Content uploaded by Steven R Seidner
Author content
All content in this area was uploaded by Steven R Seidner on Mar 21, 2018
Content may be subject to copyright.
Direct fetal glucocorticoid treatment alters postnatal
adaptation in premature newborn baboons
M. GORE ERVIN,1STEVEN R. SEIDNER,2M. MICHELLE LELAND,3
MACHIKO IKEGAMI,1AND ALAN H. JOBE1
1Perinatal Research Laboratories, Departments of Obstetrics and Gynecology and Pediatrics,
University of California, Los Angeles School of Medicine, Harbor-UCLA Medical Center,
Torrance, California 90502; 2Department of Pediatrics, University of Texas Health Science Center
at San Antonio, San Antonio 78284-7812; and 3Department of Physiology and Medicine,
Southwest Foundation for Biomedical Research, San Antonio, Texas 78227
Ervin, M. Gore, Steven R. Seidner, M. Michelle Le-
land, Machiko Ikegami, and Alan H. Jobe. Direct fetal
glucocorticoidtreatmentalterspostnataladaptationinprema-
ture newborn baboons. Am. J. Physiol. 274 (Regulatory
Integrative Comp. Physiol. 43): R1169–R1176, 1998.—
Abnormalities of premature newborn adaptation after pre-
term birth result in significant perinatal mortality and
morbidity. We assessed the effects of short-term (24 h) fetal
betamethasone exposure on preterm newborn baboon pulmo-
nary and cardiovascular regulation and renal sodium han-
dling during the first 24 h after birth. Male fetal baboons
(Papio) (124-day gestation, term 185 days) received ultra-
sound-guided intramuscular injections of saline (n55) or
betamethasone (0.5 mg/kg; n55). Fetuses were cesarean
delivered 24 h later, treated with 100 mg/kg surfactant, and
ventilated by adjusting peak inspiratory pressures to main-
tainPCO2values of 35–50 mmHg for 24h. Betamethasone- vs.
saline-treated mean 6SE newborn body weights (0.45 60.02
vs.0.4160.01kg)weresimilar.Although prenatal betametha-
sone did not affect postnatal lung function (PCO2, arterial/
alveolar O2gradient, or dynamic compliance), plasma hor-
mone (cortisol or thyroxine), or catecholamine levels, mean
arterial pressure (25 61 vs. 32 61 mmHg), plasma sodium
concentration (132 62 vs. 138 61 meq/l), glomerular
filtration rate (0.07 60.02 vs. 0.16 60.02 ml·min21·kg21),
and renal total sodium reabsorption (1.5 60.5 vs. 16.0 63.0
µeq·min21·kg21) values were significantly lower in saline-
treated than in betamethasone-treated newborns at 24 h. We
conclude that despite the fact that there are no pulmonary
and endocrine effects, antenatal glucocorticoid exposure al-
ters premature newborn baboon vascular and renal glomeru-
lar function and improves sodium reabsorption after preterm
delivery.
lung; kidney; blood pressure; betamethasone
THE MAJOR CAUSES OF INFANT morbidity and mortality in
the United States are diseases associated with prema-
turity. Because lung immaturity is the principal con-
tributor to premature newborn mortality, the lungs
have been the primary focus of strategies to improve
premature newborn survival (7). Although glucocorti-
coid-induced fetal lung maturation was demonstrated
in a large clinical trial over two decades ago (20),
widespread application of this therapy has occurred
only recently (12, 13). In addition to their pulmonary
effects, prenatal glucocorticoids improve postnatal out-
comes with their pleiotropic effects, including reduced
incidences of patent ductus arteriosus and intraven-
tricular hemorrhage and improved postnatal blood
pressure regulation (22). We have demonstrated that
antenatal glucocorticoid treatment significantly en-
hances renal sodium reabsorption in premature new-
born sheep (9). Kidney immaturity and the resulting
inability to limit excessive fluid and electrolyte losses is
a common problem in the clinical management of the
preterm newborn and is a significant source of postna-
tal morbidity (12, 13). Thus, in contrast to term new-
borns in which postnatal renal adaptations include
marked increases in glomerular filtration rate (GFR)
and sodium reabsorption (33), preterm newborns (par-
ticularly those ,30 wk gestation) do not appropriately
increase renal sodium reabsorption at birth (3) and
retain a fetal pattern of excessive natriuresis. The
effects of prenatal glucocorticoids on postnatal adapta-
tion have been examined in detail in preterm sheep,
and limited, nonrandomized observations regarding
steroid effects on renal function in preterm newborns
are available. However, renal function and its hor-
monal regulation have not been evaluated in prema-
ture humans or other primates. Preterm baboons can
be delivered as early as 125 days (0.67 of total length of
gestation, with term 185 days) and will survive with
the development of bronchopulmonary dysplasia if
surfactant treated at birth (29). The purpose of the
present study was to determine the effect of antenatal
steroid administration on postnatal lung, kidney, and
endocrine function in the premature newborn primate.
METHODS
Animal Selection and Fetal Treatment
The fetal treatments and delivery studies were performed
at the Southwest Foundation for Biomedical Research, San
Antonio, Texas. All animal husbandry, animal handling, and
procedures were reviewed and approved to conform with the
American Association forAccreditation of Laboratory Animal
Care guidelines as detailed in the Guide for the Care and Use
ofLaboratoryAnimals (National Research Council). Pregnan-
cies were dated using cycle dates and growth parameters
from prenatal ultrasounds at 70 and 120 days estimated fetal
gestational age. The pregnant baboons were sedated with
intramuscular ketamine (10 mg/kg) for each of the prenatal
ultrasounds. Amniocentesis was performed under ultrasound
guidance at 70 days fetal gestational age, and Y-specific DNA
amplification of cultured amniocytes was used to determine
fetal gender (27). Male fetuses were used in an attempt to
minimize variation due to possible differential maturation
between males and females. At 124 62 days gestation (term
185 days), the fetuses were again imaged using ultrasound
0363-6119/98 $5.00 Copyright r1998 the American Physiological Society R1169
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (154.016.160.096) on March 21, 2018.
