Danish women with singleton pregnancies,
those who reported one or more highly stressful life events had a 1.76
times greater risk of preterm delivery than those without stressful
events (Hedegaard et al. 1996
). The association was
strongest for events occurring between 16 and 30 wk. Sandman et al. (1997)
found that higher maternal stress at 28–30 wk of
gestation was a significant predictor of both gestational age at birth
and birth weight. Maternal stress in the third trimester was associated
with higher levels of maternal ACTH and cortisol.
Pregnancy outcome in other hypoxic situations.
Pregnant women who are used to living at high altitude have a
substantially higher incidence of IUGR but not of preterm deliveries
(McCullough and Reeves 1977
, Moore et al. 1982b
, Sabrevilla et al. 1968
). The same is true
in sheep kept in conditions of chronic high altitude hypoxia
(Ducsay 1998
). The higher risk of IUGR occurs even
though there are maternal and fetal adaptations to high altitude,
including maternal hyperventilation and increased hemoglobin
concentrations (Moore et al. 1982a
). At high altitude,
placentas have more fetal capillaries at the periphery of the villi,
larger intervillous spaces, a reduced volume of villi, and in some
studies, an increase in placental size (Jackson et al. 1988
). In anemia, there is also an inverse correlation between
maternal hemoglobin concentration with placental volume that is evident
by at least 18 wk of gestation (Godfrey et al. 1991
).
Not all infants born at high altitude are small. At an elevation over
3000 m in the United Statues, three maternal characteristics were
identified as related to birth weight: hypoventilation, no increase in
ventilation and arterial oxygen saturation from early to late gestation
and a falling hemoglobin concentration during pregnancy that leads to a
lower arterial oxygen content in the last trimester (Moore et al. 1982a
). This study illustrates the limited adaptability to
lower maternal arterial oxygen concentration and suggests a strong
influence of the oxygen concentration on birth weight.
The partial pressure of oxygen is lower in the blood of smokers. In
women who smoke, the placenta is smaller and there is a reduction in
the capillary volume of the villi, as well as other adverse changes
(Jauniaux and Burton 1992
). Smoking is the most
important predictor of IUGR in developing countries (Kramer 1998
), where it explains much of the disparity in IUGR
prevalence among women of lower and upper socioeconomic status.
In preeclampsia with hypertension, the partial pressure of oxygen in
maternal and fetal blood is low. Women with preeclampsia have higher
concentrations of CRH and lower concentrations of CRH-BP
(Perkins et al. 1995
). CRH concentrations are elevated
before the development of preeclampsia. In a relatively small sample,
18 patients with spontaneous preterm delivery and 23 patients who
developed pregnancy hypertension were matched to women who delivered at
term. The preterm delivery cases had significantly higher
concentrations of CRH and significantly lower CRH-BP–CRH dimer complex
concentrations at 18–20, 28–30 and 35–36 wk of gestation. The
hypertension cases also had elevated CRH as early as 18–20 wk (but not
as high as the preterm delivery cases) and low CRH-BP (but not as
low as the preterm cases) (Hobel et al. 1999a
).
Iron-deficiency anemia may increase oxidative stress.
The fetoplacental unit is very susceptible to oxidative damage induced
by reactive oxygen species. Oxidative stress is one mechanism thought
to cause preeclampsia, pregnancy-induced hypertension and
pregnancy-induced diabetes (Cester et al. 1994
,
Poranen et al. 1996
). Interest is increasing in the
possibility that antioxidant nutrients might improve pregnancy outcome
by reducing oxidative stress (Scholl and Stein 2000
,
West et al. 2000
).
The lipids of the brush border membrane of the placental syncitioblast are susceptible to peroxidation by being in contact with oxidants in the maternal blood. Scavengers of highly reactive oxygen species, including antioxidant enzymes such as superoxide dismutase, catalase and glutathione reductase, protect against peroxidation. These enzymes inhibit lipid peroxidation. Placental oxidant-antioxidant imbalance might cause release of products of lipid peroxidation into the fetal circulation with subsequent damage to endothelial cell membranes.
