 |
|
 |
|
 |
 |
|
 |
 |
 |
 |
 |
 |
 |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
PEDIATRICS Vol. 107 No. 3 March 2001, pp. 494-498
From the Department of Pediatrics, University of Pennsylvania and Children's Hospital of Philadelphia, Philadelphia, Pennsylvania, and the North American Trial Collaborators.
 |
ABSTRACT |
|---|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
Background. It has been suggested that preterm infants may have developmental immaturity of the hypothalamic-pituitary-adrenal axis, and that decreased cortisol response to stress increases risk of chronic lung disease (CLD) secondary to inflammatory lung injury.
Methods. To investigate the relationship between endogenous corticosteroid and CLD, we measured plasma cortisol during the first 28 days of life in a subset of neonates in the North American Thyrotropin-Releasing Hormone (TRH) Collaborative Trial. Analyses were performed on 314 infants, 24 to 32 weeks' gestation, whose mothers received 1 or 2 courses of antenatal corticosteroids plus TRH or placebo.
Results. Mean cortisol was 3.1 µg/dL (range: 0.1-17.9) at birth, reached maximal levels at 24 hours (19.4 µg/dL, range: 0.8-124.6), and decreased to 5.9 µg/dL (range: 0.2-24.7) at 14 to 28 days of age; levels during the first week were not associated with gestational age. The Clinical Risk Index for Babies (CRIB), a neonatal assessment tool that is correlated with risk of mortality, was positively associated with cortisol level on days 1 and 3 through 7. TRH versus placebo treatment did not influence cortisol levels at any time point. To examine the relationship between cortisol and adverse outcome of death or CLD at 36 weeks' postmenstrual age (CLD36), logistic regression models adjusting for known contributing clinical factors (gestational age and CRIB score) were fit. There was a statistically borderline negative association between median cortisol level at 3 to 7 days and CLD36. After adjusting for gestational age and CRIB score, the predicted probability of CLD36 was only minimally influenced by the cortisol concentration.
Conclusion. In preterm infants, basal plasma cortisol concentration during the first week is a weak predictor for CLD36. Possible benefits as well as risks of supplemental, low-dose cortisol treatment of high-risk preterm infants remain to be determined. Key words: cortisol, premature infant, bronchopulmonary dysplasia, infant chronic lung disease, CRIB score.
Chronic lung disease (CLD), defined as a continued
requirement for supplemental oxygen beyond 36 weeks' postmenstrual age (PMA), is a major source of morbidity and mortality among preterm infants. Exposure to mechanical ventilation, elevated oxygen levels, and infection can trigger a cascade of inflammatory changes in the
immature lung. In the first days of life, neonates at risk of
developing CLD demonstrate increased proinflammatory chemokines, complement, and neutrophil counts in tracheobronchial
aspirate.1 These inflammatory changes, with associated
release of proteolytic enzymes, may damage the structural integrity of
the lung and promote tissue fibrosis.2
Preterm neonates may have developmental immaturity of the
hypothalamic-pituitary-adrenal axis as assessed by elevated cortisol precursors, low plasma cortisol, and/or decreased responsiveness to
adrenocorticotropic hormone (ACTH).3-7 It has been
proposed that an inability to secrete adequate amounts of cortisol in
response to the stress of illness contributes to the vulnerability of
preterm infants to inflammatory injury in the lung. Administration of
corticosteroids to ventilated preterm infants has been shown to
decrease markers of inflammation in tracheal aspirate
samples.8 However, although early postnatal
corticosteroids may initially improve pulmonary function and gas
exchange and decrease the risk of CLD, it is associated with adverse
effects.9
The hypothesis of this study was that some very preterm infants have
developmental immaturity of their hypothalamic-pituitary-adrenal axis
as reflected by low cortisol levels and a blunted increase in cortisol
with severe illness in the first week of life. We measured serial
plasma cortisol concentrations during the first 28 postnatal days in a
subset of 314 preterm neonates enrolled in the North American
Thyrotropin-Releasing Hormone (TRH) trial10 and examined
associations with initial illness severity (Clinical Risk Index for
Babies [CRIB])11 and predicted probability of CLD or
death at 36 weeks' PMA.