Copyright © 1998 American Physiological Society. All rights reserved.
after maternal sedation with 10 mg/kg ketamine. Each
animal was randomly assigned to receive betamethasone
(Celestone Soluspan; Schering Pharmaceuticals, Kenilworth,
NJ) at a fetal dose of 0.5 mg/kg estimated fetal weight in 1 ml
or an equivalent volume of vehicle (0.15 M saline). Each
treatment was injected under ultrasound guidance into the
fetal thigh with a 22-gauge spinal needle. The female baboon
was then returned to her cage.
Delivery
The pregnant baboons were sedated with ketamine (10
mg/kg im) 24 h after fetal injection, intubated, and anesthe-
tized with 1.5% halothane. The preterm fetuses were deliv-
ered by cesarean section, weighed, and intubated using 2.0-
or 2.5-mm endotracheal tubes. The newborns received 100
mg/kg surfactant (Survanta, Ross Products, Columbus, OH)
by tracheal instillation, and ventilation was initiated using
pressure-limited infant ventilators. After delivery of the fetus
and repair of the maternal incisions, recovery of the female
baboons was monitored daily for 2 wk, with all animals
released to outside gang cages after 4 wk.
Management
The newborn baboons were maintained sedated with intra-
muscular ketamine (10 mg/kg) and intravenous diazepam
(0.1–0.2 mg/kg) if needed. An arterial catheter was placed
either by percutaneous insertion into the radial artery or via
an umbilical artery into the descending aorta for blood
pressure monitoring and blood gas sampling. A deep venous
catheter was placed percutaneously via the saphenous vein
into the inferior vena cava for administration of fluids and
drugs. All animals received a single intravenous injection of
[3H]inulin (6 µCi) for measurements of GFR (9). The animals
were cared for on servo-controlled infrared warmers. The
animals were not fed and were given parenteral fluids
containing amino acids and multivitamins. Intravenous flu-
ids were administered with appropriate electrolytes at 12.5
ml·kg21·h21initially, with infusion rates increased when
heart rates were .180 beats/min or for hematocrit increases
of .10% or with a small cardiac silhouette on the chest
radiograph and an increasing base deficit. Sodium bicarbon-
ate was administered (2 meq/kg) when the base deficit
exceeded 28 meq/kg. Ampicillin (50 mg·kg21·day21in 2
divided doses) and gentamicin (5 mg·kg21·day21in 2 divided
doses) were given intravenously. Local anesthesia with 2%
lidocaine and additional ketamine were administered for any
invasive procedures.
Blood sampling and cardiovascular measurements. Inaddi-
tion to routine arterial blood samples for assessments of pH,
PO2,PCO2, and hematocrit, blood samples (4–5 ml) were
collected from the umbilical cord at delivery and at 2, 6, 12,
18, and 24 h for plasma electrolytes, osmolality, [3H]inulin
concentrations, and hormone analysis. Not all measurements
were made at all times to minimize blood sampling. An
additional plasma sample (1 ml) for determination of plasma
catecholamine levels was collected 30 min after delivery.
Blood gases and hematocrit values also were monitored at
regular intervals. Blood samples were replaced volumetri-
cally with heparinized adult baboon blood. Arterial blood
pressure, heart rate, oxygen saturation, and electrocardio-
gram were monitored continuously with a Gould P23 pres-
sure transducer (Gould Instrument Systems, Cleveland, OH)
and a cardiorespiratory monitor (model 700; Biomedical
Systems, Branford, CT).
Pulmonary management and measurements. To standard-
ize management, ventilatory rate was held constant at 40
breaths/min, and the peak end-expiratory pressure was set at
4 cmH2O, with an inspiratory time of 0.6 s. Arterial PCO2
values within the target range of 35–45 mmHg were main-
tained by adjusting peak inspiratory pressures. Oxygenation
was regulated by adjusting inspired oxygen content. The
arterial/alveolar (A/a) gradient was calculated as PO2/[(FIO23
713) 2(PCO2/0.8)], where FIO2is the fractional concentrat-
ion of inspired O2, 713 is the atmospheric pressure corrected
for the partial pressure of water vapor at physiological
temperature, and 0.8 is the respiratory quotient. Tidal vol-
umesanddynamiccompliancesweremeasuredwithaVT1000
Vital Station Neonatal Plethysmograph (Vitaltrends Technol-
ogy, Wallingford, CT). The dynamic compliance was calcu-
lated as C 5dP/dV, where dP is the difference between the
airway pressures at the end and beginning of inspiration and
dV is the difference between the airway volumes at the end
and beginning of inspiration.
At 22 h, each newborn received a single intravascular
injection of 125I-labeled albumin for postnatal assessment of
lung protein leak. After 24 h of ventilation, each baboon
received 50 mg/kg pentobarbital sodium to achieve deep
anesthesia and was ventilated for 2 min with 100% O2. The
endotracheal tube was disconnected from the ventilator to
allow passive deflation of the lungs and was clamped. Cardiac
activity continued for 2 min to permit absorption atelectasis
andwasfollowedbypentobarbital(50mg/kg)andexsanguina-
tion. Static deflation pressure-volume curves were measured
in situ in the open chest by inflating the lungs with air to 35
cmH2O, followed by sequential volume measurements with
incremental decreases in pressure as previously outlined in
premature newborn lambs (6). The lungs were then removed,
weighed, and thoroughly lavaged with cold saline (16). Alveo-
lar wash protein and lung homogenate protein and hemoglo-
bin levels were assessed to determine postnatal lung protein
leak (17).
