Effect of dexamethasone and betamethasone on fetal heart rate variability in preterm labour: a randomised study

Authors


Correspondence: Dr Y Ville, Fetal Medicine Unit, St George's Hospital, Cranmer Terrace, London SW 17 0RE, UK.

Abstract

Objective To compare the effects of betamethasone and dexamethasone on fetal heart rate in appropriately grown fetuses.

Methods Eighty-two pregnant women (97 fetuses) with preterm labour were randomly allocated to receive betamethasone (n= 42) or dexamethasone (n= 40) for fetal lung maturation in a nonblinded fashion. Computerised cardiotocogram (CTG) parameters were compared before, during and after treatment.

Results A decrease in fetal heart rate variability was found with betamethasone but no significant changes were found with dexamethasone. Fetal heart rate variability returned to pre-treatment values within a week after cessation of treatment with betamethasone. Neonatal outcome was similar in the two groups.

Conclusions These findings might prove useful in the management of compromised fetuses with decreased fetal heart rate variability in which the CTG should be used together with other parameters to assess fetal wellbeing during corticosteroid treatment. Dexamethasone may be preferable as the drug of choice since it was associated with significantly less alteration in fetal heart rate variability compared with betamethasone.

INTRODUCTION

Antenatal corticosteroid treatment has a beneficial effect on fetal lung maturation when given to the woman at least 24 hours before preterm delivery between 24 and 34 weeks of gestation1–4. Potential side effects include neonatal infection and adrenal suppression, but these do no appear to be clinically significant in appropriately grown fetuses5,6. However, fetal heart rate variation or variability can be affected by corticosteroids7–10; decrease in fetal heart rate variability has been reported with betamethasone8,9, and the reverse was found with dexamethasone7. Since different regimens were used in heterogeneous population7–10 we examined the effect of equivalent dosage regimens of betamethasone and dexamethasone on computerised fetal heart rate traces in appropriately grown fetuses in a prospective randomised study.

METHODS

During a 12–month period (October 1994 to October 1995) 2150 women were delivered in our institution, and 210 pregnant women were diagnosed as having mild to moderate premature contractions with intact membranes between 25 and 33 weeks of gestation. They were admitted to hospital and bedrest was advised in all cases; salbutamol (Glaxo, Boulogne, France) was given intravenously for at least 48 hours when > 5 contractions per hour were recorded at presentation. All women who presented at < 34 weeks of gestation were given steroids for enhancement of fetal lung maturation. In utero transfers with uncertain pregnancy history and women who presented with clinical infection, vaginal bleeding or suspicion of premature rupture of the membranes were not included in the study. Women who received any additional treatment, such as antihypertensive drugs or benzodiazepines, other than salbutamol were not included as these treatments could interfere with the analysis. Eighty-two women were therefore randomly allocated by a table of random numbers held by an independent investigator to either one of two different corticosteroid regiments in a nonblinded fashion: 1. four intramuscular injections of 3 mg betamethasone sodium and 3 mg betamethasone acetate (Célestene Chronodose; Roche, Paris, France) were given 12 hours apart in 42 pregnancies (53 fetuses); or 2. four intramuscular injections of 4 mg dexamethasone acetate (Soludécadron; Merck, Paris, France) were given 12 hours apart in 40 pregnancies (44 fetuses). Steroids were given at weekly intervals until preterm labour resolved. In women who received more than one course of steroids, only the first one was kept for analysis. In multiple pregnancy, one fetus was randomly selected for analysis. The outcome of pregnancy was assessed by recording gestational age at delivery, birthweight and neonatal complications such as collapse, respiratory distress syndrome, enterocolitis, intraventricular haemorrhage, and periventricular leukomalacia. All patients had cardiotocogram (CTG) records before (baseline), 24 to 48 hours and 4 to 7 days (mean 5) after the first injection of corticosteroids was given to the mother. Records were obtained in 97 fetuses from a median gestational age of 29 (range 25–33) weeks of gestation until premature contractions resolved.