Study Population
Plasma cortisol levels were measured on 314 infants, with
gestational ages between 24 and 32 weeks, who were born to mothers enrolled in a multicenter, blinded, randomized trial of antenatal TRH
for the prevention of lung disease in preterm infants (North American
TRH Trial). The clinical outcome for infants in this trial has been
reported.10 Women were eligible for enrollment if they
were in active labor between 24 and 30 weeks' gestation. They were
excluded from the study if they had evidence of bleeding, infection
including clinical chorioamnionitis, hypertension, or known
life-threatening fetal anomalies. Women with premature rupture of
membranes, diabetes, and multiple gestations were not excluded from the
study.
The TRH treatment group received 400 µg of TRH (Thypinone, Abbott
Laboratories, Chicago, IL, or Thyrel, Ferring Laboratories, Tarreytown,
NY) administered intravenously in 50 mL of saline over 20 minutes;
unless delivery occurred, treatment was repeated at 8-hour intervals
for a total of 4 doses. Women in the placebo (control) group received
infusions of saline on the same schedule. Within 72 hours of the first
TRH infusion, all women received either betamethasone (Celestone
Soluspan, Schering Corporation, Kenilworth, NJ), given as two 12-mg
intramuscular injections at a 24-hour interval, or 6 mg of
dexamethasone given at 12-hour intervals for 4 doses. Additional
courses of antenatal corticosteroid treatment and their timing were
given at the discretion of the attending obstetrician. Women received
TRH or placebo at the time of their first corticosteroid treatment
only. The clinical trial and associated studies were approved by
institutional review boards at each institution.
The TRH trial enrolled 996 women and 1123 neonates at 13 institutions
between 1992 and 1996. Ten of these centers participated in collection
of infant blood samples between birth and 28 days of life. Analysis of
plasma cortisol levels was performed on infants of 24 to 32 weeks'
gestation. Because of a previously described association between number
of courses of antenatal corticosteroids and the degree and duration of
suppression of plasma cortisol,12 neonates who had
received 3 or more courses of antenatal steroids were excluded from
this analysis (n = 37, representing ~10% of the
infants with plasma samples). In addition, plasma samples obtained
after administration of postnatal corticosteroids were excluded from
analysis; 46 infants received postnatal steroids with treatment
beginning before 7 days of age in 9 of the infants (2.5% of the
total). It was a TRH study protocol violation to administer steroids
before 2 weeks of age.
Clinical Data
Detailed information was collected on the pregnancy history,
perinatal course, and infant outcome until discharge. Adverse outcome
was defined as death or CLD (requiring supplemental oxygen or
mechanical ventilation) at 36 weeks' PMA. As recently
reported,10 there was no effect of antenatal TRH treatment
on respiratory outcomes (surfactant treatment, respiratory distress
syndrome, CLD at 28 days and at 36 weeks' PMA), death or other
morbidities of prematurity (intraventricular hemorrhage, retinopathy of
prematurity, necrotizing enterocolitis, and patent ductus arteriosus).
Initial illness severity was assessed using the CRIB11
scoring system, developed to assess initial neonatal mortality risk. The score is based on factors including birth weight, gestational age,
maximum and minimum inspired oxygen concentration and maximum base
deficit in the first 12 hours of life, and the presence of congenital
malformations. CRIB scores range from 0 to 23, with higher values
associated with increased probability of death.
Cortisol Assay
Blood samples were obtained at birth (cord blood, either
umbilical venous or arterial) and at 2 and 24 hours and at 3, 7, 14, 21, and 28 days of age. The blood was obtained from an umbilical arterial catheter when available or from a peripheral vein at the time
of a routine clinical blood draw. All blood samples were collected into
tubes containing ethylenediaminetetraacetic acid and plasma was
obtained by centrifugation and stored at Immunoassay for cortisol was performed using an enhanced
chemiluminescense kit (Nichols Institute, San Juan Capistrano, CA). The
antibody has substantial cross reactivity (18%-30% of cortisol) only
for 11-deoxycortisol, corticosterone, and cortisone. Although 11-deoxycortisol is elevated in sick premature infants,13
cross-reactivity in the assay would contribute no more than ~2% to
the measured cortisol value at 3 to 7 days. Cortisone is present in
appreciable levels at birth; however, concentrations decreased rapidly
after delivery in infants of ~32 weeks' gestation.14
The maximal contribution of cortisone to measured cortisol values is
estimated to be ~6% during the first day of life. Assays were performed in duplicate on each plasma sample and the volume of plasma
was adjusted as needed to give data within the linear range of the
assay. No consistent differences were observed for results in cord
blood using venous or arterial blood and all data were used.