Iron deficiency may cause increased oxidative stress because it causes
erythrocytes to be more susceptible to oxidative damage. Erythrocytes
are normally protected from oxidative stress caused by free radicals
released from the potentially dangerous combination of iron and oxygen.
Mechanisms that achieve this protection include superoxide dismutase,
reduced glutathione and catalase. However, microcytic erythrocytes have
a shorter survival time partly because of oxidative damage to the
erythrocyte membrane (Diez-Ewald and Layrisse 1968
,
Vetore and Tedesco 1975
). In addition, in
ß-thalassemia, for example, excess globin chains (due to low
hemoglobin) autoxidize and release heme and produce superoxide eight
times faster than does normal hemoglobin. In a comparison of control
subjects with those with iron-deficiency anemia and ß-thalassemia
trait, the group with iron-deficiency anemia had significantly
higher malondialdehyde (by 50%) and superoxide dismutase (by 40%)
concentrations per unit hemoglobin. Also, concentrations of erythrocyte
pyrimidine 5'-nucleotidase, the enzyme most sensitive to sulfhydryl
group damage in vivo, are lower in iron-deficiency anemia
(Vives Corrons et al. 1995
). In 26 Israeli patients with
iron-deficiency anemia compared with 10 healthy control subjects,
erythrocytes had higher levels of reduced glutathione and normal
malondialdehyde concentrations (Bartal et al. 1993
).
Iron-deficient cells were more sensitive than control cells to high
concentrations of peroxides.
When the effect of iron deficiency on lipid peroxidation and
antioxidant enzyme activity was examined in rats, the
iron-deficient group had significantly lower liver production of
malondialdehyde and activity of superoxide dismutase and higher
catalase activity (Rao and Jagadeesan 1996
). When
treated with a carcinogen (dimethyl hydrazine, which produces active
oxygen species), unlike controls, the iron-deficient rats did not
respond with an increase in glutathione peroxidase activity. Similarly,
superoxide dismutase activity fell much more in the iron-deficient
group than in iron-replete controls when treated with the
carcinogen. The investigators speculated that the iron-sufficient
rats were better able to eliminate oxygen radicals, whereas iron
deficiency increased susceptibility to oxidative stress.
Iron deficiency may increase the risk of maternal infections.
There is some evidence that iron deficiency adversely affects immune function. For example, it can alter the proliferation of T and B cells, reduce the killing activity of phagocytes and neutrophils, and lower bactericidal and natural killer cell activity.
Infection is one of the main pathological risk factors for preterm
labor. The presence of bacteria or inflammatory cytokines in amniotic
fluid or chorioamnionitic membranes is strongly associated with preterm
labor and premature rupture of the membranes. The bacteria are believed
to come from the vagina. Early bacterial vaginosis (before 16 wk) is
associated with a relative risk of preterm delivery of 5–7.5. If this
condition occurs after 26 wk, the risk is 1.4–1.9 (Kurki et al. 1992
). There is potential for CRH to regulate inflammatory
responses and vice versa. The cytokine interleukin-1 stimulates
production of CRH, and CRH in turn regulates cytokine production by
immune effector cells. Because maternal stress is associated with
preterm birth, abnormalities in the regulation of CRH and the
production of inflammatory cytokines may be a mechanism that could form
the pathophysiological basis for this association.
Falkenberg et al. (1999)
examined the effect of maternal
infections on the fetal hypothalamic-pituitary-adrenal axis. Subjects
were 361 women with normal pregnancy (including some with preterm
delivery) and 110 with infections. Cord blood was analyzed at delivery,
which occurred between 24 and 44 wk of gestation. The infants born to
women with infections were born 1.5 wk earlier, and cord blood
concentrations of cortisol and dehydroepiandrosterone sulfate were
significantly higher. The authors suggested that products of the
activated immune system of mothers with infections may have crossed the
placenta and activated the fetal hypothalamic-pituitary-adrenal axis.