All women were studied in standardised conditions: observations were completed in an average of 30 minutes, two to three hours after a meal between 10:00 h and 18:00 h. The women were not allowed to smoke, drink or eat during the cardiotocogram. Simultaneous records were made on twins using two separate fetal monitors and computers. The fetal heart rate was recorded with a fetal heart rate monitor (Sonicaid Team, Sonicaid Oxford, UK) connected to a Sonicaid system 8002 computer. The system of computerised analysis has been described previously11. Briefly, fetal heart rate irregularity (variation or variability) may be calculated; this is a mathematical approximation of true variability in the sense that the fetal heart rate source in antepartum records is derived from a processed Doppler signal rather than the fetal electrocardiogram. Long term fetal heart rate variation was defined as the mean minute range of pulse intervals about the baseline, and short term (1/16th min epochal) variation as the mean epoch-to-epoch variation, averaged over the record and expressed in milliseconds. Episodes of high or low fetal heart rate variations were identified as 5/6 consecutive minutes in each of which the range of pulse interval was > 32 ms or < 30 ms, respectively. All pregnancies were dated according to the first day of the last menstrual period and ultrasonographic measurement of crown-rump length before the 13th gestational week. Fetal heart rate computerised parameters were analysed for each fetus and compared before, during and after treatment; each fetus was taken as its own control.

All data are presented as medians and interquartile ranges. The values obtained on the control day (baseline) were examined for any significance between the independent maternal betamethasone and dexamethasone groups. This was carried out using the Student t test for each CTG parameters. Differences in the measurements (baseline, 24 to 48 hours and 4 to 7 days) were assessed by repeated measures analysis of variance. The Scheffe F-test was used to compare pairs of mean values when the analysis of variance indicated a significant trend over time. The differences between the baseline measurements and the measurements taken at 24 to 48 hours in each group were calculated to compare the effect of betamethasone and dexamethasone by using the unpaired t test.

RESULTS

Group 1: betamethasone

Forty-two pregnancies, including 32 singletons, nine twins and one triplet, were randomly allocated to be exposed to 42 antenatal courses of betamethasone at 25 to 33 weeks of gestation (median 29). Patients were advised bedrest for an average of 10 days in hospital and received 4 to 12 mg salbutamol intravenously. Delivery occurred at 30 to 40 weeks (median 37), 2 to 10 weeks (median 6) after the steroids were given. Clinical characteristics of the pregnancies are shown in Table 1. All newborn infants had a birthweight within the normal range for gestation (980 to 3840 g, median 2500). Transient respiratory distress syndrome occurred in one case that required two days of assisted ventilation; no neonatal deaths occurred. One infant had a bowel volvulus; surgery was performed on the first day after birth and the post-operative course was uneventful.

Table 1.  Comparison of baseline clinical characteristics and fetal outcome in patients randomly allocated to betamethasone (Group 1) or dexamethasone (Group 2).
 Group 1Group 2
  1. * Pulmonary hypoplasia related to cystic adenomatoid malformation of the lungs

No. of patients4240
Singleton pregnancies3236
Twin pregnancies94
Triplet pregnancies10
Parity
Nulliparae2321
Multiparae1919
Gestational age (weeks)
At recording29[25–33]29 [25–33]
At delivery37[30–40]37[30–40]
≤ 2700
28–3376
34–361210
≥ 372324
Birthweight (g)2500[980–3840]2550[1130–3910]
Fetal outcome
Death01*
RDS10
Collapse00
Enterocolitis00
Intraventricular haemorrhage00
Periventricular leukomalacia00

Significant changes in CTG parameters were found; number of accelerations, long term and short term variation (Table 2 and Table 3). The number of accelerations increased significantly (Fig. 1) after treatment was stopped. Long term variation increased significantly after cessation of treatment and returned to pre-treatment values within a week (Fig. 2). Short term variation decreased significantly during the course of betamethasone, and this was followed by a significant increase after cessation of treatment to return to pre-treatment values (Fig. 3).