Statistical Analysis
Analyses of cortisol data were limited to infants of gestational
age 24 to 32 weeks and for whom at least 3 blood samples were collected
and assayed. In those infants who had >1 plasma sample obtained at 3 to 7 days or 14 to 28 days, the median of the values during that time
period was used. Results for placebo-treated and TRH-treated groups
were not different and final analyses on 1247 values used the entire
population of 314 infants.
Data for cortisol levels and CRIB scores followed a more normal
distribution when square root transformation, compared with a log
transformation, was applied to the values. The square root transformed
variables were used in all statistical tests to determine significance
(P = .05). To examine the relationship between the CRIB
score and the cortisol levels, the significance of the Pearson correlation coefficients were examined. To determine whether there was
a difference between specific categories (eg, gestational age groups)
for cortisol levels an analysis of variance or Student's t
test was used where appropriate. To examine the effect of the median
cortisol levels at 3 to 7 days on the adverse outcome of CLD or death
at 36 weeks' PMA, a logistic model was first developed for all of the
possible contributing clinical factors without the inclusion of the
cortisol values. Maternal factors included age, race, substance abuse,
medical history (diabetes, hypertension, heart disease), labor history
(duration, preterm or prolonged rupture of membranes, duration and type
of tocolytic, placenta previa, chorioamnionitis developing after
enrollment and placenta abruption), and delivery route. Infant factors
included gestational age, birth weight, multiple births, number of
courses of antenatal corticosteroids, CRIB score, patent ductus
arteriosus, pressor use, and respiratory distress syndrome (RDS). After
finding the best model of clinical predictors of adverse outcome, the
median of the cortisol level at 3 to 7 days was added to the model to determine whether the levels contributed to the model for CLD. A
likelihood-ratio test was used to examine whether the cortisol was
significant when included in the model.
Characteristics of Study Infants
Table 1 compares the characteristics
of study infants, 24 to 32 weeks' gestation, who did and did not have
plasma samples obtained. The median gestational age for both groups was 27.4 weeks. There was no significant difference in the percent randomized to receive TRH, the incidence of RDS, or the incidence of
CLD at 36 weeks' PMA. A slightly higher mortality rate was seen for
infants without cortisol levels, probably reflecting early postnatal
death and unavailability of plasma samples.
TABLE 1
METHODS
Top
Abstract
Methods
Results
Discussion
References
70°C.
RESULTS
Top
Abstract
Methods
Results
Discussion
References
Characteristics of Study Infants With and Without Plasma Cortisol
Levels
Cortisol Levels by Postnatal Age
The mean plasma cortisol levels at various time points, stratified by gestational age, are presented in Fig 1. For the entire study group (24-32 weeks), the mean cortisol at birth was 3.1 µg/dL, levels were maximal at 24 hours of life (19.5 µg/dL), and decreased to 5.9 µg/dL at 14 to 28 days of age. For statistical comparisons we examined data for infants with cortisol values at each of 2 time points. The increase in cortisol between birth and 1 day (n = 167) and the decrease between 1 day and 3 to 7 days (n = 244) were both significant (P < .001). The interval from the last dose of maternal corticosteroid treatment to birth was inversely correlated with plasma cortisol level at 0, 2, and 24 hours, as previously reported,15 but not at 3 to 7 days (data not shown). The pattern of change in cortisol levels with postnatal age was similar within each of the 3 gestational age categories: 24 to 25 weeks' gestation, 26 to 27 weeks' gestation, and 28 to 32 weeks' gestation, indicating that even the earliest gestation neonates increase their cortisol concentrations after birth. Statistical testing by logistic regression for each time point indicated a significant (inverse) association for gestational age and cortisol only at 14 to 28 days.
|
Effect of Initial Illness Severity on Plasma Cortisol
To determine the effect of initial illness severity on cortisol level, we assessed the association between the CRIB score at 12 hours of age and median cortisol level at 3 to 7 days of life (Fig 2). Because of skewed data, both the cortisol levels and CRIB scores were transformed to the square root to achieve a more normal distribution. Higher CRIB scores were associated with a slight increase in cortisol (r = 0.17; P = .01). However, the magnitude of the association is small. There is significant variability in cortisol levels at any given CRIB score, and some neonates with severe illness (CRIB >9; square root >3) did not have elevated cortisol levels. A similar weak association was seen between CRIB score and cortisol levels at 1 day of life (r = 0.16, P = .01; data not shown).
