By continuing to browse this site you agree to us using cookies as described in About Cookies
Notice: Wiley Online Library will be unavailable on Saturday 7th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 08.00 EDT / 13.00 BST / 17:30 IST / 20.00 SGT and Sunday 8th Oct from 03.00 EDT / 08:00 BST / 12:30 IST / 15.00 SGT to 06.00 EDT / 11.00 BST / 15:30 IST / 18.00 SGT for essential maintenance. Apologies for the inconvenience.
* Professor J. J. Morrison, Department of Obstetrics and Gynaecology, National University of Ireland Galway, University College Hospital, Newcastle Road, Galway, Ireland.
Objective To investigate the direct effects of corticosteroids on human umbilical artery resistance, in vitro.
Design Prospective laboratory study.
Setting University teaching hospital.
Samples and methods Umbilical artery samples were obtained following normal, term deliveries (n= 50) and dissected rings were suspended for isometric recording under physiological conditions. The effects of hydrocortisone (10−9–10−4 M), dexamethasone (10−9–10−4 M) and betamethasone (10−9–10−4 M) on umbilical artery resistance were measured in vitro.
Main outcome measures Changes in umbilical artery resistance, in vitro.
Results Hydrocortisone (n= 12) exerted a vasodilatory effect on human umbilical artery at all concentrations studied compared with vehicle control experiments (n= 12) (P < 0.0001). The mean net relaxant effect of hydrocortisone ranged from 11.77% (10−9 M) to 57.01% (10−4). Both exogenous compounds, dexamethasone (n= 12) and betamethasone (n= 12), similarly exerted a significant relaxant effect on human umbilical artery tone (P < 0.05–0.01), compared with vehicle control experiments (n= 12). The mean net relaxant effect of dexamethasone ranged from 14.43% (10−9 M) to 38.12% (10−4) and that of betamethasone ranged from 6.02% (10−9 M) to 42.30% (10−4), in a cumulatively increasing fashion. There was a non-significant trend towards a greater vasodilatory effect of dexamethasone than betamethasone at lower bath concentrations studied.
Conclusion Corticosteroids exert a direct and potent vasodilatory effect on human umbilical artery resistance in vitro, thus providing an explanation for the previously unexplained vascular effects associated with antenatal administration of corticosteroids.
Preterm delivery is a major cause of infant morbidity and mortality1,2. Despite intensive clinical and research efforts investigating novel methods of intervention, the prevalence of preterm delivery has remained unchanged1–3. One intervention, the maternal administration of synthetic corticosteroids (dexamethasone or betamethasone) antenatally, clearly confers benefit by reducing morbidity and mortality from respiratory distress syndrome for preterm infants4–6. This has resulted in widespread use of corticosteroids antenatally in clinical practice. However, what has been less clear until recently is the possibility that such treatments may have adverse fetal effects with potential short and/or long term implications for the fetus and the developing child7,8. While knowledge concerning possible adverse fetal effects is limited, it has been recognised for some time that corticosteroids exert significant effects on the fetal cardiovascular system. Dexamethasone and betamethasone result in alterations in baseline fetal heart rate and fetal heart rate variability9–11. Wallace and Baker12 have reported that betamethasone administered during pregnancy is associated with decreased placental vascular resistance as measured by return of end-diastolic flow using Doppler ultrasonography in the umbilical artery in pregnancies complicated by absent end-diastolic flow. Similarly, after maternal dexamethasone or betamethasone treatment, a significant decrease in fetal middle cerebral artery impedance (pulsatility index) has been demonstrated in third trimester healthy fetuses13 and in third trimester high risk pregnancies14. No clear explanation exists for this decreased vascular resistance but theories have focussed on the possibility that exogenous corticosteroids increase placental secretion of corticotrophin releasing hormone (CRH), which is a potent vasodilator in the fetal placental circulation12,14,15. There are no data to our knowledge outlining the possible direct effects of exogenous or endogenous corticosteroids on feto-placental vascular tone. The aim of this study was to investigate the effects of hydrocortisone, dexamethasone and betamethasone on human umbilical artery resistance in vitro.