Cortisol has been reported to inhibit the activity of natural killer
cells both in vitro and in vivo. In a study of the effect of delivery
method on cortisol in cord blood, natural killer cell activity in the
cord blood of the group with the low cortisol concentrations (because
of cesarean delivery and general anesthesia) was twice that of the two
groups with higher cortisol concentrations (De Amici et al. 1999
).
|
Implications for future studies |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
Dr. Rasmussen: The low birth weight or intrauterine growth retardation rate varies considerably worldwide as does the rate of iron deficiency, but the preterm rate does not vary a lot from country to country. So, I am having trouble linking those two. If I have iron status that varies a lot and I have a preterm rate that is not so variable, I have trouble figuring out how the two of those might be associated. Can you speak to that at all?
Dr. Allen: I agree. There is about double the preterm rate in developing countries, but you would think if iron deficiency caused that much preterm, you would see an enormous preterm rate. Remember that preterm rates are counted only in babies weighing less than 2500 g. My suspicion is that there are a lot of children born between 2500 and 3000 g or 3100 g who might be slightly preterm or preterm. They might be born earlier than usual and, therefore, have a slightly lower birth weight. You will not pick it up in low-birth-weight statistics.
Dr. Rasmussen: No, but if you use decent low birth weights and decent statistics on intrauterine growth retardation, low birth weight varies significantly around the world when preterm is much more constant.
Dr. Allen: Right, but I am saying that it does not have to be preterm by that much. I think we lack really good data, especially from places with a high prevalence of iron deficiency, such as south Asia. So, I just do not think we quite know yet. I agree that it is something that has to be sorted out.
|
FOOTNOTES |
|---|
2 This article was commissioned by the World
Health Organization (WHO). The views expressed are those of the author
alone and do not necessarily reflect those of WHO.
3 Abbreviations used: ACTH, adrenocorticotropic
hormone; CRH, corticotropin-releasing hormone; CRH-BP,
CRH-binding protein; IGF, insulin-like growth factor; IUGR,
intrauterine growth retardation; LC, locus ceruleus; NE,
norepinephrine;
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONAssociations between iron...Prevalence of preterm delivery...Risk factors for low...Biological mechanisms involved...The importance of corticotropin...Hormones that influence fetal...Mechanisms that might underlie...Implications for future studiesDISCUSSIONREFERENCES |
|---|
1. Bakketeig L. Ultrasound dating of pregnancies changes dramatically the observed rates of pre-term, post-term, and small-for-gestational-term. Iatrogenics 1991;1:174-175
2. Bartal M., Mazor D., Dvilansky A., Meyerstein N. Iron deficiency anemia: recovery from in vitro oxidative stress [see comments]. Acta Haematol 1993;90:94-98[Medline]
3. Bassett N. S., Oliver M. H., Breier B. H., Gluckman P. D. The effect of maternal starvation on plasma insulin-like growth factor I concentrations in the late gestation ovine fetus. Pediatr. Res. 1990;27:401-404[Medline]
4. Behan D. P., Potter E., Sutton S., Fischer W., Lowry P. J., Vale W. W. Corticotropin-releasing factor-binding protein. A putative peripheral and central modulator of the CRF family of neuropeptides. Ann. N.Y. Acad. Sci. 1993;697:1-8
5. Calogero A. E., Gallucci W. T., Chrousos G. P., Gold P. W. Catecholamine effects upon rat hypothalamic corticotropin-releasing hormone secretion in vitro. J. Clin. Investig. 1988;82:839-846
6. Campos M. S., Barrionuevo M., Alférez M. J., Gómez-Ayala A. E., Rodríguez-Matas M. C., Lopez Aliaga I., Lisbona F. Interactions among iron, calcium, phosphorus and magnesium in the nutritionally iron-deficient rat. Exp. Physiol. 1998;83:771-781[Abstract]
7. Cester N., Staffolani R., Rabini R. A., Magnanelli R., Salvolini E., Galassi R., Mazzanti L., Romanini C. Pregnancy induced hypertension: a role for peroxidation in microvillus plasma membranes. Mol. Cell. Biochem. 1994;131:151-155[Medline]
8. Challis J. R., Richardson B. S., Rurak D., Wlodek M. E., Patrick J. E. Plasma adrenocorticotropic hormone and cortisol and adrenal blood flow during sustained hypoxemia in fetal sheep. Am. J. Obstet. Gynecol. 1986;155:1332-1336[Medline]
9. Chrousos, G. P. (1998) Stressors, stress, and neuroendocrine integration of the adaptive response. The 1997 Hans Selye Memorial Lecture. Ann. N.Y. Acad. Sci. 851: 311–335.
