Main

Glucocorticoids play an important role in perinatal and neonatal medicine. Clinical benefits associated with glucocorticoids include, but are not limited to, antenatal administration for lung, brain and gastrointestinal protection of preterm infants, and postnatal use in ventilator dependence and pressor-resistant hypotension. The precise mechanisms underlying the beneficial effects of perinatal glucocorticoids have not yet been clarified. The genomic actions of glucocorticoids occur via binding to the glucocorticoid receptor, a member of the nuclear receptor family of ligand-dependent transcription factors (1). After activation by its ligand, the receptor can act as a transcription factor and alter the expression of specific target genes. Whether specific actions of glucocorticoids occur via this mechanism or via other nongenomic effects is unclear given the variety of uses and lack of mechanistic studies in specific disease models. What is clear, however, is that from adult neurologic and cardiovascular disorders to chronic glucocorticoid-induced hypertension, steroids have a significant impact on the human vascular system (1,2,3).

Mechanisms of Action

The effects of glucocorticoids on vascular health and dysfunction may be mediated in part by activation of endothelial nitric oxide synthase (eNOS). Dysfunction of eNOS or its downstream signaling targets has been implicated in many disease states. Endothelial NOS catalyzes the formation of NO and citrulline from the nitrogens of L-arginine via a complex oxidation-reduction reaction, requiring molecular oxygen and NADPH plus numerous cofactors (flavin adenine dinucleotide (FAD), Flavin mononucleotide (FMN), heme, and tetrahydrobiopterin), as well as the activator calmodulin (4). Nitric oxide diffuses extracellularly to bind with soluble guanylyl cyclase in neighboring smooth muscle cells to increase cyclic GMP concentrations. Cyclic GMP activates G kinase and decreases smooth muscle cell intracellular calcium levels, eliciting relaxation in most conductance vessels. Activation of eNOS and release of NO have been shown to play important roles in normal neonatal circulatory transition; however the role of NO in various neonatal disease states, such as bronchopulmonary dysplasia (BPD), chronic pulmonary hypertension associated with BPD, neonatal sepsis, and necrotizing enterocolitis, are not much well understood (5). How the vascular processes involved in these physiologic and pathologic states may be influenced by steroid administration is incompletely understood, but is important that given the protection conferred by antenatal steroids and the clinical effects of postnatal steroids.

Vascular responses are known to be influenced by eNOS expression and activity. Antenatal dexamethasone (DEX) treatment increases eNOS expression in large vessel endothelium and large airway and small airway epithelium of fetal rat lungs (6). In rat models of congenital diaphragmatic hernia in which eNOS protein levels in the lung are decreased, maternal antenatal DEX administration results in offspring lung tissue that contained eNOS protein amounts equal to that of control, noncongenital diaphragmatic hernia animals (7). Antenatal betamethasone (BMZ) stimulates an increase in eNOS protein levels in lambs born with pulmonary hypertension induced by fetal ductal ligation (8). Similarly, repeat doses of antenatal BMZ increased total amounts of eNOS protein in the lung of newborn lambs (9). Of note, steroids also decrease baseline values of IL6 and reactive oxidative species suggesting other mechanisms may contribute to improved pulmonary outcomes following antenatal BMZ (10). Thus, through its effect on pulmonary vascular physiology, biochemistry and molecular signaling, antenatal steroids improve the pulmonary transition and adaptation at birth in a number of animal models.

Given the pivotal role of eNOS in vascular homeostasis, it is also important to consider the impact of steroids on eNOS cofactors that have been implicated in disease processes of vascular dysfunction. Tetrahydrobiopterin (BH4) is one such cofactor or stabilizer of eNOS , which is altered by corticosteroid administration. In adult rats, glucocorticoids downregulate guanosine triphosphate cyclohydrolase, the generator of tetrahydrobiopterin, causing attenuation of eNOS-dependent regulation of vascular contractility (11). Glucocorticoids are known to interact with Hsp90, a molecular chaperone required for eNOS phosphorylation and coupled eNOS activity (12). In a lamb model of pulmonary hypertension, antenatal BMZ restored Hsp90 interactions with eNOS in hypertensive pulmonary artery endothelial cells and corrected the increase in superoxide levels and decrease in NO bioavailability, resulting in decreased oxidative stress in pulmonary artery endothelial cells and improved eNOS protein expression (8). Steroids may affect other eNOS cofactors implicated in dysregulation of vascular tone such as caveolin-1 and calmodulin, however their contribution requires further study (13).

