Antenatal betamethasone administration in pregnancies at risk for preterm delivery before 34 weeks of gestation is an established and effective procedure to improve neonatal outcome by decreasing neonatal mortality and infant morbidity.1 However, in different animal models, excessive prenatal glucocorticoid administration reduces birth weight and increases blood pressure and glucose levels and alters the activity of the hypothalamic–pituitary–adrenal axis in adults.2 Although long-term studies in the human after antenatal corticosteroid treatment have not found adverse effects on general health and cardiovascular risk factors at 20 and 30 years, there was evidence for insulin resistance in these individuals3,4 and clinically apparent symptoms for cardiovascular disease may even appear at older ages. Moreover, hypothalamic–pituitary–adrenal axis regulation has not been assessed in these studies.
Therefore, concerns have been raised as to whether prenatal treatment with corticosteroids may entail long-term consequences for the health of these individuals. The hypothalamic–pituitary–adrenal axis, which is considered to be one of the major systems involved in the fetal origin of adult diseases hypothesis, has been shown to be vulnerable to excess glucocorticoid exposure during its maturational stage.5,6 Endogenous glucocorticoids are involved in mediation of developmental steps of endocrine and neurohumoral systems,7,8 and interference with this fine neuroendocrine balance may alter the function of these systems permanently.
Studies in premature neonates have shown that a single course of antenatal betamethasone causes suppression of stress reactivity because these neonates failed to increase cortisol levels in response to a stressor.9,10 Prematurity per se, prematurity-associated morbidity, and the stressful environment of a neonatal intensive care unit (NICU) may have a considerable effect on neonatal stress regulation.11 Thus, allocation of these effects to a specific event such as antenatal betamethasone treatment appears difficult, especially when considering that a pregnancy-associated complication usually precedes preterm delivery.
We therefore analyzed the hypothalamic–pituitary–adrenal axis reactivity of healthy, late preterm or term newborns who had received a single course of antenatal betamethasone treatment for imminent preterm delivery before 34 weeks of gestation and in whom preterm delivery was postponed for more than a mean of 8 weeks. We hypothesized that stress reactivity in these newborns is altered in comparison with those in the control group.
MATERIALS AND METHODS
The study was approved by the Research Ethics Committee of the University of Zurich and the federal ethics commission of the canton of Zurich, and written maternal consent was obtained from all participants in the study.
Only healthy newborns delivered after 34 weeks of gestation (more than 238 days) were included in the study. Forty-four neonates who had received antenatal betamethasone administration (2×12 mg within 24 hours) for imminent preterm delivery were delivered after 34 weeks of gestation during the study period (September 2006 to April 2007). In 21 cases, either parents refused to participate in the study (16 cases) or saliva collection was inadequate (five cases). Saliva samples of 23 neonates (study group) could be included in our analysis. The experience from a previous study allowed us to presume that, with regard to the variability of cortisol levels, approximately half of the number of controls provides a statistically adequate study group sample size. Accordingly, 40 healthy neonates, whose cases have been published previously,12 served as controls (control group). These neonates were born in the same institution between April and September 2005, and methods for saliva collection and laboratory analysis were identical. Neonatal weight between the 10th and 90th percentiles was required for both study groups. Newborns were excluded from analysis when intensive care treatment or invasive procedures had been necessary, when malformations were present, or when insufficient amounts of saliva were collected. Furthermore, maternal substance abuse (nicotine, alcohol) during pregnancy was an exclusion criterion. No signs of clinically apparent infections were present in either group. Two pregnancies in the study group were complicated by preterm premature rupture of membranes and four pregnancies by vaginal bleeding, leading to betamethasone treatment for imminent preterm delivery. The remaining cases were treated for preterm contraction and cervical ripening. Ultrasonographic measurements of fetal crown-rump length and biparietal diameter during the first trimester served as parameters for accurate determination of gestational age.