|
Association Between Plasma Cortisol and Adverse Outcome at 36 Weeks' PMA
A subset regression model of clinical predictors of death or CLD at 36 weeks' PMA was developed. Clinical risk factors evaluated for inclusion in this model included both maternal and infant factors (see "Methods"). The only significant clinical predictors of adverse outcome were gestational age and CRIB score. As expected, increasing gestational age was associated with a decreased risk of CLD (odds ratio [OR]: 0.8; P = .007). Increasing CRIB score, square root transformed, was associated with an increased risk of CLD (OR: 2.5; P < .001).
Median cortisol level at 3 to 7 days was then added to this best logistic regression model predicting CLD or death (Table 2). There was a borderline significant association between low cortisol levels and an increased risk of CLD or death (P = .08). Similar results were found examining cortisol at 3 to 7 days and occurrence of CLD in surviving infants at 36 weeks' PMA (OR: 0.65; confidence interval: 0.4-1.0; P = .052). This model was also examined with the median cortisol level at 14 to 28 days; the OR was 1.1 (P = .75). Similar analyses were not performed for cortisol at birth, 2 hours, 1 day, and incremental change between birth and 24 hours because of the suppressive effect of antenatal betamethasone treatment at these time points.
|
In Fig 3 the logistic regression model was used to calculate predicted probabilities of CLD or death at 36 weeks' PMA (adverse outcome). Each curve demonstrates the association between adverse outcome and gestational age at representative day 3 to 7 cortisol levels and CRIB scores. A CRIB score of 0 is represented by the lower set of curves, and a CRIB score of 5 is shown in the upper curves. Each of the 3 curves within a category represent cortisol levels of 5, 7, and 11 µg/dL, which are the median, and approximately 25th and 75th percentile levels, respectively, at 3 to 7 days for study infants with CRIB scores of 0. Based on this analysis, a 25 weeks' gestation infant with a CRIB score of 5 and a day 3 to 7 cortisol level of 7 would have a 60% chance of dying or developing CLD by 36 weeks' PMA. These curves demonstrate that the probability of adverse outcome is highest for lower gestational age infants with high CRIB scores. Taking these 2 factors into account, the contribution of cortisol level to the predicted probability is small, as demonstrated by the fact that the 3 cortisol curves are tightly grouped.
|
| |
DISCUSSION |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
In this study of plasma cortisol levels in 314 preterm infants, we found that even the earliest gestation infants (24-25 weeks) have an increase in cortisol after delivery, reaching maximal levels at 24 hours of life. Plasma cortisol then decreased gradually over 14 to 28 days. Levels were not associated with gestational age, and were only slightly correlated with illness severity, as indicated by CRIB score. After adjusting for clinical risk factors, low cortisol at 3 to 7 days of life, but not at 14 to 28 days, contributed minimally to increasing the risk of CLD.
The increase in cortisol concentration during the first day of life
suggests that many very premature infants have a functional hypothalamic-pituitary-adrenal axis and respond to the stress of
delivery and/or respiratory illness. Increasing cortisol after birth
also likely reflects, in part, reduced metabolism and clearance of
corticosteroid secondary to both immaturity of the newborn liver and
the absence of placental and maternal pathways. Although we found a
slight association between initial illness severity (as assessed by
CRIB score) and higher cortisol levels, it was not a strong
correlation. Similarly, in a study of 31 preterm infants >28 weeks'
gestation conducted by Hanna and colleagues,16 there was
no significant correlation between cortisol in the first week and
illness severity as reflected by the Score for Neonatal Acute
Physiology Perinatal Extension; 5 of 31 infants with a high score had
cortisol levels 
3.0 µg/dL. Lee et al3 reported similar
plasma cortisol concentrations in premature infants of 31 to 35 weeks'
gestation with severe versus mild respiratory disease. These findings
may reflect an inability of some preterm infants to mount an
appropriate cortisol stress response when exposed to serious illness in
addition to the stress of labor and delivery. However, earlier studies
with more mature premature infants reported higher cortisol and/or
total corticoid concentrations among preterm infants with severe RDS
compared with infants with milder disease.17,18
The relationship between adrenocortical function in the first week of life, as assessed by basal cortisol level, and development of CLD has been examined previously. Korte et al7 reported that infants <32 weeks' gestation with a cortisol concentration <414 nM (<15 µg/dL) were more likely to develop CLD at 36 weeks. These investigators argue that the cortisol levels in most sick premature infants, although similar to normal term values, are insufficient under the stressed conditions. By contrast, 2 other studies found no significant relationship between basal plasma cortisol and CLD.4,19
Response to ACTH stimulation may be more useful than isolated basal cortisol levels in identifying preterm neonates with evidence of adrenal insufficiency. Watterberg and colleagues4 studied preterm infants with ACTH stimulation testing at 5 to 7 days of life. All study infants increased their cortisol concentration after ACTH administration, and neonates with lower birth weight had a decreased response. A decreased cortisol response was also associated with an increased probability of subsequently developing CLD independent of gestational age. In the study by Korte et al,7 however, there was no association between responsiveness to ACTH and CLD at 28 days.