Renalmeasurements.Urinesampleswerecollectedcontinu-
ously into an inverted syringe barrel placed around the
newborn penis, and the total volume of urine produced was
measured at 4-h intervals during the 24 h of study. Urine
samples were assessed for osmolality, electrolytes (Na, K, and
Cl) and [3H]inulin specific activity. The GFR was determined
from the calculated plasma [3H]inulin clearance.
Analytic Techniques
Blood pH, PO2, and PCO2values were determined with a
model 995 blood gas analyzer (AVL Scientific, Roswell, GA).
Plasma and urine osmolalities were measured by freezing
point depression (Advanced Digimatic Osmometer, model
MO; Advanced Instruments, Needham Heights, MA). Blood
and urine electrolyte (Na, K, and Cl) concentrations were
determined with a Nova electrolyte analyzer (Nova Biomedi-
cals, Waltham, MA). Plasma and urine [3H]inulin specific
activities were assessed by measuring radioactivity in ali-
quots (0.1 ml) of plasma and urine.
Blood samples were divided immediately after withdrawal
into chilled test tubes, vortexed, and centrifuged immedi-
ately. Plasma aliquots were frozen (220°C) for determina-
tions of cortisol, thyroxine (T4), triiodothyronine (T3), argi-
ninevasopressin(AVP), aldosterone, and plasma renin activity
(PRA) (lithium heparin, 40 µg/ml blood), angiotensin II (ANG
II), and atrial natriuretic factor (ANF) levels (aprotinin, 500
KIU/ml blood and K2EDTA, 1 mg/ml) and catecholamines (4
mmol/l EGTA and 3 mmol/l reduced glutathione). Plasma
cortisol,T4,andT3levelsweredeterminedwithchemilumines-
cence kits (Nichols Diagnostics, San Juan Capistrano, CA)
standardized for fetal plasma. Plasma AVP extraction and
radioimmunoassay (RIA) were performed as previously de-
R1170 GLUCOCORTICOIDS AND PRETERM NEWBORN ADAPTATION
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (154.016.160.096) on March 21, 2018.
Copyright © 1998 American Physiological Society. All rights reserved.
scribed (10, 36); assay sensitivity is 0.8 pg of AVP per tube,
with intra- and interassay coefficients of variation of 6 and
9%,respectively. PlasmaANF levels were determined by RIA,
with an assay sensitivity of 2 pg/tube and intra-assay and
interassaycoefficientsof variation of 11and 13%, respectively
(11). Plasma ANG II levels were determined from the ANF
plasma extracts by use of RIA kits obtained from Peninsula
Laboratories (Belmont, CA). Intra-assay and interassay coef-
ficients of variation for the ANG II assay averaged 6 and 9%,
respectively, with an overall assay sensitivity of 2 pg/tube.
PRA was determined with commercially available reagents,
and plasma aldosterone levels were determined by use of RIA
kits obtained from ICN Radiochemicals (Costa Mesa, CA).
Plasma catecholamine levels were determined by radioenzy-
matic assay (30).
Data Analysis
All values are expressed as means 6SE. Differences over
time and differences between saline and betamethasone-
treated groups were assessed by two-way repeated-measures
analysis of variance (ANOVA), with time as the between-
subjects factor and treatment as the among-subjects factor.
Multiple comparisons to identify differences among groups
were conducted with the Student-Neuman-Keuls procedure.
Paired or unpaired t-tests were also used as appropriate.
Statistical significance was accepted at P,0.05.
RESULTS
Pulmonary
There were no differences in mean body weights
betweenthecontrol (0.41 60.01 kg)andbetamethasone-
treated (0.45 60.02 kg) preterm newborn baboons.
Indexes of ventilatory status (arterial PCO2, A/a gradi-
ent, and dynamic compliance) did not differ between
control and betamethasone-treated preterm newborn
baboons over the 24-h study period (Fig. 1). At 24 h,
there were no differences, respectively, between control
and betamethasone-treated values for arterial pH
(7.27 60.03 vs. 7.27 60.03), PO2(59 68 vs. 62 64
mmHg), PCO2(43 65 vs. 42 62 mmHg), peak inspira-
tory pressure (25 62 vs. 25 62 cmH2O), or maximal
lung volumes measured at 35 cmH2O (17 64 vs. 15 64
ml/kg body wt). Radiolabeled albumin accumulated
equivalently in alveolar wash (0.6%) and lung tissue
(,2%), indicating minimal development of lung injury.
Dry-to-wet lung weights were 0.141 60.119 and
0.14160.005 g/g in control andbetamethasone-treated
animals, respectively. Prenatal betamethasone expo-
sure did not alter overall postnatal lung function.
Cardiovascular, Electrolytes, and Catecholamines
Mean blood pressure values were not different be-
tween control and betamethasone-treated newborns 2
h after delivery (Fig. 2). However, by 12 h, mean blood
pressurewas significantly higher inthe betamethasone-
treated group and remained higher at 24 h (Fig. 2).
Mean heart rate values significantly increased in both
groups between 2 and 12 h of life and remained
elevated at 24 h. Over the 24-h observation period,
least-squares regression analysis revealed an inverse
relationship between heart rate and mean blood pres-
sure in both groups, suggesting the presence of an
intact baroreflex response in these very premature
newborn baboons. Although heart rate values were
significantly higher in the control animals after 12 h of
ventilation, heart rate values were not different be-
tween the two study groups at 24 h.