Synthetic glucocorticoids, specifically DEX and BMZ, cross the placenta without degradation by 11 beta-hydroxysteroid dehydrogenase and reach significant pharmacological levels in the fetus, in the range of 30–50% of the maternal levels (14). The mechanisms underlying the clinical benefits of antenatal corticosteroids may, at least in part, be related to their effects on the fetal vasculature. The vascular effects of steroids are described below according to their pre and postnatal uses, with separate consideration of the pulmonary and systemic vascular effects of various steroid preparations and dosing regimens.

Antenatal Vascular Effects

Antenatal steroids have been shown to have many beneficial effects on the fetus. In multiple randomized controlled trials and meta-analyses, maternal administration of a single course of antenatal corticosteroids, commonly BMZ or DEX, is associated with an overall reduction in neonatal death and morbidity, including respiratory distress syndrome, intraventricular hemorrhage, NEC, and neonatal intensive care unit admissions (15,16). Routine clinical use of antenatal glucocorticoids has previously been limited to mothers with threatened or impending preterm birth between 24 and 34 wk of gestation (17). However, infants born between 18 and 23 wk gestation may also benefit from antenatal corticosteroids with decreased mortality and improved neurodevelopmental outcomes (18). More recently, it has been demonstrated that antenatal BMZ treatment decreases respiratory morbidity and NICU admission rates among late preterm infants born between 34 and 36 wk of gestation (19). Glucocorticoids may have maximal effect during a critical window of development, an important area for future research.

Infants with congenital pulmonary abnormalities may also benefit from antenatal BMZ (20). Studies on the potential advantages of antenatal glucocorticoid in other clinical scenarios including fetuses with antenatally diagnosed congenital diaphragmatic hernia (20) and fetuses of diabetic mothers at or near term are needed.

Pulmonary Effects of Antenatal Steroids

Prenatal glucocorticoids have significant maturational effects on the developing fetal lung, which include enhanced alveolar differentiation, thinning of alveolar septae and capillary walls, and upregulation of surfactant production (21,22). In fetal rat and lamb models of pulmonary development, antenatal glucocorticoids enhance normal development with increased antioxidant activity and decreased formation of reactive oxygen species (6,7). In a premature lamb model of lipopolysaccharide induced chorioamnionitis, antenatal steroids provide partial recovery of lung structure via the sonic hedgehog pathway (23).

Antenatal corticosteroids enhance fetal pulmonary adaptation at birth, although the mechanism is unclear. In a late gestation lamb model, fetal pulmonary vascular reactivity is increased by antenatal corticosteroid administration in response to vasodilatory stimuli, such as catecholamines, prostaglandins, and NO (24). In a preterm lamb model, antenatal corticosteroids specifically enhance fetal pulmonary blood flow, an effect that does not persist after birth (25). In near-term lambs with hypoplastic lungs, antenatal steroids return pulmonary blood flow to control levels, restoring the impaired perinatal adaptation of the pulmonary circulation in this model (26). Analysis of Doppler indices in extremely premature human fetuses following maternal BMZ administration suggests decreased pulmonary vascular resistance (27). Thus, the effect of antenatal corticosteroids on pulmonary blood flow may vary based on gestation and perinatal cardiopulmonary physiology.

Systemic Effects of Antenatal Steroids

Within days of maternal steroid administration, changes are observed in human fetal hemodynamics (28,29). Corticosteroids dilate the human umbilical circulation both in vivo and in vitro and may blunt fetal response to hypoxia (30,31). Corticosteroids increase fetal systemic blood pressure while cerebral blood flow is decreased due to increased cerebral vascular resistance (25). In clinical practice, fewer extremely premature infants who receive antenatal steroids require blood pressure support after birth (32).