10. Chrousos G. P. Reproductive placental corticotropin-releasing hormone and its clinical implications. Am. J. Obstet. Gynecol. 1999;180:S249-S250[Medline]
11. Chrousos G. P., Gold P. W. The concepts of stress and stress system disorders. Overview of physical and behavioral homeostasis [published erratum appears in J. Am. Med. Assoc. 268: 200]. J. Am. Med. Assoc. 1992;267:1244-1252[Abstract]
12. Cianfarani S., Germani D., Rossi L., Argirò G., Boemi S., Lemon M., Holly J. M., Branca F. IGF-I and IGF-binding protein-1 are related to cortisol in human cord blood. Eur. J. Endocrinol. 1998;138:524-529[Abstract]
13. Coulter C. L., Giraud A. S., Hooper S. B., Parker L., McMillen I. C. Fetal asphyxia stimulates an increase in fetal plasma catecholamines and [Met]-enkephalin-arg6-phe7 in the late-gestation sheep fetus. J. Dev. Physiol. 1990;14:267-272[Medline]
14. Dallman P. R. Biochemical basis for the manifestations of iron deficiency. Annu. Rev. Nutr. 1986;6:13-40[Medline]
15. De Amici D., Gasparoni A., Chirico G., Ceriana P., Bartoli A., Ramajoli I., Moretta A. Natural killer cell activity and delivery: possible influence of cortisol and anesthetic agents. A study on newborn cord blood. Biol. Neonate 1999;76:348-354[Medline]
16. Diez-Ewald M., Layrisse M. Mechanisms of hemolysis in iron deficiency anemia. Further studies. Blood 1968;32:884-894
17. Ducsay C. A. Fetal and maternal adaptations to chronic hypoxia: prevention of premature labor in response to chronic stress. Comp. Biochem. Physiol. A Mol. Integr. Physiol. 1998;119:675-681[Medline]
18. Emanuel R. L., Robinson B. G., Seely E. W., Graves S. W., Kohane I., Saltzman D., Barbieri R., Majzoub J. A. Corticotrophin releasing hormone levels in human plasma and amniotic fluid during gestation. Clin. Endocrinol. 1994;40:257-262[Medline]
19. Falkenberg E. R., Davis R. O, DuBard M., Parker C. R., Jr Effects of maternal infections on fetal adrenal steroid production. Endocr. Res. 1999;25:239-249[Medline]
20. Fowden A. L., Szemere J., Hughes P., Gilmour R. S., Forhead A. J. The effects of cortisol on the growth rate of the sheep fetus during late gestation. J. Endocrinol. 1996;151:97-105[Abstract]
21. Gluckman P. D., Harding J. E. Fetal growth retardation: underlying endocrine mechanisms and postnatal consequences. Acta Paediatr. (suppl.) 1997;422:69-72
22. Gluckman P. D., Johnson-Barrett J. J., Butler J. H., Edgar B. W., Gunn T. R. Studies of insulin-like growth factor-I and -II by specific radioligand assays in umbilical cord blood. Clin. Endocrinol. 1983;19:405-413[Medline]
23. Godfrey K. M., Redman C. W., Barker D. J., Osmond C. The effect of maternal anaemia and iron deficiency on the ratio of fetal weight to placental weight. Br. J. Obstet. Gynaecol. 1991;98:886-891[Medline]
24. Goland R. S., Wardlaw S. L., Stark R. I., Brown L. S., Jr, Frantz A. G. High levels of corticotropin-releasing hormone immunoactivity in maternal and fetal plasma during pregnancy. J. Clin. Endocrinol. Metab. 1986;63:1199-1203[Abstract]
25. Gülmezoglu A. M., Mahomed K., Hofmeyr G. J., Nikodem V. C., Kramer T. Fetal and maternal catecholamine levels at delivery. J. Perinat. Med. 1996;24:687-691[Medline]
26. Hedegaard M., Henriksen T. B., Secher N. J., Hatch M. C., Sabroe S. Do stressful life events affect duration of gestation and risk of preterm delivery?. Epidemiology 1996;7:339-345[Medline]
27.