At delivery, umbilical cord blood hematocrit and
plasma osmolality, sodium, and potassium values were
similar between control and betamethasone-treated
preterm newborn baboons (Table 1). However, plasma
chloride concentrations were significantly higher in the
betamethasone-treated lambs. Because plasma sodium
and chloride concentrations decreased significantly in
Fig. 1. Arterial PCO2, arterial/alveolar (A/a) gradient, and dynamic
compliance in control and betamethasone-treated premature new-
born baboons ventilated for 24 h.
Fig. 2. Mean arterial blood pressure, heart rate, and plasma sodium
values in control and betamethasone-treated premature newborn
baboons ventilated for 24 h. *Different from 2-h value, P,0.05;
**different from control, P,0.05.
R1171GLUCOCORTICOIDS AND PRETERM NEWBORN ADAPTATION
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (154.016.160.096) on March 21, 2018.
Copyright © 1998 American Physiological Society. All rights reserved.
the control newborns during the 24 h of ventilation,
plasmasodium and chloride concentrationswere signifi-
cantly higher in the betamethasone-treated baboons
relative to controls (Fig. 2 and Table 1). There were no
differences in total fluid administered, hematocrit,
plasma osmolality, or plasma potassium concentrations
betweencontrol and betamethasone-treated baboons at
any time during the 24 h of ventilation.
Mean umbilical cord plasma epinephrine and norepi-
nephrine levels were low and were not different be-
tweencontrol and betamethasone-treatedanimals (Fig.
3). Plasma epinephrine and norepinephrine levels sig-
nificantly increased in both groups by 30 min after
delivery, remained elevated at 2 h, and were further
elevated by 24 h. Plasma epinephrine and norepineph-
rine levels were similarly elevated in both groups from
2to24h.
Renal
The measurements of renal function are summarized
in Table 2 and Fig. 4. During the first 12 h, GFR values
were not different between groups. By 16 h, GFR was
significantlyhigherin the betamethasone-treatedlambs
and remained above the control values at 24 h. Urine
flow values also were significantly higher in the beta-
methasone-treated baboons during the final urine col-
lection period (20–24 h), and urine osmolalities and
sodium concentrations were significantly lower than in
the control animals (Table 2). In the betamethasone-
treated baboons, total sodium reabsorption was signifi-
cantlyhigher (Fig. 4), and osmolar clearance per100 ml
GFR values were significantly below (Table 2) the
control animal values. Free water clearance also was
significantly elevated in the betamethasone-treated
Table 1. Control and betamethasone-treated preterm
newborn baboon plasma osmolality, electrolyte, and
hormone values at delivery and after 24 h of ventilation
Control Betamethasone
Cord 24 h Cord 24 h
Hematocrit, % 43.460.7 40.861.5 46.060.9 41.262.5
Osmolality, mos-
mol/kgH2O 29265 28965 29461 29463
Sodium, meq/l 13762 13262* 14161 13861†
Potassium, meq/l 4.460.2 4.560.3 4.960.6 4.360.2
Chloride, meq/l 10861.2 10461* 11261 111 61†
Triiodothyronine,
µg/dl 55.168.2 41.263.9 66.9611.1 48.663.6
Atrial natriuretic
factor, pg/ml 376863614 17610 83613*
Plasma renin
activity, ng ANG
I·ml21·h216.461.8 131627* 19610 90631*
Aldosterone, pg/ml 65617 90629 2156104 398690†
Values are means 6SE; n55 animals for both control and
betamethasone. *Different from umbilical cord (Cord) value, P,
0.05; †different from control value, P,0.05.
Fig. 3. Plasma epinephrine and norepinephrine levels in control and
betamethasone-treated premature newborn baboons ventilated for
24 h.
Table 2. Control and betamethasone-treated
preterm newborn baboon renal parameters
after 24 h of ventilation
Control Betamethasone
Urine
Flow, ml·min21·kg212.8461.4 5.5961.25*
Osmolality, mosmol/kgH2O 30866 213633*
Sodium, meq/l 1376688614*
Potassium, meq/l 7.1360.46 5.0461.09
Potassium excretion,
µeq·min21·kg213.0461.06 4.4861.01
Osmolar clearance, ml/100 ml
GFR 44610 2065*
Free water clearance, ml/min 0.01560.017 0.06360.016*
Values are means 6SE; n55 animals for both control and
betamethasone. GFR, glomerular filtration rate. *Different from
control, P,0.05.
Fig. 4. Glomerular filtration rate (GFR) and total sodium reabsorp-
tion in premature newborn baboons delivered 24 h after fetal saline
or betamethasone (0.5 mg/kg) treatment and ventilated for 24 h.
*Different from control, P,0.05.
R1172 GLUCOCORTICOIDS AND PRETERM NEWBORN ADAPTATION
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (154.016.160.096) on March 21, 2018.
Copyright © 1998 American Physiological Society. All rights reserved.
animals (Table 2).Although control animal urine potas-
sium concentration and urinary potassium excretion
values were 40% higher and 32% lower, respectively,
than the betamethasone-treated animal values, the
values were not statistically different between groups
(Table 2).
Endocrine
Plasma cortisol levels were similar between groups
at delivery (Fig. 5). By 24 h, there were similar signifi-
cant decreases in plasma cortisol levels in both groups.
T4levels also were similar between control and beta-
methasone-treated newborns at delivery, and in both
groupsplasma T4levels significantly decreased.Umbili-
cal cord plasma T3levels were similar in both groups
after delivery and did not change in either group at
24 h.
Umbilical cord plasma endocrine values including
AVP, ANF, and ANG II and aldosterone and PRA values
were not different between control and betamethasone-
treated newborns (Fig. 5 and Table 1). Relative to the
umbilical cord values, plasma AVP, ANG II, and PRA
levelsincreased significantly in both groupsat 24 h, but
there were no differences between control and beta-
methasone-treated groups. Plasma ANF levels were
not different between groups at delivery. Although
there was a statistically significant increase in plasma
ANF levels in the betamethasone-treated group by 24
h, plasma ANF levels did not differ between groups.