The study was carried out in the Department of Obstetrics and Gynaecology, University College Hospital Galway, Ireland between April 2001 and June 2002. Sections of human umbilical cord of approximately 10 cm in length were excised from the proximal segment of the cord (i.e. closest to the placental attachment), after delivery at term. All samples were collected within 20 minutes of delivery. Samples were immediately placed in cold, buffered Krebs Henseleit physiological salt of the following composition: potassium chloride 4.7 mmol/L, sodium chloride 118 mmol/L, magnesium sulphate 1.2 mmol/L, calcium chloride 1.2 mmol/L, potassium phosphate 1.2 mmol/L, sodium bicarbonate 25 mmol/L and glucose 11 mmol/L (Sigma-Aldrich, Dublin). Cord samples were refrigerated at 4°C and used within 24 hours of collection. The study was approved by the Research Ethics Committee at the University College Hospital Galway.
Umbilical arteries were carefully dissected free of Wharton's Jelly and cut in rings 4–5 mm in axial length. Rings were suspended individually on stainless steel hooks inserted into their lumens and mounted under 2 g (30 mN) of isometric tension, in glass-jacketed tissue baths. Each bath contained 20 mL of oxygenated (95% O2 and 5% CO2) Krebs Henseleit physiological salt solution at 37°C and pH 7.4. Rings were allowed to equilibrate for 90 minutes with regular washouts. During this interval, spontaneous tone developed. After the equilibration period, the vessel rings were challenged with 60 mM KCl. Washouts were carried out once the maximum response had reached a plateau, and a 20-minute recovery period was allowed in order that baseline be attained again. The KCl challenge was performed three times. After the last KCl challenge, 40 minutes recovery was allowed. Contraction was then stimulated by bath exposure of the vessel rings to 5-hydroxytryptamine (5-HT) (10−7 M). Once maximum contractile response to 5-HT was attained, the rings were allowed to remain at plateau for 20 minutes.
Concentration–effect experiments were performed by cumulative additions of hydrocortisone, dexamethasone or betamethasone, or respective vehicle, to produce one log molar increase in the bath concentration per addition. Bath concentrations in the range of 10−9–10−3 M were cumulatively attained at 20-minute intervals. The effects of hydrocortisone, betamethasone or dexamethasone, and of respective vehicle for each, were then assessed by calculation of the mean amplitude of selected areas for each 20-minute interval (i.e. 10−9, 10−8, 10−7etc.) using the Powerlab hardware package and Chart version 3.7 software (AD instruments, UK). The effects were demonstrated by expressing the mean amplitude calculated during the 20-minute period following addition of each drug concentration as a percentage of the mean amplitude obtained in the 20 minutes prior to any drug addition. This measurement represents percentage contractility, and subtracted from 100%, provides the percentage relaxation value for each bath concentration of vehicle and study compounds. The net relaxant effect of each corticosteroid was calculated by subtracting the percentage contractility value for each compound, at each bath concentration, from the percentage contractility value calculated for its respective vehicle control experiment, at each similar bath concentration. All of the umbilical artery samples used for experimentation for each corticosteroid were obtained from different women (i.e. n= 12 for hydrocortisone, for example, was achieved by using umbilical artery samples from 12 different women). The allocation of umbilical artery samples for the different experiments was entirely random.
Fresh Krebs Henseleit physiological salt solution was made up and buffered daily. KCl solutions were made up on the day of experimentation. A stock solution of 5-HT (Sigma-Aldrich) was made up in deionised water and diluted with Krebs solution. Dexamethasone (Sigma-Aldrich) was made up as a 10−2 M stock solution in methanol and stored at 4°C. Stock solutions of both hydrocortisone (Sigma-Aldrich) and betamethasone (Sigma-Aldrich) were made up in a 1:1 ratio of chloroform/ethanol and stored at room temperature. Serial dilutions of hydrocortisone, dexamethasone and betamethasone were made using Krebs Henseleit physiological salt solution.