The effects of antenatal corticosteroids on central nervous system development are complex. Among appropriate for gestational age premature infants, antenatal corticosteroid use is associated with advanced cortical maturation as measured by diffusion tensor imaging at term equivalent age (33). Normally grown, preterm lamb fetuses exhibit transient decreases in carotid flow after maternal antenatal BMZ but Intrauterine growth retardation (IUGR) fetuses and those who have experienced a hypoxic insult have exaggerated cerebral reperfusion which correlates with immunohistochemical evidence of oxidative stress and cerebral apoptosis in the fetal brain (34,35).

While these effects may arise from direct damage to neurons, altered cerebrovascular regulation may also be a contributing factor (36,37). In humans, both BMZ and DEX have been implicated in disturbances in fetal middle cerebral artery (MCA) flow after single course maternal antenatal steroid administration (38,39). Transient reductions in MCA flow after maternal glucocorticoid administration have been reported in normal human fetuses; prolonged reductions in MCA blood flow have been reported in fetuses <32 wk of gestation and in the IUGR fetus (38,40,41,42). A majority of IUGR fetuses (from one-half to two-thirds) transiently respond within 24–96 h to antenatal BMZ as evidenced by reversal of absent end-diastolic flow in the uterine artery (43). However, a subset of severely affected infants show no change in umbilical artery or MCA pulsatility in response to BMZ, and may be at heightened perinatal risk with longer duration of assisted ventilation and supplemental oxygen (44,45).

Long-term assessments of neurodevelopmental outcomes following antenatal exposure to glucocorticoids in humans are lacking in this population of fetuses with severe IUGR. Although meta-analyses report less overall risk of IVH with antenatal steroids; concern has been raised whether this high-risk population should receive antenatal steroids; given the lack of difference in morbidity and mortality between IUGR fetuses that receive antenatal steroids vs. those that do not; and the differential responses of umbilical and cerebral blood flow parameters between IUGR fetuses and those with appropriate intrauterine growth (46,47). In preterm fetuses without growth restriction, there are also reports of an increased risk of neurodevelopmental disability associated with antenatal DEX compared with antenatal BMZ (48).

Antenatal steroid use has been associated with multiorgan vascular effects, which may not manifest until adulthood. For example, antenatal steroid administration improves urinary output in the newborn (49,50). In the kidney, renal blood flow in newborn preterm lambs is improved after single course antenatal BMZ but long-term assessment of these animals reveals an association between antenatal steroid exposure and hypertension and altered renal development in adult life that may be gender-specific (51,52,53). The adult cerebral circulation of prenatally-exposed sheep to clinically relevant doses of antenatal BMZ exhibits attenuation of pressure-induced vasoconstriction and altered vessel reactivity (54). Fetal programming upon exposure to high dose antenatal (or postnatal) corticosteroids undoubtedly influences the long-term impact of corticosteroids on the developing vascular system and is an important area for further animal and human research.

Human and animal studies using multiple courses of antenatal steroids show that repeated exposure can adversely affect multiple organ systems. Repeated antenatal steroid doses may reduce the severity and frequency of neonatal lung disease in humans but has been associated with decreased fetal growth in some studies (55,56). In lambs, repeat maternal doses of antenatal BMZ cause growth retardation, whereas in utero fetal administration of repeat doses does not alter growth but also demonstrates less improvement in postnatal lung function, suggesting a specific role for maternal metabolism of glucocorticoids in mediating its beneficial effects on the fetus (57). Repeated doses of antenatal BMZ are associated with transient increases in cardiac wall mass at birth though single courses are not (55,58). In the follow-ups between 6 and 8 y of age, there was no difference in cardiovascular risk factors between those who received a single antenatal course of BMZ compared with multiple courses of BMZ (59). Repeated doses of antenatal steroids do not appear to change significantly survival free of major neurologic disability or BMI at 24-month follow-up, although a nonsignificant trend toward more cerebral palsy continues to suggest the need for caution and further research (60,61).