Mean umbilical cord and 24-h plasma aldosterone
levels did not differ between groups. However, plasma
aldosterone levels were significantly elevated in the
betamethasone-treated newborns compared with con-
trols at 24 h (Table 1).
DISCUSSION
In these very premature newborn baboons given
prenatal glucocorticoids as a single fetal dose 24 h
before delivery, an increase in postnatal blood pressure
and improved kidney function were evident after 12 h
postnatal age and were sustained to 24 h of postnatal
life. However, the anticipated effect of improved lung
function consistently observed after prenatal glucocor-
ticoid exposure in preterm sheep was absent. Several
possible explanations for this apparent absence of
pulmonary effects in the premature baboon model
warrant consideration. First, the present preterm new-
born baboon studies are unique because the animals
were studied very early in gestation (0.67), earlier than
previously reported for most other species. In preterm
sheep, prenatal glucocorticoids improve renal, cardio-
vascular, and lung function after maternal or fetal
treatments after ,0.8 of gestation (6, 32). Postnatal
effects of prenatal glucocorticoid treatment before 120
days (0.8) gestation have not been evaluated in sheep
because survival is limited. Thus pulmonary effects
may not have been detected in the present studies
because the preterm baboons at 0.67 of gestation were
too immature to respond. In apparent contradiction to
this hypothesis, striking alterations in lung structure
follow preterm or term delivery of monkeys exposed to
high-dose glucocorticoids over 3 days at midgestation
(4).Although speciesdifferencesintiming of responsive-
ness might be one explanation, a second possibility is
that a treatment-to-delivery interval of 24 h may be
insufficient in the baboon to manifest lung matura-
tional effects. In sheep delivered at 0.85 of gestation,
glucocorticoid-induced effects on lung function occur
within 15 h of fetal or maternal treatment with 0.5
mg/kg betamethasone. In both baboons and monkeys,
glucocorticoids also enhance lung function, but only
paradigmsof 3 daysof prenatal maternalglucocorticoid
exposure have been studied (18, 19). Thus glucocorti-
coid treatment-to-delivery intervals of .24 h may be
necessary to initiate improvements in lung function in
primates. In favor of this explanation is the lack of
Fig. 5. Plasma cortisol, thyroxine (T4), arginine vaso-
pressin (AVP), and ANG II concentrations from umbili-
cal cord (Cord) blood and after 24 h of ventilation in
control and betamethasone-treated (0.5 mg/kg) prema-
ture newborn baboons. *Different from Cord value, P,
0.05.
R1173GLUCOCORTICOIDS AND PRETERM NEWBORN ADAPTATION
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (154.016.160.096) on March 21, 2018.
Copyright © 1998 American Physiological Society. All rights reserved.
effect of prenatal glucocorticoids on the incidence or
severity of respiratory distress syndrome in the human
(7) and the observation that the blood pressure and
renal effects detected in the current studies were not
apparent until 12 h after delivery. It is possible that the
postnatal cardiovascular and renal responses observed
were delayed relative to the time of delivery because of
the short treatment-to-delivery interval.
A third explanation for the absence of pulmonary
effects is that lung function in both glucocorticoid-
treatedand control animalswas favorably influencedto
asimilar degree by the stress associated withrepetitive
maternal anesthesia and ultrasound examination. In
theovine model, a single maternalexposure to anesthe-
sia and laparotomy can increase circulating fetal corti-
sol levels sufficiently to induce fetal lung maturation
(34). Thus, in our attempt to optimize the quality
control of the pregnant baboons through ultrasound
examinations performed at 70 and 120 days for mea-
surements of fetal growth, and again at 124 days for
fetal therapy, the procedures alone may have been
sufficient to evoke a stress response in both groups.
Consistent with this hypothesis, the umbilical cord
cortisol levels (Fig. 5) in the control animals were high
relative to reported newborn sheep or human values
and also higher than anticipated values for early
gestationcontrols in this species, giventhe limited fetal
potential for cortisol production at this early gestation
(25). Another indicator consistent with chronic fetal
stress was the high umbilical cord plasma T3value.
Thus handling and anesthesia of the pregnant baboons
may have been sufficient to evoke similar elevations in
cortisol and T3values in both control and betametha-
sone-treated animals, thereby enhancing lung matura-
tion in both groups. The predominant source of the
cortisol in the umbilical cord blood may have been from
the mother via placental transfer (25). Assuming the
latter hypothesis is correct (induced lung maturation in
both groups), it is interesting to note that a single fetal
dose of betamethasone was associated with marked
effects on preterm newborn blood pressure and kidney
function.
Despite a lack of differences in lung function, the
betamethasone-treated preterm baboons were charac-
terized by significantly higher mean arterial blood
pressure values by 12 h after delivery, and this trend
continued through 24 h. The higher plasma sodium
concentration in these animals may have been suffi-
cient to sustain intravascular volume better than that
of the control animals (22). Because fluid administra-
tion was similar in both groups, the higher plasma
sodium concentration and larger urinary losses in the
betamethasone-treated animals suggest that dehydra-
tion might account for the higher plasma sodium
concentration. However, the similar hematocrit values
and higher rate of sodium reabsorption suggest that
overall sodium management was altered in favor of a
higherplasma sodium concentration.Thus one specula-
tion is that glucocorticoid-induced changes in vascular
permeability may have contributed to a decrease in
extravascular fluid volume, consistent with the postna-
tal adaptation pattern in term newborns.