All women in our trial received antenatal corticosteroid therapy, which causes a transient suppression of the fetal adrenal axis. After a single course of betamethasone, cortisol concentration in cord plasma is maximally reduced by ~55% at 6 to 18 hours after treatment and returns to the control level by ~5 days.15 Because the degree and duration of fetal adrenal suppression increases with repetitive antenatal corticosteroid treatment,12 we limited our analysis to infants who received only 1 or 2 courses of antenatal therapy. The interval from last corticosteroid dose to delivery was similar for infants with and without CLD and we found no effect of treatment-delivery interval on levels of cortisol at 3 to 7 days. Thus, it is unlikely that our results regarding cortisol concentration at 3 to 7 days as a predictor of adverse outcome are influenced by antenatal corticosteroid treatment. In evaluating the status of the adrenal cortex in individual newborn infants, however, it is important to consider the number of courses and timing of antenatal corticosteroid treatment.
Based on our analysis of basal cortisol levels, we would speculate that routine early cortisol supplementation, achieving levels within the physiologic range, is unlikely to have a major impact on the incidence of CLD. Most trials involving early postnatal treatment of preterm infants with pharmacologic doses of dexamethasone have demonstrated modest decreases in the incidence of CLD and the relative risk for benefit is significant on meta analysis (OR: 0.82; P = .01; Banks, unpublished data). However, serious adverse effects of this treatment regimen have been described including gastrointestinal bleeding, intestinal perforation, hyperglycemia, hypertension, decreased growth, and poor long-term neurodevelopmental outcome.8,20-23 Two recent large trials of early postnatal dexamethasone (National Institute of Child Health and Human Development Neonatal Network and Vermont Oxford Network) were stopped because of significant adverse effects in the steroid treated groups.23,24 By contrast, a recent pilot trial of early low-dose hydrocortisone involving 40 preterm infants 500 to 999 g found an increase in survival without CLD5 without apparent side effects of treatment.
Defining whether there is a higher-risk subpopulation of preterm infants who would benefit from early low-dose cortisol, as well as the optimal dose and duration of treatment, will be important areas of future research. Restricting treatment to seriously ill preterm infants with low cortisol levels or an inappropriate response to ACTH may optimize the potential for a beneficial response while minimizing the likelihood of significant adverse effects.
| |
ACKNOWLEDGMENTS |
|---|
This study was supported by Grants 5 RO1 HD29201, 1 P50 HL56401, and M01-RR00240 and by Perinatal Associates.
In addition to the authors, the following members of the North American TRH Study Group participated in this study: C. Coburn, M. McCarthy, and E. Escobar at The Children's Hospital of Philadelphia, Philadelphia, PA; M. Morgan, E. Anday, K. Mooney, and M. Johnson at the University of Pennsylvania Hospital, Philadelphia PA; R.H. Phibbs, J.T. Parer, N. Newton, and J. Milar at the University of California San Francisco, San Francisco CA; M. Ross, D. Polk, J. Padbury, and S. Harrington at Harbor-UCLA Medical Center, Torrance, CA; D. J. Davis, K. Ash and J. Frank at Ottawa General Hospital, Ottawa, Ontario, Canada; E. Tyrala and L. Chan at Temple University Hospital, Philadelphia PA; J. Lioy and R. Librizzi at West Jersey Hospital, Voorhees, NJ; M. C. Hart, J. Garbaciak, and E. Ramthun at St Joseph's Hospital, Phoenix, AZ; F. L. Mannino, T. Moore,and E. Milan at University of California San Diego, San Diego, CA; J. Keith and M. Rivera-Alsina, Naval Medical Center, San Diego CA; and V. Bhutani, S. Weiner, and M. Grous at Pennsylvania Hospital, Philadelphia PA.