Comparisons of measurements of amplitude for each bath concentration of hydrocortisone, dexamethasone and betamethasone or respective vehicle, were made using a 3 × 2 ANOVA. Post hoc comparisons were made using the Newman–Keuls protected t test. A P value <0.05 was accepted as statistically significant. The statistical package GBSTAT version 6.5 was used for statistical calculations.
Umbilical artery samples were obtained from 50 women after delivery at term. All pregnancies were uncomplicated and there was no evidence of hypertensive disease or intrauterine fetal growth restriction. The mean maternal age was 30.2 years (range 17–42 years). The median period of gestation was 39 weeks plus four days (range 36–42 weeks). The mode of delivery was as follows: normal vaginal delivery (n= 19), caesarean section (n= 21), instrumental vaginal delivery (n= 12). At the time of recruitment, 23 women were nulliparous and 27 women were parous.
A recording from control experiments (i.e. without addition of vehicle or corticosteroid), demonstrating umbilical artery resistance at maximum plateau for the time duration of experiments, is shown in Fig. 1A. The spontaneous relaxation values calculated are shown in Table 1. The endogenous corticosteroid hydrocortisone exerted a potent vasodilatory effect on human umbilical artery. Representative recordings of umbilical artery resistance in a vehicle control experiment (i.e. with vehicle only), and the relaxant effect of hydrocortisone, are shown in Figs 1B and 1C, respectively. The dose–response curves for hydrocortisone (n= 12), its vehicle control (n= 12) and control experiments without vehicle or corticosteroid are shown in Fig. 2. The percentage relaxation values calculated for hydrocortisone and for vehicle control experiments, at the range of bath concentrations outlined, are shown in Table 1. The net percentage relaxant effects of hydrocortisone (i.e. after consideration of the vehicle control effects) are shown in Table 2. The mean net relaxant effect of hydrocortisone varied from 11.77% (10−9 M) to 57.01% (10−4 M) in a cumulatively increasing fashion. Compared with vehicle control experiments, hydrocortisone exerted a significant vasodilatory effect at all bath concentrations (P < 0.05–0.01).
The observed relaxant effect on umbilical artery resistance in control rings (without vehicle or corticosteroid), vehicle-exposed rings (methanol or chloroform/ethanol) and corticosteroid-exposed rings (hydrocortisone, dexamethasone or betamethasone) are shown. Values are expressed as percentage [%] relaxation (SEM). Analysis by ANOVA for each bath concentration reveals a significant difference in relaxant effect (P < 0.0001) (n= 12 for all observations).
Table 2. Net percentage (%) relaxant effect of corticosteroids on umbilical artery resistance.
Values for the net relaxant effect (i.e. corrected for observed effect of respective vehicle) of hydrocortisone, dexamethasone and betamethasone on umbilical artery resistance, at each bath concentration of compound, are shown. Values are expressed as percentage [%] relaxation (SEM).
Both exogenous corticosteroids, dexamethasone and betamethasone, similarly exerted a potent relaxant effect on human umbilical artery tone. Examples of recordings of umbilical artery resistance after bath exposure to dexamethasone or betamethasone, compared with recordings after bath exposure to respective vehicle, are shown in Fig. 3 and Fig. 1 (vehicle for betamethasone is chloroform and is shown in Fig. 1B). The relaxant response was cumulative and immediate in nature. The dose–response curves for the net effects of dexamethasone (n= 12) and betamethasone (n= 12) on percentage contractility are shown in Fig. 4. The mean net relaxant effect of dexamethasone varied from 14.43% (10−9 M) to 38.12% (10−4 M) in a cumulatively increasing fashion (Table 2). Similarly, the mean net relaxant effect of betamethasone varied from 6.02% (10−9 M) to 42.30% (10−4 M). Both dexamethasone and betamethasone exerted a significant vasodilatory effect at all bath concentrations investigated (P < 0.05–P < 0.01) (Table 1).