Glucocorticoids also affect vascular reactivity and
vasomotortone.For example, glucocorticoidadministra-
tion increases overall cardiac contractility, vascular
tone, and cardiovascular receptor expression for cate-
cholamines and ANG II (28, 35). In premature newborn
sheep,single-dosefetal or maternal prenatalbetametha-
sone treatment increases mean arterial blood pressure
despitecausing marked suppression of circulatingcate-
cholamines and ANG II levels. Thus, in the lamb,
betamethasone-inducedincreases in catecholamine and
ANG II receptor expression and/or postreceptor re-
sponse mechanisms may augment overall vascular
responsiveness to circulating vasoactive agents even
when circulating levels are reduced. However, the
response pattern was different in these premature
newborn baboons. The betamethasone-treated new-
borns had higher mean blood pressure values with no
differences in circulating catecholamine and ANG II
levels between betamethasone and control groups.
Rather than being suppressed as in the lamb, catechol-
amine and ANG II levels increased over 24 h.
At birth, the kidneys must rapidly shift from a fetal
pattern of high fractional excretion of water and so-
diumto highly efficientwater and sodium conservation.
In term human newborns, adaptations including
marked increases in GFR and sodium reabsorption are
well underway within 24–48 h after delivery. Although
preterm newborns also will eventually adapt appropri-
ately if maintained viable for a sufficient period of time
(3), delays in shifting from the fetal pattern of excessive
water and sodium losses relative to the low GFR often
predispose preterm infants to hyponatremia and hypo-
volemia.Glucocorticoid-induced kidney maturation(in-
creases in GFR and sodium reabsorption) has been
reported previously in fetal (15) and near-term new-
bornsheep (31) inresponse to prolonged cortisol admin-
istration. We have also reported increases in GFR and
sodium reabsorption in preterm newborn lambs deliv-
ered24 or 48 hafter single fetalor maternal betametha-
sone treatment (2, 9). An acute effect of glucocorticoids
to increase GFR is widely accepted and principally
reflects selective renal near-parallel afferent and effer-
ent vessel vasodilation to increase glomerular plasma
flow and filtration fraction (1). Although postnatal
increases in GFR also reflect increases in filtration
fraction, the exact mechanism(s) have not been defined.
The two- to threefold increase in GFR noted in the
betamethasone-treated baboons (Fig. 4) is not attribut-
able to the higher blood pressure in these animals
because blood pressure did not increase concurrently
with the increase in GFR.
Prenatal glucocorticoid treatment in sheep also in-
creases postnatal sodium reabsorption (31, 32). Proxi-
mal rather than distal tubular function appears to be
the principal location limiting fetal and perhaps pre-
termnewborn sodium reabsorption (21).Glucocorticoid-
inducedincreases in preterm newbornsodium reabsorp-
tion may reflect a direct effect to increase proximal
tubular sodium transporter function and expression
R1174 GLUCOCORTICOIDS AND PRETERM NEWBORN ADAPTATION
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (154.016.160.096) on March 21, 2018.
Copyright © 1998 American Physiological Society. All rights reserved.
(14) and an indirect effect to increase basolateral
Na-K-ATPase activity (8). Thus glucocorticoid-induced
changes in proximal tubular function may also be
important in the premature newborn baboon. However,
thesignificantlyhigher plasma aldosteronelevels(Table
1) measured in the betamethasone-treated fetuses sug-
gest that increases in distal tubular function may have
contributed to the overall increase in renal sodium
reabsorption (Fig. 4). Given the higher mean blood
pressure and GFR observed in the betamethasone-
treated premature newborn baboons and the marked
suppression of the renin-ANG II-aldosterone axis ob-
served in newborn lambs after prenatal glucocorticoid
treatment, the basis for the heightened renin-ANG
II-aldosteroneaxis activity measuredin the betametha-
sone-treated preterm newborns is not clear. Nonethe-
less, the present results indicate that glucocorticoid
exposure for as little as 24 h before delivery can
significantly improve preterm newborn baboon renal
sodium handling and plasma electrolyte regulation.
The umbilical cord plasma cortisol levels were not
suppressed in the betamethasone-exposed fetuses, a
surprising result suggesting lack of feedback inhibi-
tion, a long plasma cortisol half-life at this gestation, or
perhapssimply placental cortisol transfer.Thevery low
umbilical cord plasma catecholamine and AVP levels
(Figs. 3 and 5) indicate that the preterm baboons were
not ‘‘stressed’’at the time of delivery. The absence of an
effect of antenatal betamethasone exposure on umbili-
cal cord plasma ANG II levels is consistent with fetal
sheepdatademonstrating that cortisol-induced suppres-
sion of fetal angiotensinogen production is gestation
dependent (23). The pronounced increases in catechol-
amines, AVP, and ANG II levels and PRA at 24 h
indicate the significant stress associated with preterm
delivery and postnatal ventilation. In contrast, cortisol
and T4levels had decreased similarly in both groups by
24 h, with no changes in plasma T3levels. In fetal
sheep, the thyroid axis response to glucocorticoids
differs in that T3levels increase with little change in T4
levels (5, 26). However, a thyroid response to exogenous
corticosteroidexposure is variably seen in humans (24).
Although changes in cortisol and thyroid hormone
metabolism were likely important to the changes in
circulating levels noted between delivery and 24 h of
postnatal life, these patterns did not appear to be
influencedby prenatal betamethasone exposure(Fig. 5).
In summary, these are unique data in very preterm
baboons for which there is no comparison information
available. A limitation to studies in the baboon is the
maternal response to handling and the need for seda-
tion and/or anesthesia associated with procedures.