Stress reactivity was tested in newborns using a standardized procedure with an automated heel lance (heel-prick test) as a stress event. This test is a routinely applied screening test in which blood is sampled and analyzed for metabolic disorders (Guthrie test). Saliva samples were collected from each neonate 10 minutes before and 20 minutes after the stress event 72–96 hours postpartum between 8 am and 1 pm and analyzed for salivary cortisol and cortisone levels. Baseline levels were obtained in an undisturbed environment. Collection time was based on experiments revealing peak cortisol responses between 20 and 30 minutes postmanipulation.13 In detail, a cotton swab was placed in the neonate’s mouth for a collection time of 5 minutes. Salivary cortisol reflects the unbound, active fraction of cortisol and is highly correlated with plasma cortisol levels.14,15 Samples were placed in saliva-collection tubes (Salivette, Sarstedt, Nümbrecht, Germany), immediately frozen, and stored at -20°C until further analysis.
Saliva cortisol and cortisone were determined simultaneously by liquid chromatography tandem mass spectrometry, with atmospheric pressure chemical ionization in the positive ion mode, according to a modified method of Rauh et al16 using 100 microliters of the samples and calibrators, as previously described in detail.12
We applied STATA 9 statistics/data analysis software (Stata Corporation, College Station, TX) for statistical analysis according to Altman’s recommendations for repeated measurements that rise from the same individuals studied under different circumstances.17 Baseline characteristics were compared using the Mann-Whitney test and χ2 test when appropriate. Because basal levels of cortisol vary considerably in individual newborns and children,13 absolute values and median relative alterations are presented. Because cortisol and cortisone values were not distributed normally as analyzed by the Shapiro-Francia W’ test, we analyzed the difference between cortisol and cortisone baseline levels and time point “20 minutes post” using the Wilcoxon signed rank test. Mann-Whitney test was used for comparison of baseline cortisol and cortisone levels. To account for within-subject variations, a two-way analysis of variance for repeated measurements was conducted. A stepwise multiple regression was applied to test for putative influencing factors on cortisol alterations such as gestational age at betamethasone administration, gestational age at delivery, neonatal weight, and sex. The resulting regression coefficient indicated the predicted increase in the dependent variable for each unit increase in the explanatory variable. To validate our statistical analyses, a power calculation was conducted for the given sample size (n=23), assuming the mean increase in cortisol levels of neonates in the control group as relevant. Finally, cortisol and cortisone levels in the study group were tested for correlation using the Spearman rank correlation test. The level of statistical significance of all analyses was set at P<.05.
The mean interval between betamethasone administration and birth was 60 (+23) days. Gestational age, birth weight, weight percentile, and head circumference at birth did not differ significantly between the study group and control group. Betamethasone was administered at a mean gestational age of 29.4 (+2.6) weeks. A summary of the characteristics of the study population is given in Table 1.
The median baseline levels for cortisol and cortisone did not differ significantly between neonates in the study group and those in the control group (1.39 ng/mL [range 0.09–9.82] compared with 1.175 ng/mL [range 0.09–15.7] for cortisol, P=.42; 14.8 ng/mL [range 2.6–36] compared with 11.35 ng/mL [range 5.83–44.3] for cortisone, P=.34, respectively). Individual baseline and stimulated cortisol and cortisone levels are indicated in Figure 1.
In neonates in the control group, analysis of the physiological stress response revealed a significant increase in cortisol and cortisone levels in response to the heelstick, as was expected (2.4 ng/mL [range 0.3–12.2], 18.15 ng/mL [range 2.8–43.1], respectively, P<.05, Fig. 1). In contrast, neonates in the study group displayed a noticeably blunted stress response—cortisol and cortisone release was not significantly different from baseline levels (1.6 ng/mL [range 0.2–11.3], P=.76 and 17.1 ng/mL [range 7–32.9], P=.69, respectively, Fig. 1). A two-way analysis of variance again revealed that there was no significant alteration of cortisol after the stress stimulus in the study group (P=.65) as opposed to control group (P=.003). A one-sample t-test power calculation validated the absent cortisol response in neonates in the study group with a power of 99% at a significance level of 0.05. For better visualization of alterations before and after stress stimulus medians of relative alterations for cortisol and cortisone, see Figure 2.