Table 2.  Comparison of selected computerised CTG parameters in 42 antenatal courses of betamethasone between: baseline, 24 to 48 hours, and 4 to 7 days after the course. Medians and interquartile ranges are given. Difference in the measurements were assessed by repeated measures analysis of variance.
 Baseline24 to 48 h4 to 7 daysP
Fetal heart rate (bpm)145.0 (141.0–148.0)147.0(141.0–154.0)148.0 (142.0–154.0)0–.5
Long term variation (ms)37.0 (30.0–47.0)34.2 (28.6–39.0)38.5 (33.4–43.4)0.01
Short term variation (ms)6.9 (5.9–9.0)6.0 (5.4–7.3)7.0(6.0–8.3)0.01
High variation episode (min)13.0(6.0–19.0)9.5 (6.0–22.0)14.0(8.0–18.0)0.69
Low variation episode (min)0.0 (0.0–6.2)7.0(0.0–11.2)0.0(0.0–7.2)0.12
Acceleration ≤ 10 bpm (n)5.0 (3.0–12.0)7.0(3.0–10.0)9.5(6.0–12.0)0.01
Deceleration ≤ 10 bpm (n)0.0 (0.0–0.0)0.0 (0.0–0.0)0.0 (0.0–0.0)0.16
Uterine contractions (n)1.8 (0.0–2.0)1.0(0.0–3.0)1.0(0.0–2.0)0.45
Table 3.  Comparison of computerised CTG parameters in 42 antenatal courses of betamethasone at baseline and 24 to 48 hours, 24 to 48 hours and 4 to 7 days, baseline and 4 to 7 days, using Scheffe F-test for multiple comparisons. Values are given as mean difference. Key as for Table 2.
 Baseline & 24 to 4824 to 48 hours & 4 to 7 daysBaseline & 4 to 7 days
  1. * Significant at 5% level.

Long term variation (ms)3.3−4.6*−1.3
Short term variation (ms)1.1*−1.0*0.1
Accelerations > 10 bpm (n)0.1−2.4*−2.3*
Figure 1.

Accelerations (n) during the study period in Group 1: betamethasone. Medians and interquartile ranges are given. **P < 0.05

Figure 2.

Long term fetal heart rate (FHR) variation (ms) during the study period in Group 1 : betamethasone. Medians and interquartile ranges are given. **P < 0.05

Figure 3.

Short term fetal heart rate (FHR) variation (ms) during the study period in Group 1: betamethasone. Medians and interquartile ranges are given. **P < 0,05

Short term variation decreased by 28% of the initial value during the course (n= 26,62%). There was a concomitant increase in basal fetal heart rate by 2.6% and a decrease in long term variation by 20%. Short term variation increased by 13% of the initial value during the course (n = 16,38%). There was a concomitant decrease by 2.6% in basal fetal heart rate and an increase by 14% in long term variation. Uterine contractions and heart rate decelerations did not change throughout the course of betamethasone.

Group 2: dexamethasone

Forty pregnancies, including 36 singletons and four twins, were randomly allocated to 40 antenatal courses of dexamethasone at 25 to 33 weeks of gestation (median 29). Bedrest was advised for an average of 10 days in hospital and the women received 4 to 12 mg salbutamol intravenously. Delivery occurred at 30 to 40 weeks (median 37), 1 to 11 weeks (median 6) after steroids were given. Birthweight of all infants was within the normal range for gestational age (1130 to 3910 g, median 2550). There was no respiratory distress syndrome (Table 1). One baby died in the neonatal period after a normal delivery at 38 weeks from pulmonary hypoplasia related to cystic adenomatoid malformation of the lungs; this case was excluded from the study and statistics were therefore derived from the analysis of 39 pregnancies. None of the computerised CTG parameters including accelerations, long term and short term variation (Figs 4, 5 and 6 respectively) changed throughout the course of dexamethasone (Table 4)