Nonetheless, these studies provide the first integrative
information regarding premature newborn pulmonary,
cardiovascular, renal, and endocrine responses during
postnatal adaptation and the effects of fetal therapy in
the severely premature (0.67 gestation) newborn pri-
mate. A single fetal dose of betamethasone as little as
24 h before delivery can have pronounced effects on
preterm newborn GFR, sodium reabsorption, plasma
sodium regulation, and blood pressure stability. The
interpretative difficulty is that we do not know how the
fetal environment (high cortisol and T3) may have
modulated the betamethasone effects. This difficulty is
also common to clinical situations in which many
preterm deliveries are associated with chronic and/or
acutefetal stress. Theclinically important conclusionis
that prenatal glucocorticoids may augment postnatal
adaptation even if the fetus has been stressed in utero
and is at a decreased risk of respiratory distress
syndrome.
Perspectives
Thenowwidely accepted practiceofprenatal glucocor-
ticoid administration in cases of threatened premature
labor has profoundly improved postnatal outcomes in
premature infants. Nonetheless, prematurity-associ-
ateddiseases remain the leadingcause of infant morbid-
ity and mortality in the United States. Studies assess-
ing prenatal glucocorticoid exposure in the premature
newborn model may provide insight into why some
infants apparently fail to respond to prenatal glucocor-
ticoid exposure. In addition, glucocorticoids influence a
diverse array of systems and functions. However, the
actual effect(s) and the sequence of events that improve
or augment postnatal adaptation are not known. Thus
studies of preterm newborn adaptation and the diverse
effects of prenatal glucocorticoid exposure on this pro-
cess represent an important step toward optimizing
therapeutic approaches to improve postnatal outcomes
in premature infants.
This work was supported in part by National Heart, Lung, and
Blood Institute Grants HL-052635 and HL-53636 (The Southwest
Foundation for Biomedical Research BPD Resource Center) and an
Established Investigatorship Award to M. G. Ervin from the Ameri-
can Heart Association.
Address for reprint requests: M. G. Ervin, Dept. of Biology, Box 60,
Middle Tennessee State Univ., Murfreesboro, TN 37132.
Received 15 May 1997; accepted in final form 12 January 1998.
REFERENCES
1. Baylis, C., R. K. Handa, and M. Sorkin. Glucocorticoids and
control of glomerular filtration rate. Semin. Nephrol. 10: 320–
329, 1990.
2. Berry, L. M., D. H. Polk, M. Ikegami, A. H. Jobe, J. F.
Padbury, and M. G. Ervin. Preterm newborn renal and cardio-
vascular responses after fetal or maternal antenatal betametha-
sone. Am. J. Physiol. 272 (Regulatory Integrative Comp.Physiol.
41): R1972–R1979, 1997.
3. Bueva, A., and J. P. Guignard. Renal function in preterm
neonates. Pediatr. Res. 36: 572–577, 1995.
4. Bunton, T. E., and C. G. Plopper. Triamcinolone-induced
structural alterations in the development of the lung of the fetal
rhesus macaque. Am. J. Obstet. Gynecol. 148: 203–215, 1984.
5. Celsi, G., W. Zheng-Ming, G. Akusjarvi, and A. Aperia.
Sensitive periods for glucocorticoids’ regulation of Na1,K1-
ATPase mRNA in the developing lung and kidney. Pediatr. Res.
33: 5–9, 1993.
6. Chen, C. M., M. Ikegami, D. Polk, and A. H. Jobe. Fetal
corticosteroid and T4treatment effects on lung function of
surfactant-treated preterm lambs. Am. J. Respir. Crit. Care Med.
151: 21–26, 1995.
7. Crowley, P., I. Chalmers, and M. Keirse. The effects of
corticosteroid administration before preterm delivery: an over-
view of the evidence from controlled trials. Br. J. Obstet. Gynae-
col. 97: 11–25, 1990.
R1175GLUCOCORTICOIDS AND PRETERM NEWBORN ADAPTATION
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (154.016.160.096) on March 21, 2018.
Copyright © 1998 American Physiological Society. All rights reserved.
8. Dobrovic-Jenik, D., and S. Milkovic. Regulation of fetal
Na1,K1-ATPase in rat kidney by corticosteroids. Biochim. Bio-
phys. Acta 942: 227–235, 1988.
9. Ervin, M. G., L. M. Berry, M. Ikegami, A. H. Jobe, J. F.
Padbury, and D. H. Polk. Single dose fetal betamethasone
administration stabilizes glomerular filtration rate and alters
endocrine function in premature lambs. Pediatr. Res. 40: 645–
651, 1996.
10. Ervin, M. G., R. D. Leake, M. G. Ross, and D. A. Fisher.
Arginine vasotocin in ovine maternal and fetal blood, fetal urine
and amniotic fluid. J. Clin. Invest. 75: 1696–1701, 1985.
11. Ervin, M. G., K. A. Terry, G. Calvario, R. Castro, M. G. Ross,
R. D. Leake, and D. A. Fisher. Vascular effects alter preterm
fetal renal responses to vasopressin. Am. J. Physiol. 266 (Regula-
tory Integrative Comp. Physiol. 35): R722–R729, 1994.
12. Gilstrap, L. C. Effect of corticosteroids for fetal maturation on
perinatal outcomes, February 28–March 2, 1994. Am. J. Obstet.
Gynecol. 173: 246–252, 1995.
13. Goldenberg, R. L. Effects of corticosteroids for fetal maturation
on perinatal outcomes. JAMA 273: 413–418, 1995.
14. Guillery, E. N., L. P. Karniski, M. S. Mathews, W. V. Page, J.
Orlowski, P. A. Jose, and J. E. Robillard. Role or glucocorti-
coids in the maturation of renal cortical Na1/H1exchanger
activity during fetal life in sheep. Am. J. Physiol. 268 (Renal
Fluid Electrolyte Physiol. 37): F710–F717, 1995.
15. Hill, K. J., E. R. Lumbers, and I. Elbourne. The actions of
cortisol on fetal renal function. J. Dev. Physiol. (Eynsham) 10:
85–96, 1988.