To test whether the reduced cortisol response in the neonates in the study group could be influenced by gestational age at steroid administration, a stepwise multiple regression model was applied, revealing no significant association (P=.76). Subsequently, neonatal weight, sex, and gestational age at delivery were tested, and none of them had a significant influence on cortisol response levels in the study group (P=.71, P=.74, P=.71, respectively) (Table 2). Finally, we tested whether the decreased cortisol response in the study group was mediated by an increased conversion of cortisol to cortisone. Cortisol and cortisone levels were strongly correlated in the study group, thus excluding this possibility (P=.76).
Our results indicate that a single course of betamethasone administration in pregnancies at risk for preterm delivery entails alterations in hypothalamic–pituitary–adrenal axis responsiveness that persist into postnatal life, even if delivery can be avoided until or near term.
Previous studies in premature neonates (28–30 weeks of gestation) have suggested that antenatal betamethasone treatment persistently suppresses cortisol responsiveness to a stressor for at least 4–6 weeks after birth.10 Owing to the setting of that study, however, prematurity itself as well as the effect of stressful postnatal intensive care unit handling may have had considerable influence on these results. Unfortunately, owing to the lack of an appropriate control group of equal gestational age, these neonates were compared with a control cohort delivered at 33–34 weeks of gestation, significantly later than the study group. Thus, these cofactors could not be controlled for. In a small study population (n=9), however, the same group of investigators analyzed the hypothalamic–pituitary–adrenal response in a NICU situation in preterm neonates born at 33–34 weeks of gestation, again revealing a decrease in cortisol response after antenatal betamethasone administration.9
Baseline cortisol levels after antenatal corticosteroid treatment have been shown to be suppressed transiently for 2–7 days in preterm neonates,18,19 and cortisol response levels to a pharmacologic challenge were lower in preterm very low birth weight neonates after a single course of antenatal betamethasone administration 7 days postpartum. Curiously, this could not be shown for multiple doses of betamethasone administration. However, at 14 days postpartum, these differences no longer could be observed, thus indicating a transient effect.11 In fact, there is evidence, that premature newborns have an immature hypothalamic–pituitary–adrenal axis so that the hypothalamus may fail to recognize the stimulatory signal or adrenal steroidogenesis may be ineffective yet.20–22
The present study is the first to analyze stress physiology in healthy term or near-term neonates after a single course of betamethasone administration at a much earlier time in pregnancy. Consequently, these neonates did not experience the potentially stressful environment of a NICU, and prematurity-associated effects with related morbidity can be excluded. Likewise, putative influencing factors such as smoking during pregnancy or intrauterine compromise of the fetus had been excluded explicitly, in contrast to studies in preterm neonates. Furthermore, the mean interval between betamethasone administration and hypothalamic–pituitary–adrenal axis analysis was more than 8 weeks, thereby excluding transient short-term effects.
Animal models in different species collectively show that corticosteroid application in pregnancy permanently affects hypothalamic–pituitary–adrenal axis function and behavior in offspring.5 Unlike endogenous glucocorticoids, synthetic glucocorticoids predominantly bind to glucocorticoid receptors because the mineralocorticoid receptors have low affinity to exogenous glucocorticoids.23 Thus, the glucocorticoid receptor is likely to be affected predominantly. In the rodent, a single course of antenatal betamethasone induced a decrease in hippocampal glucocorticoid receptor expression.24 Interestingly, disruption of the glucocorticoid receptor gene in the central nervous system in knock-out mice results in an impaired behavioral response to stress.25 Constant receptor alterations have been described at the level of the limbic system in the hippocampus (involved in negative feedback), the amygdala (involved in activation), and the paraventricular nucleus.26–30
In guinea pigs, a single equivalent dose of glucocorticoids used in pregnant women caused significantly lower hippocampal glucocorticoid receptor mRNA levels in females, accompanied by an attenuated stress response. In contrast, glucocorticoid receptor mRNA levels were elevated in males that showed elevated basal but not stress-induced cortisol levels.31 Repeated doses of glucocorticoids induced sex-specific differential effects on mineralocorticoid receptor mRNA and glucocorticoid receptor mRNA levels in different regions.32 Animal models have shown further that females might be more sensitive to hypothalamic–pituitary–adrenal axis programming than males.33,34 We did not observe a sex-dependent difference in hypothalamic–pituitary–adrenal axis reactivity in our study, just as a study in adults analyzing postmenopausal women.35 Sex-specific cortisol reactivity in general depends on different levels of corticosteroid-binding globulin and sex steroids.36 We therefore cannot exclude sex-specific hormonal influences that may manifest at a later time during adolescence and adulthood in these neonates.