Hobel C. J., Arora C. P., Korst L. M. Corticotrophin-releasing hormone and CRH-binding protein. Differences between patients at risk for preterm birth and hypertension. Ann. N.Y. Acad. Sci. 1999a;897:54-65
28. Hobel C. J., Dunkel-Schetter C., Roesch S. C., Castro L. C., Arora C. P. Maternal plasma corticotropin-releasing hormone associated with stress at 20 weeks’ gestation in pregnancies ending in preterm delivery. Am. J. Obstet. Gynecol. 1999b;180:S257-S263[Medline]
29.
Hooper S. B., Coulter C. L., Deayton J. M., Harding R., Thorburn G. D. Fetal endocrine responses to prolonged hypoxemia in sheep. Am. J. Physiol. 1990;259:R703-R708
30. Institute of Medicine Preventing Low Birth Weight 1985 National Academy Press Washington, DC.
31. Jackson M. R., Mayhew T. M., Haas J. D. On the factors which contribute to thinning of the villous membrane in human placentae at high altitude. II. An increase in the degree of peripheralization of fetal capillaries. Placenta 1988;9:9-18[Medline]
32.
Jauniaux E., Burton G. J. The effect of smoking in pregnancy on early placental morphology. Obstet. Gynecol. 1992;79:645-648
33. Karalis K., Goodwin G., Majzoub J. A. Cortisol blockade of progesterone: a possible molecular mechanism involved in the initiation of human labor. Nat. Med. 1996;2:556-560[Medline]
34. Kramer M. S. Socioeconomic determinants of intrauterine growth retardation. Eur. J. Clin. Nutr. 1998;52:S29-S33
35.
Kramer M. S., McLean F. H., Eason E. L., Usher R. H. Maternal nutrition and spontaneous preterm birth. Am. J. Epidemiol. 1992;136:574-583
36.
Kramer M. S., Platt R., Yang H., Joseph K. S., Wen S. W., Morin L., Usher R. H. Secular trends in preterm birth: a hospital-based cohort study. J. Am. Med. Assoc. 1998;280:1849-1854
37.
Kurki T., Sivonen A., Renkonen O. V., Savia E., Ylikorkala O. Bacterial vaginosis in early pregnancy and pregnancy outcome. Obstet. Gynecol. 1992;80:173-177
38. Li J., Saunders J. C., Gilmour R. S., Silver M., Fowden A. L. Insulin-like growth factor-II messenger ribonucleic acid expression in fetal tissues of the sheep during late gestation: effects of cortisol. Endocrinology 1993;132:2083-2089[Abstract]
39. Lin C. C., Santolaya-Forgas J. Current concepts of fetal growth restriction: part I. Causes, classification, and pathophysiology. Obstet. Gynecol. 1998;920:1044-1055
40. Linton E. A., Behan D. P., Saphier P. W., Lowry P. J. Corticotropin-releasing hormone (CRH)-binding protein: reduction in the adrenocorticotropin-releasing activity of placental but not hypothalamic CRH. J. Clin. Endocrinol. Metab. 1990;70:1574-1580[Abstract]
41. Linton E. A., Wolfe C. D., Behan D. P., Lowry P. J. A specific carrier substance for human corticotrophin releasing factor in late gestational maternal plasma which could mask the ACTH-releasing activity. Clin. Endocrinol. 1988;28:315-324[Medline]
42. Lobel M., Dunkel-Schetter C., Scrimshaw S.C.M. Prenatal maternal stress and prematurity: a prospective study of socioeconomically disadvantaged women. Health Psychol 1992;11:32-40[Medline]