Table 4.  Comparison of selected computerised CTG parameters in 40 antenatal courses of dexamethasone between: baseline, 24 to 48 hours, and 4 to 7 days after the course. Medians and interquartile ranges are given. Difference in the measurements were assessed by repeated measures analysis of variance.
 Baseline24 to 48 h4 to 7 daysP
Fetal heart rate (bpm)146.0 (140.0–151.0)142.0 (137.0–149.0)139.0(136.0–146.0)0.16
Long term variation (ms)36.8(31.3–44.2)36.0(29.4–49.6)33.0 (27.7–47.7)0.58
Short term variation (ms)6.7 (5.9–8.2)6.8 (5.7–8.8)6.0(5.1–7.8)0.42
High variation episode (min)12.0(6.0–20.2)9.0 (5.0–14.0)14.0(8.0–19.2)0.50
Low variation episode (min)0.0 (0.0–9.0)0.0(0.0–9.0)6.0(0.0–9.2)0.60
Acceleration ≤ 10 bpm (n)6.0(3.0–8.0)5.0(2.2–10.0)6.0 (4.0–9.2)0.60
Deceleration ≤ 10 bpm (n)0.0(0.0–0.0)0.0 (0.0–0.0)0.0 (0.0–0.0)0.30
Uterine contractions (n)1.0(1.1–4.0)1.0(0.0–3.0)1.0(1.0–3.0)0.80

Comparison of groups 1 and 2

There was no difference in gestational age at diagnosis and at delivery, incidence and severity of prematurity, birthweight and fetal outcome (Table 1). The incidence of neonatal respiratory distress syndrome was similar in both groups. Salbutamol was the only drug used. All baseline computerised CTG parameters were compared in the two groups and none of the differences were significant.

The difference in short variation between the baseline measurement and 24–48 hours after the first injection was significantly different for the two steroids. (Table 5) The same was found for long term variation. (Table 5) Basal fetal heart rate changed in opposite directions in the two groups, with an increase during betamethasone (Fig. 7A) and a decrease during dexamethasone (Fig. 7B) courses. None of the steroids used was associated with any effect on uterine contractions and fetal heart rate decelerations. (Table 5).

Table 5.  Comparison of mean (standard deviation) difference between baseline measurements and at 24 to 48 hours for each selected computerised CTG parameters in Group 1 (betamethasone) and Group 2 (dexamethasone). Comparisons are made by using unpaired t test.
 Group 1Group 2 
 Mean difference between baseline and 24 to 48 h(SD)Mean difference between baseline and 24 to 48 h(SD)P
Fetal heart rate (bpm)−1.4(8.8)3.7(8.8)0.01
Long term variation (ms)3.4(9.8)−3.0(14.8)0.03
Short variation (ms)1.2(2.2)−0.4(2.4)0.01
High variation episode (min)−0.2(10.6)1.7(9.5)0.40
Low variation episode (min)−1.7(8.1)−0.8(9.3)0.60
Acceleration ≤ 10 bpm (n)0.1(4.8)−1.2(5.4)0.30
Deceleration ≤ 10 bpm (n)0.2(0.5)0.1(0.5)0.16
Uterine contractions (n)0.4(3.1)0.7(2.6)0.65
Figure 7(a).

Fetal heart rate (bpm) during the study period in Group 1: betamethasone. Medians and interquartile ranges are given.

Figure 7(b).

Fetal heart rate (bpm) during the study period in Group 2: dexamethasone. Medians and interquartile ranges are given.