The values for the net relaxant effect of dexamethasone and betamethasone on umbilical artery resistance (i.e. after correction for relaxant effect of respective vehicle) at all bath concentrations studied are shown in Table 2. There was a trend towards a greater vasodilatory effect of dexamethasone over betamethasone at lower bath concentrations studied but this difference was not statistically significant. Finally, there was no detected difference in the vasodilatory effects of the corticosteroids in relation to parity or mode of delivery.
This paper demonstrates that corticosteroids exert a direct and marked vasodilatory effect on human umbilical artery resistance. This provides a possible explanation for the hitherto unexplained variations in fetal heart rate and umbilical artery Doppler waveforms associated with antenatal administration of corticosteroids12. These findings raise further questions about the safety of antenatal corticosteroids, and particularly the clinical practice of repeated doses.
No previous studies have investigated the effects of corticosteroids on umbilical artery resistance in vitro. However, the effects of corticosteroids on other vascular smooth muscle types have been outlined. Numerous studies have demonstrated that corticosteroids potentiate contractile responses to catecholamines in various animal and human vascular smooth muscle types in vitro and in vivo16–18. This has led to the general belief that corticosteroids are predominantly involved in augmenting and maintaining vascular tone. However, it is recognised that in some vascular tissue types, corticosteroids exert a relaxant effect18–20, and hence, their role in the regulation of vasomotor tone is poorly understood. In addition, the mechanisms by which corticosteroids may alter vascular tone are not clearly defined. They have biochemical actions with the potential for both vasoconstriction (e.g. reduced prostacyclin production, reduction in expression or function of calcium-dependent potassium channels, increased α-adrenoreceptor numbers, decreased cGMP production, increased Ca2+ influx, inhibition of nitric oxide synthase)20–22 and vasodilatation (e.g. increased cAMP responses, reduction in expression of endothelin receptors, increased β-adrenoreceptor numbers)20,23. This study reports that corticosteroids exert a potent vasodilatory effect in human umbilical artery in vitro. While the mechanism of this effect requires further evaluation, our findings provide a possible explanation for the previously reported alterations in fetal heart rate9–11 and umbilical artery waveforms12 that occur in association with antenatal administration of corticosteroids. These effects had previously been unexplained or attributed to placental secretion of CRH, a potent vasodilator of the feto-placental circulation24,25.
Serotonin is a reproducible vasoconstricting agent for human umbilical artery and is hence used as a standard in experiments of this nature in order to compare the effects of other vasoactive compounds on the umbilical vasculature26–28. The relaxant effects of the corticosteroids were observed in all experiments and, using our computerised data acquisition system, the amplitude measurements obtained were highly accurate. A difference in the effects on umbilical artery resistance exerted by the two most commonly used therapeutic agents, dexamethasone and betamethasone, was not demonstrated despite the fact that betamethasone is believed to be more therapeutically effective and safer than dexamethasone29. Current treatment regimens with antenatal corticosteroids result in elevated fetal levels within 2 hours of administration30. These elevated levels are maintained during the course of treatment, resulting in near-maximal occupancy of receptors and induction of target proteins. Values return to normal by two days after the last dose30. This time scale of corticosteroid kinetics is similar to that of the observed variations in short term and long term fetal heart rate variability, the decreased fetal movements and altered umbilical artery Doppler waveform patterns reported in association with administration of antenatal corticosteroids10,11.
These findings demonstrate that corticosteroids exert a prolonged significant and direct vasodilatory effect on umbilical artery resistance in vitro. The potential implications of this for fetal wellbeing in the short or long term are unknown. There is a need for future research to focus on the regulation of vasomotor tone in the fetus.
This research was funded by the Health Research Board (HRB) of Ireland. The authors would like to thank the Midwifery Staff at University College Hospital Galway for their assistance in the collection of umbilical cord samples.