The mechanisms underlying hypothalamic–pituitary–adrenal axis balance thus are complex and depend on timing and dose of glucocorticoid exposure.29 The variability in neurodevelomental profiles of different species further adds to the difficulty in drawing conclusions for the human situation.
Limited data on prenatal glucocorticoid treatment in nonhuman primates indicate structural and morphological alterations in the area of the hippocampus and an altered set point for basal and stress-induced cortisol levels at the age of 9 months.37 Furthermore, alterations in cytoskeletal proteins and presynaptic terminals involved in neuroplasticity were observed after equivalent doses of betamethasone in the baboon.38
In the human, betamethasone-induced alterations in hippocampal mineralocorticoid receptor mRNA expression levels could not be observed in preterm neonates who died 4 days after delivery (n=9); however, glucocorticoid receptor mRNA alterations that may be of major relevance could not be analyzed reliably, which makes it difficult to draw conclusions.39 Different gestational ages and variabilities in glucocorticoid exposure as well as other factors, such as stress due to intensive care settings, further add to the difficulty of interpreting these data.
The strength of the present study is that study groups are largely free from maternal- and neonatal-derived interfering factors such as prematurity-associated interventions and morbidities. Furthermore, the long interval between betamethasone administration and delivery (mean 60 days) does exclude transient short-term effects of betamethasone on the hypothalamic–pituitary–adrenal axis. Nevertheless, betamethasone administration implies the presence of a risk for preterm delivery, and we cannot exclude the possibility that these risk factors, such as preterm contractions and their treatment, bleeding, undetectable infections, and event- or hospitalization-associated stress situations for the mother, add to the alteration of hypothalamic–pituitary–adrenal axis responsiveness. Indeed, median maternal hospital stay during pregnancy was 8 days. Furthermore, whether our observed effects transform into long-term consequences has to be established by follow-up studies.
We would like to point out that antenatal glucocorticoid administration in pregnancies at risk for preterm delivery has established benefits in reducing neonatal mortality and morbidity,1 and it is not clear to what extend our observed alterations associated with this procedure really are relevant determinants for diseases in later life. However, if a permanent alteration of the hypothalamic–pituitary–adrenal axis occurs in the human fetus after the application of a single course of betamethasone treatment, hypothalamic–pituitary–adrenal axis function would be expected to be modified already in the neonate. This precondition for a persistent reprogramming of the hypothalamic–pituitary–adrenal axis has been demonstrated in our study. Long-term follow-up studies analyzing hypothalamic–pituitary–adrenal axis reactivity are not yet available. In a 30-year follow-up study of Dalziel et al,3 basal plasma cortisol levels were unaltered after a single course of maternal betamethasone administration; stress reactivity, however, was not analyzed, and clear indicators for the presence of insulin resistance were present in these individuals.
1. Roberts D, Dalziel S. Antenatal corticosteroids for accelerating fetal lung maturation for women at risk of preterm birth. The Cochrane Database of Systemic Reviews 2006, Issue 3. Art. No.: CD004454. DOI: 10.1002/14651858.CD004454.pub2.