DISCUSSION

Recently, the American College of Obstetricians and Gynecologists’ Committee on Obstetric Practice Recommendation3 and the National Institute of Health Consensus Conference4 confirmed that short term antenatal steroid therapy is effective in preventing neonatal respiratory distress syndrome and appears to have no detrimental effects on adrenal function of newborn infants5 However, the effect of synthetic steroids on the fetal nonstress test are still controversial. Computerised fetal heart rate analysis provides quantification of baseline fetal heart rate elements and overcomes the difficulty of visual assessment to distinguish trends in variability or acceleration counts. Dawes et al.7 reported an increase in short term fetal heart rate variation with dexamethasone (12 mg × 2 within 48 h) in a group of high risk pregnancies. Derks et al.8 reported a decrease in long term and short term fetal heart rate variation with betamethasone (Celestone 12 mg × 2, 12 hours apart) in high risk pregnancies including seven intrauterine growth retarded fetuses and five pre-eclamptic women with appropriately grown fetuses. We have shown that both steroids were likely to affect fetal heart rate variability in opposite directions and that fetal heart rate variability was decreased by betamethasone.

Betamethasone and dexamethasone are believed to act on the same receptors. Irrespective of their effect on fetal lung maturation, different effects with betamethasone and dexamethasone might be explained by different doses and schedules of administration of the two steroids as well as by the heterogeneity of the populations studied7–11. The consensus panel recommended treatment regimens of either two doses of 12 mg of betamethasone (acetate + phosphate forms) given intramuscularly 24 hours apart1. or four doses of 6 mg of dexamethasone given intramuscularly 12 hours apart2. However, these choices were empirical. All synthetic corticosteroids have a greater affinity for glucocorticoid receptors than cortisol, and dexamethasone and betamethasone are 7.1 -fold and 5.4–fold higher, respectively; this gives a 1.5–fold higher activity for dexamethasone when compared with betamethasone12. We have therefore used four doses of 4 mg dexamethasone given intramuscularly 12 hours apart in order to reach a pharmacological effect equivalent to our betamethasone regimen (four doses of 6 mg given intramuscularly 12 hours apart). Our patients were receiving salbutamol together with steroids but no other drugs; this could potentially affect our results by superimposing the effect of the tocolytic on the fetal heart rate variability. However, randomisation should allow for a valid comparison; this is frequently the treatment in clinical practice.

Postnatally, Dorr et al.5 reported a marked suppression of all glucocorticoids in infants delivered during a course of betamethasone; return to normal was preceded by a rebound effect with plasma cortisol levels reaching two to three times control levels. The same has been reported with dexamethasone2. In a previous study in normal twin fetuses9 we have found a significant decrease in fetal heart rate variability during the course of betamethasone; these changes returned to normal within a week, following a rise in accelerations compatible with a rebound effect after cessation of treatment. This is confirmed by the data presented here. Dawes et al.7 also suggested an initial rise that was not constant and limited to the subsequent day with dexamethasone.

We have shown that with both steroids an increase in short term variation was seen together with a decrease in basal fetal heart rate, and the reverse was seen when short term variation was decreased. This is compatible with fetal cardiovascular response to cortisol infusion described in sheep fetuses together with a rise in blood pressure13. Administration of synthetic steroids is likely initially to increase corticosteroid-related cardiovascular effects and should be followed by a decrease due to a damping effect of adrenal function. Since dexamethasone has a higher glucocorticoid effect than betamethasone, it should have a higher initial flare-up effect; the subsequent decrease in variability might therefore be attenuated when compared with initial values. This could provide an explanation for the discrepancy in the effects of the two steroids on fetal heart rate variability with an overall decrease with betamethasone and no change with dexamethasone. These results are therefore unlikely to be clinically significant in normally grown fetuses; however, a 30% decrease in fetal heart rate variability could be considered as a sign of fetal distress in an already compromised fetus (ie, intrauterine growth retardation). It is therefore important to consider each fetus as its own control to monitor these transient alterations of fetal heart rate variability. Assessment of fetal wellbeing in high risk fetuses should combine the use of Doppler examination and biophysical profile with that of CTG in order to avoid iatrogenic preterm delivery during the course of steroids. Dexamethasone has proved to be as effective as betamethasone to prevent neonatal respiratory distress syndrome and is less likely to affect fetal heart rate variability.

Ancillary