We thank all the physicians, nurses, respiratory therapists, and pharmacists at the 10 participating institutions who participated in the patient enrollment and collection of blood samples and clinical data used in this study. We would also like to thank Christine Coburn, RN, for excellence in directing the TRH trial and Sandy Mosiniak for preparation of this manuscript.
| |
FOOTNOTES |
|---|
The names of the individuals from the North American TRH Study Group can be found in the "Appendix."
Received for publication Feb 16, 2000; accepted Jul 13, 2000.
Reprint requests to (P.L.B.) Department of Pediatrics, Children's Hospital of Philadelphia, 416 Abramson Research Center, 34th St and Civic Center Blvd, Philadelphia, PA 19104-4318. E-mail: ballardp{at}email.chop.edu
| |
ABBREVIATIONS |
|---|
CLD, chronic lung disease; PMA, postmenstrual age; ACTH, adrenocorticotropic hormone; TRH, thyrotropin-releasing hormone; CRIB, Clinical Risk Index for Babies; RDS, respiratory distress syndrome; OR, odds ratio.
| |
REFERENCES |
|---|
|
Top
Abstract
Methods
Results
Discussion
References
|
|---|
This article has been cited by other articles:
 |
|
 |
|
  O. M. Peltoniemi, M. A. Kari, O. Tammela, L. Lehtonen, R. Marttila, E. Halmesmaki, P. Jouppila, M. Hallman, and for the Repeat Antenatal Betamethasone Study Group Randomized Trial of a Single Repeat Dose of Prenatal Betamethasone Treatment in Imminent Preterm Birth Pediatrics, February 1, 2007; 119(2): 290 - 298. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. D. Bruder, P. C. Lee, and H. Raff Dexamethasone treatment in the newborn rat: fatty acid profiling of lung, brain, and serum lipids J Appl Physiol, March 1, 2005; 98(3): 981 - 990. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
K. L. Watterberg, J. S. Gerdes, C. H. Cole, S. W. Aucott, E. H. Thilo, M. C. Mammel, R. J. Couser, J. S. Garland, H. J. Rozycki, C. L. Leach, et al. Prophylaxis of Early Adrenal Insufficiency to Prevent Bronchopulmonary Dysplasia: A Multicenter Trial Pediatrics, December 1, 2004; 114(6): 1649 - 1657. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 P C Ng, C H Lee, C W K Lam, K C Ma, I H S Chan, E Wong, and T F Fok Early pituitary-adrenal response and respiratory outcomes in preterm infants Arch. Dis. Child. Fetal Neonatal Ed., March 1, 2004; 89(2): F127 - F130. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 Statement on the Care of the Child with Chronic Lung Disease of Infancy and Childhood Am. J. Respir. Crit. Care Med., August 1, 2003; 168(3): 356 - 396. [Full Text] [PDF]
|
|
|
|
|
|
P. Ballabh, J. Kumari, A. N. Krauss, J. J. Shin, A. Jain, P. A. M. Auld, M. L. Lesser, and S. Cunningham-Rundles Soluble E-Selectin, Soluble L-Selectin and Soluble ICAM-1 in Bronchopulmonary Dysplasia, and Changes With Dexamethasone Pediatrics, March 1, 2003; 111(3): 461 - 468. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 E. Kajantie, L. Dunkel, U. Turpeinen, U.-H. Stenman, P. J. Wood, M. Nuutila, and S. Andersson Placental 11{beta}-Hydroxysteroid Dehydrogenase-2 and Fetal Cortisol/Cortisone Shuttle in Small Preterm Infants J. Clin. Endocrinol. Metab., January 1, 2003; 88(1): 493 - 500. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
 |
||
|
||
Home | My Pediatrics | Journal Information | Current Issue | Past Issues | Subscriptions & Services | Contact Us Click here for faster international access © 2001 American Academy of Pediatrics. All rights reserved. |
||
|
||