2. Seckl JR, Holmes MC. Mechanisms of disease: glucocorticoids, their placental metabolism and fetal ‘programming’ of adult pathophysiology. Nat Clin Pract Endocrinol Metab 2007;3:479–88.
3. Dalziel SR, Walker NK, Parag V, Mantell C, Rea HH, Rodgers A, et al. Cardiovascular risk factors after antenatal exposure to betamethasone: 30-year follow-up of a randomised controlled trial. Lancet 2005;365:1856–62.
4. Dessens AB, Haas HS, Koppe JG. Twenty-year follow-up of antenatal corticosteroid treatment. Pediatrics 2000;105:E77.
5. Kapoor A, Petropoulos S, Matthews SG. Fetal programming of hypothalamic-pituitary-adrenal (HPA) axis function and behavior by synthetic glucocorticoids. Brain Res Rev 2008;57:586–95.
6. Owen D, Andrews MH, Matthews SG. Maternal adversity, glucocorticoids and programming of neuroendocrine function and behaviour. Neurosci Biobehav Rev 2005;29:209–26.
7. Young JB. Programming of sympathoadrenal function. Trends Endocrinol Metab 2002;13:381–5.
8. Matthews SG. Early programming of the hypothalamo-pituitary-adrenal axis. Trends Endocrinol Metab 2002;13:373–80.
9. Davis EP, Townsend EL, Gunnar MR, Georgieff MK, Guiang SF, Ciffuentes RF, et al. Effects of prenatal betamethasone exposure on regulation of stress physiology in healthy premature infants. Psychoneuroendocrinology 2004;29:1028–36.
10. Davis EP, Townsend EL, Gunnar MR, Guiang SF, Lussky RC, Cifuentes RF, et al. Antenatal betamethasone treatment has a persisting influence on infant HPA axis regulation. Journal of Perinatology 2006;26:147–53.
11. Ng PC, Lam CW, Lee CH, Ma KC, Fok TF, Chan IH, et al. Reference ranges and factors affecting the human corticotropin-releasing hormone test in preterm, very low birth weight infants. J Clin Endocrinol Metab 2002;87:4621–8.
12. Schäffer L, Müller-Vizentini D, Burkhardt T, Rauh M, Ehlert U, Beinder E. Blunted stress response in small for gestational age neonates. Pediatr Res 2008 October 22 [Epub ahead of print].
13. Gunnar MR. Reactivity of the hypothalamic-pituitary-adrenocortical system to stressors in normal infants and children. Pediatrics 1992;90:491–7.
14. Calixto C, Martinez FE, Jorge SM, Moreira AC, Martinelli CE Jr. Correlation between plasma and salivary cortisol levels in preterm infants. J Pediatr 2002;140:116–8.
15. Gunnar MR. Studies of the human infant’s adrenocortical response to potentially stressful events. New Dir Child Dev 1989:3–18.
16. Rauh M, Groschl M, Rascher W, Dorr HG. Automated, fast and sensitive quantification of 17 alpha-hydroxy-progesterone, androstenedione and testosterone by tandem mass spectrometry with on-line extraction. Steroids 2006;71:450–8.
17. Altman D. Practical statistics for medical research, first ed. London (UK): Chapman & Hall; 1991.
18. Parker CR Jr, Atkinson MW, Owen J, Andrews WW. Dynamics of the fetal adrenal, cholesterol, and apolipoprotein B responses to antenatal betamethasone therapy. Am J Obstet Gynecol 1996;174:562–5.
19. Ballard PL, Gluckman PD, Liggins GC, Kaplan SL, Grumbach MM. Steroid and growth hormone levels in premature infants after prenatal betamethasone therapy to prevent respiratory distress syndrome. Pediatr Res 1980;14:122–7.
20. Hanna CE, Keith LD, Colasurdo MA, Buffkin DC, Laird MR, Mandel SH, et al. Hypothalamic pituitary adrenal function in the extremely low birth weight infant. J Clin Endocrinol Metab 1993;76:384–7.