A summary of the vascular effects of antenatal glucocorticoid administration in human and animal research can be found in Table 1 .

Table 1 Summary of the antenatal effects of glucocorticoids

Postnatal Vascular Effects

In vitro experiments in vascular cells and tissues display a dose-dependent response to steroids across multiple experimental endpoints. DEX increases human endothelial cell proliferation and migration and is associated with increased eNOS phosphorylation and cyclic GMP levels after 2–4 h of exposure (62). Similar results following acute glucocorticoid exposure have been reported in endothelial cells isolated from multiple species and vascular beds (1,3,62). The common denominator in this effect is the high- dose and short duration of exposure (<36 h) to corticosteroid. Whether the effect is associated with altered eNOS protein amount, mRNA, or nitric oxide production, this effect is not sustained with long-term corticosteroid treatment, even when a 10-fold increase in steroid dose is used chronically (63). A glucocorticoid responsive element (GRE) has been reported for eNOS in human umbilical vein endothelial cells and may help to explain the corticosteroid effects on vascular cells (64). Local tissue metabolism of steroids must also be considered when interpreting vascular responses to glucocorticoids. Steroid dehydrogenases, which regenerate and deactivate bioactive forms of endogenous steroids, have been reported to play significant and distinct roles in modulating the effect of glucocorticoids on eNOS (64).

Pulmonary Effects of Postnatal Steroids

To the neonatologist, the lung is the first litmus test of successful adaptation to extra uterine life. Robust data suggest that steroids have beneficial effects on this organ. The clinical impact of postnatal corticosteroids on lung function is well-recognized. Postnatal steroids (primarily DEX) improve the likelihood of successful extubation and decrease the incidence and severity of chronic lung disease. Some of the early meta-analyses on follow-up data of infants receiving postnatal steroids for chronic lung disease treatment or prevention presented concerning evidence of increased cerebral palsy and poorer neurodevelopmental performance in infants exposed to postnatal steroids. However, more recent reports of low dose postnatal steroids that were started after the first week of life fail to demonstrate adverse effects or benefits at longer follow-up durations (65,66). The potential benefits of postnatal steroids have generally been reserved for the infant at highest risk of pulmonary mortality.

Postnatal steroids seem to offer some protective advantages in animal models of lung inflammation or disease. Adult rabbits chronically exposed to DEX (3–4 d) are protected from the detrimental effects of chronic hypoxia in a model of pulmonary hypertension. DEX treatment prevents the hypoxia-induced decreases in phosphorylation of both eNOS and Akt, a protein kinase that phosphorylates serine 1,177 residues contributing to eNOS activation. In organ-cultured pulmonary arteries of adult rabbits, DEX recovered eNOS mRNA expression and maintained eNOS distribution similar to normoxic controls in hypoxia-induced pulmonary hypertension (13).

Systemic Effects of Postnatal Steroids

The acute actions of corticosteroids on the systemic circulation appear to have biologic and clinical relevance. Acute effects of eNOS activation are evident in adult disease states and are modulated by high-dose and brief exposure to corticosteroids, effects that are not transcriptionally mediated. Specifically, in adult mouse stroke models, eNOS is involved in the neuroprotective effect of steroids. Endothelial NOS mRNA and protein levels do not acutely change, yet NO generation from eNOS is increased and stroke size is reduced. These rapid, nontranscriptional effects are dose-dependent, exclusive to glucocorticoid receptor activation and are not estrogen or mineralocorticoid receptor-responsive (1). Similarly, in models of myocardial infarction, glucocorticoid-mediated reduction in infarction size is not seen in eNOS null mice and is evident only under conditions of high-dose and short-exposure times (3).