16. Ikegami, M., D. Polk, B. Tabor, J. Lewis, T. Yamada, and A.
Jobe. Corticosteroid and thyrotropin-releasing hormone effects
on preterm sheep lung function. J. Appl. Physiol. 70: 2268–2278,
1991.
17. Jobe, A. H., D. H. Polk, M. G. Ervin, J. F. Padbury, C. M.
Rebello, and M. Ikegami. Preterm betamethasone treatment
of fetal sheep: outcome after term delivery. J. Soc. Gynecol.
Invest. 3: 250–258, 1996.
18. Kessler, D. L., W. E. Truog, J. H. Murphy, S. Palmer, T. A.
Standaert, D. E. Woodrum, and W. A. Hodson. Experimental
hyaline membrane disease in the premature monkey; effects of
antenatal dexamethasone. Am. Rev. Respir. Dis. 126: 62–69,
1982.
19. Kotas, R. V., O. R. Kling, M. F. Block, J. F. Soodsma, R. D.
Harlow, and W. M. Crosby. Response of immature baboon fetal
lung to intra-amniotic betamethasone. Am. J. Obstet. Gynecol.
130: 712–717, 1978.
20. Liggins, G. C., and R. N. Howie. A controlled trial of antepar-
tum glucocorticoid treatment for prevention of the respiratory
distress syndrome in premature infants. Pediatrics 50: 515–520,
1972.
21. Lumbers, E. R., K. J. Hill, and V. J. Bennett. Proximal and
distal tubular activity in chronically catheterized fetal sheep
compared with the adult. Can. J. Physiol. Pharmacol. 66: 697–
702, 1987.
22. Moise, A. A., M. E. Wearden, C. A. Kozinetz, A. L. Gest, S. E.
Welty, and T. N. Hansen. Antenatal steroids are associated
withlessneed for blood pressure support in extremelypremature
infants. Pediatrics 95: 845–850, 1995.
23. Olson, A. L., S. Perlman, and J. E. Robillard. Developmental
regulation of angiotensinogen gene expression in sheep. Pediatr.
Res. 28: 183–185, 1990.
24. Padbury, J. F., M. G. Ervin, and D. H. Polk. Extrapulmonary
effects of antenatally administered steroids. J. Pediatr. 128:
167–172, 1996.
25. Pepe, G. J., and E. D. Albrecht. Actions of placental and fetal
adrenal steroid hormones in primate pregnancy. Endocr. Rev. 16:
608–648, 1995.
26. Polk, D. H., M. Ikegami, A. H. Jobe, J. Newham, P. Sly, K.
Rolland, and R. Kelly. Postnatal lung function in preterm
lambs: effects of a single exposure to betamethasone and thyroid
hormones. Am. J. Obstet. Gynecol. 172: 872–881, 1995.
27. Samollow, P. R., L. M. Cherry, S. M. Witte, and J. Rogers.
Interspecific variation at the Y-linked RPS4Y locus in hominoids:
implications for phylogeny. Am. J. Phys. Anthropol. 101: 333–
343, 1996.
28. Segar, J. L., K. Bedell, W. V. Page, J. E. Mazursky, A. M.
Nuyt, and J. E. Robillard. Effect of cortisol on gene expression
of the renin-angiotensin system in fetal sheep. Pediatr. Res. 37:
741–746, 1995.
29. Seidner, S., D. McCurnin, J. Coalson, D. Correll, and R.
Castro. A new model of chronic lung injury in surfactant-treated
preterm baboons delivered at very early gestations (Abstract).
Pediatr. Res. 33: 344A, 1993.
30. Stein, H. M., K. Oyama, A. Martinez, B. A. Chappell, E.
Buhl, L. Blount, and J. F. Padbury. Effects of corticosteroids
in preterm sheep on adaptation and sympathoadrenal mecha-
nisms at birth. Am. J. Physiol. 264 (Endocrinol. Metab. 27):
E763–E769, 1993.
31. Stonestreet, B. S., N. B. Hansen, A. R. Laptook, and W. Oh.
Glucocorticoid accelerates renal functional maturation in fetal
lambs. Early Hum. Dev. 8: 331–341, 1983.
32. Strandhoy, J. W., C. E. Giammattei, and J. C. Rose. Stimula-
tion of renal medullary cAMP production by vassopressin (AVP)
in fetal sheep (Abstract). FASEB J. 6: 1745A, 1992.
33. Svenningsen, N. W., and A. S. Aronson. Postnatal changes in
proximal and distal tubular sodium reabsorption in healthy
very-low-birth-weight infants. Biol. Neonate 25: 230–241, 1976.
34. Tabor, B. L., J. F. Lewis, M. Ikegami, D. Polk, and A. H.
Jobe. Corticosteroids and fetal intervention interact to alter
lung maturation in preterm lambs. Pediatr. Res. 35: 479–483,
1994.
35. Tseng, Y. T., M. A. Tucker, K. T. Kashiwai, J. A. Waschek,
and J. F. Padbury. Regulation of b1-adrenoceptors by glucocor-
ticoids and thyroid hormones in fetal sheep. Eur. J. Pharmacol.
289: 353–359, 1995.
36. Weitzman, R. E.,A. Reviczky, T. H. Oddie, and D. A. Fisher.
Effect of osmolality on arginine vasopressin and renin release
after hemorrhage. Am. J. Physiol. 238 (Endocrinol. Metab. 1):
E62–E68, 1980.
R1176 GLUCOCORTICOIDS AND PRETERM NEWBORN ADAPTATION
Downloaded from www.physiology.org/journal/ajpregu by ${individualUser.givenNames} ${individualUser.surname} (154.016.160.096) on March 21, 2018.
Copyright © 1998 American Physiological Society. All rights reserved.