21. Hingre RV, Gross SJ, Hingre KS, Mayes DM, Richman RA. Adrenal steroidogenesis in very low birth weight preterm infants. J Clin Endocrinol Metab 1994;78:266–70.
22. Lee MM, Rajagopalan L, Berg GJ, Moshang T Jr. Serum adrenal steroid concentrations in premature infants. J Clin Endocrinol Metab 1989;69:1133–6.
23. Krozowski ZS, Funder JW. Renal mineralocorticoid receptors and hippocampal corticosterone-binding species have identical intrinsic steroid specificity. Proc Natl Acad Sci U S A 1983;80:6056–60.
24. Velisek L. Prenatal corticosteroid impact on hippocampus: implications for postnatal outcomes. Epilepsy Behav 2005;7:57–67.
25. Tronche F, Kellendonk C, Kretz O, Gass P, Anlag K, Orban PC, et al. Disruption of the glucocorticoid receptor gene in the nervous system results in reduced anxiety. Nat Genet 1999;23:99–103.
26. Matthews SG. Antenatal glucocorticoids and programming of the developing CNS. Pediatr Res 2000;47:291–300.
27. Jacobson L. Hypothalamic-pituitary-adrenocortical axis regulation. Endocrinol Metab Clin North Am 2005;34:271–92, vii.
28. Seckl JR. Prenatal glucocorticoids and long-term programming. Eur J Endocrinol 2004;151(Suppl 3):U49–62.
29. Welberg LA, Seckl JR. Prenatal stress, glucocorticoids and the programming of the brain. J Neuroendocrinol 2001;13:113–28.
30. Herman JP, Prewitt CM, Cullinan WE. Neuronal circuit regulation of the hypothalamo-pituitary-adrenocortical stress axis. Crit Rev Neurobiol 1996;10:371–94.
31. Dean F, Yu C, Lingas RI, Matthews SG. Prenatal glucocorticoid modifies hypothalamo-pituitary-adrenal regulation in prepubertal guinea pigs. Neuroendocrinology 2001;73:194–202.
32. Liu L, Li A, Matthews SG. Maternal glucocorticoid treatment programs HPA regulation in adult offspring: sex-specific effects. Am J Physiol Endocrinol Metab 2001;280:E729–39.
33. Weinstock M, Matlina E, Maor GI, Rosen H, McEwen BS. Prenatal stress selectively alters the reactivity of the hypothalamic-pituitary adrenal system in the female rat. Brain Res 1992;595:195–200.
34. McCormick CM, Smythe JW, Sharma S, Meaney MJ. Sex-specific effects of prenatal stress on hypothalamic-pituitary-adrenal responses to stress and brain glucocorticoid receptor density in adult rats. Brain Res Dev Brain Res 1995;84:55–61.
35. Reynolds RM, Walker BR, Syddall HE, Andrew R, Wood PJ, Phillips DI. Is there a gender difference in the associations of birthweight and adult hypothalamic-pituitary-adrenal axis activity? Eur J Endocrinol 2005;152:249–53.
36. Kudielka BM, Buske-Kirschbaum A, Hellhammer DH, Kirschbaum C. HPA axis responses to laboratory psychosocial stress in healthy elderly adults, younger adults, and children: impact of age and gender. Psychoneuroendocrinology 2004;29:83–98.
37. Uno H, Eisele S, Sakai A, Shelton S, Baker E, DeJesus O, et al. Neurotoxicity of glucocorticoids in the primate brain. Horm Behav 1994;28:336–48.
38. Antonow-Schlorke I, Schwab M, Li C, Nathanielsz PW. Glucocorticoid exposure at the dose used clinically alters cytoskeletal proteins and presynaptic terminals in the fetal baboon brain. J Physiol 2003;547:117–23.
39. Noorlander CW, De Graan PN, Middeldorp J, Van Beers JJ, Visser GH. Ontogeny of hippocampal corticosteroid receptors: effects of antenatal glucocorticoids in human and mouse. J Comp Neurol 2006;499:924–32.