In the clinical setting, steroids are used to augment blood pressure, particularly in vasopressor-resistant hypotension (67). Animal models of Cushing syndrome have been used to study the interaction of steroids and blood pressure. Results suggest that these effects are exposure time-dependent. Animal studies show that eNOS is down-regulated at a transcriptional and functional level under chronic glucocorticoid administration (63,68). In these mice models, the decrease in eNOS mRNA and protein in the liver and kidney (with no significant decrease in the heart), correlates with lower tissue amounts of nitrite generation. These studies suggest that proper eNOS function is an important contributor to blood pressure maintenance. Acute, high-dose, short-duration corticosteroids have beneficial effects in ischemic disease states while chronic administration (unrelated to dosage) down regulates eNOS.

Corticosteroids also reduce vascular resistance and increase blood flow in the kidney, heart, and eye (69,70). These effects are thought to be mediated via endothelial-dependent pathways involving vasodilator mediators, including prostaglandin E2 and NO. In adult rats, seven-day treatment with DEX (to prevent cyclosporine-induced nephropathy) produces a marked vasodilatory effect and is associated with increased eNOS (71). Short-term postnatal DEX causes decreased glomerular number and function in adult rats by limiting cell longevity through increased apoptosis (72). In infants treated with postnatal corticosteroids increased retinal blood flow has been noted (73). A subset of neonatal rats exposed to DEX has increased rates of systolic dysfunction as adults (74). However, in humans, a six-week tapering course of postnatal DEX was not associated with altered systolic blood pressure or BMI at school-age follow-up (75).

While the short-term benefits of postnatal steroids on systemic blood pressure in hypotensive neonates were immediately appreciated from clinical experience, the longer-term, neurologic detrimental effects, particularly of prolonged courses of steroids initiated in the first week of life, took longer to be recognized (12,76,77). While there are short-term benefits in several organ systems, the long-term effects of antenatal and postnatal glucocorticoids have yet to be fully elucidated. Table 2 provides a summary of the vascular effects of postnatal glucocorticoids with a breakdown by species, postnatal age, steroid type and organ system.

Table 2 Summary table of the vascular effects of postnatal glucocorticoids

Conclusion

Glucocorticoids are widely used in perinatal medicine. There is evidence that some of the vascular effects occur via nontranscriptional activation of eNOS. However, the vascular mechanisms of action appear to be dose-, duration- and vascular bed-dependent confounding interpretation of clinical risks and benefits. Data strongly support that antenatal glucocorticoids enhance pulmonary adaptation at birth. There is evidence antenatal steroids cause alterations in fetal systemic vascular tone as well, although the clinical impact of these changes is marginally understood. Postnatal glucocorticoids have beneficial, protective effects in animal models of pulmonary hypertension. Various adult tissue beds also benefit from postnatal steroids in high-dose and short duration, by a mechanism that appears to be nontranscriptionally mediated.

Several gaps remain in our understanding of glucocorticoid effects on the vasculature. The majority of the literature investigates the prenatal effects of glucocorticoids on the pulmonary vascular bed while the brain and gastrointestinal beds of the antenatally exposed fetus and newborn remain largely unexplored. This may be relevant given the protection that is conferred postnatally on these organ systems and the importance of understanding whether it is NOS regulation or other molecular machinery that is responsible for the beneficial effects. Large animal models, such as the neonatal baboon, piglet and lamb possess vascular beds amenable to experimentation that may shed light on the physiological and functional effects of glucocorticoids in the perinatal period while rodent and cell culture studies offer models to explore transcriptional activity, protein activation and downstream signal transduction. Whether findings in these experimental models are translatable to the human neonate is still in question. Inconsistent outcome data have been reported on the impact of antenatal steroid exposure on blood pressure, and vascular stiffness in antenatally exposed adolescents and adults (78,79,80,81). Studies on the long-term effects of perinatal glucocorticoid exposure on the cardiovascular system in adulthood are needed both in experimental animal models and in children, while adolescents and adults exposed to glucocorticoids in the perinatal period.

Statement of Financial Support

The submitting authors have no conflicts of interests to disclose in relation to the work submitted. No financial assistance was received to support this study.