Administration of a single course of antenatal steroids results in decrease in neonatal morbidity and mortality as well as substantial savings in health care costs by specifically reducing the risk of respiratory distress syndrome, intraventricular hemorrhage, and neonatal death among premature infants.1–3 The preferred corticosteroids for antenatal therapy are betamethasone administered intramuscularly as two doses of 12 mg each 24 hours apart or dexamethasone four doses of 6 mg given intramuscularly every 12 hours. These agents are favored over other forms of steroids because they are the most widely studied, seem to have identical biologic activity, and readily cross the placenta. In addition, they are devoid of mineralocorticoid activity, have relatively weak immunosuppressive actions, and have longer duration of action in comparison with cortisol and methylprednisolone.4
However, there continue to be reports from uncontrolled studies of differences in effectiveness between betamethasone and dexamethasone. A meta-analysis3 showed that although the two agents reduce the risks of respiratory distress syndrome and intraventricular hemorrhage to a comparable extent, betamethasone was more consistently associated with reduction in neonatal death than dexamethasone. Furthermore a retrospective study5 found that betamethasone but not dexamethasone was associated with a decreased risk of periventricular leukomalacia, a major precursor of cerebral palsy. On the other hand, dexamethasone is substantially cheaper, more readily available and less likely to decrease fetal breathing movements and fetal heart rate variability. In addition a recent retrospective study6 did not find any differences in effectiveness in the reduction of neonatal mortality and morbidity, including intraventricular hemorrhage and periventricular leukomalacia.
The choice of which of these agents to use is currently based on ease of administration, cost, availability, and results from conflicting observational studies.
Our hypothesis was that there are no differences between betamethasone and dexamethasone with regards to effectiveness when used in the clinical setting of anticipated preterm delivery. Our objective was to compare betamethasone with dexamethasone in terms of effectiveness in reducing perinatal morbidities and mortality among preterm infants.
MATERIALS AND METHODS
We conducted a randomized, double-blind, placebo-controlled trial involving pregnant women at risk for delivering preterm at Stony Brook University Hospital from August 1, 2002, through July 31, 2004. These included all women in preterm labor with intact membranes, those with preterm premature rupture of membranes, and those anticipated to deliver for fetal and maternal indications between 24 and 33 weeks 6 days gestation.
Preterm labor with intact membranes was diagnosed in the presence of six to eight contractions per hour or four contractions in 20 minutes, associated with cervical changes but with no prelabor rupture of membranes. Preterm PROM included cases in which delivery followed prelabor rupture of membranes, documented by pooling of fluid on sterile speculum examination, ferning, and alkaline pH of fluid collected from the posterior vaginal fornix. Fetal and maternal indications included mainly intrauterine growth restriction, preeclampsia, and chronic hypertension with superimposed preeclampsia. We excluded women with clinical chorioamnionitis, known major fetal structural anomalies, known fetal chromosomal abnormalities, prior antenatal steroid exposure, steroid use for other indications, and quadruplets, as well as women who declined enrollment.
Subjects meeting inclusion criteria were approached by a resident or attending physician who explained the study and sought consent to participate in the study. Those who elected to participate in the study signed a formal consent form approved by our institution's Committee on Research on Human Subjects.
Consenting women were randomly allocated to one of two groups. The random allocation sequence was carried out by the Pharmacy using computer-generated random numbers, and each participant was assigned to one of two groups, ie, betamethasone or dexamethasone. One group received 12 mg of betamethasone (Celestone Soluspan, Schering, Kenilworth, NJ) intramuscularly at 0 and 24 hours and similar-appearing placebo at 12 and 36 hours. The second group received 6 mg of dexamethasone (dexamethasone sodium phosphate, Baxter Healthcare Corporation Anesthesia and Critical Care, New Providence, NJ) intramuscularly at 0, 12, 24, and 36 hours. Concealment was achieved by delivering all doses in identical-looking syringes covered by opaque material. Both subjects and health care providers were blinded as to the group to which participants belonged.
After delivery, study personnel blinded to group assignment reviewed the prenatal, delivery, neonatal, and postpartum records and documented maternal and neonatal independent outcome variables. The primary outcome variables were respiratory distress syndrome, intraventricular hemorrhage, and neonatal death. Secondary outcome variables included the need for exogenous surfactant, oxygen dependency at 28 days after birth or oxygen dependency at equivalent of 36 completed weeks gestational age, retinopathy of prematurity, inotropic support for hypotension initiated during the first 24 hours of life, treatment for patent ductus arteriosus, necrotizing enterocolitis, neonatal sepsis, and histologic chorioamnionitis. Other variables recorded were maternal age, clinical diagnosis, gestational age at randomization and at delivery, gender, and birth weight.
Respiratory distress syndrome was diagnosed clinically, by the need for mechanical ventilation and oxygen for at least 48 hours, and the presence of radiologic chest findings. Each neonate had transfontanelle head ultrasound scans on days 3 and 7 of life. Neurosonograms were evaluated by an experienced radiologist blinded to the antenatal steroid exposure status of the parturient. Intraventricular hemorrhage was graded as described by Papile et al7 Periventricular leukomalacia was diagnosed by the presence of echolucent areas or persistent echogenicity in the periventricular areas on coronal and sagittal views. For study purposes, “any brain lesion” was defined as any grade of intraventricular hemorrhage or periventricular leukomalacia present exclusively or in combination." Necrotizing enterocolitis was diagnosed clinically and radiologically and confirmed at surgery or autopsy. Proven neonatal sepsis included positive blood, cerebrospinal fluid, or urine cultures.
Data analysis was performed in accordance of the intention-to-treat principle. Sample size computations for an equivalence study were performed according to the work of Blackwelder.8 We assumed that approximately 80% of neonates exposed to either betamethasone or dexamethasone will not develop respiratory distress syndrome. We also defined a deviation in effectiveness of 10% or less as been essentially equivalent. A sample size of 354 neonates (177 exposed to betamethasone compared with 177 exposed to dexamethasone) would provide more than 80% power to detect differences of more than 10% for a two-tailed test of significance at a critical level of 0.05. The distributional characteristics of the variables were examined. Continuous data were normally distributed. As such, differences between groups defined by exposure to antenatal steroids were examined using Student t test for continuous variables and χ2 for categorical variables. Fisher exact test was used when expected cell frequency was equal to or less than five. A P value of <.05 was considered statistically significant.
Five hundred forty-three women were screened for eligibility. One hundred sixty (29.5%) received antenatal steroids before their transport and subsequently had completion of their course based on what was initiated in the transferring hospital. Fifty-five (10.1%) women declined participation and were given antenatal steroids based on the discretion of the treating physician and availability, whereas 29 (5.3%) women could not be approached because of immediate or imminent delivery. Thus, the study cohort consisted of 299 (55.1%) women and their 359 neonates. The profile of the study participants is shown in Figure 1. There were no statistically significant differences between the groups in terms of maternal age, race, body mass index, smoking status, gestational age at randomization, and diagnoses between women who participated in the trial and those who did not participate. Similarly, the baseline characteristics of the women and neonates randomly assigned to betamethasone were not different from those that assigned to dexamethasone, as depicted in Tables 1 and 2. No adverse effect or side effects were reported in either group.
There were no significant differences between neonates exposed to betamethasone and those exposed to dexamethasone with regard to the rates of respiratory distress syndrome (73 of 181 [40.5%] compared with 79 of 178 [44.4%], P=.53), necrotizing enterocolitis (0 of 181 [0%] compared with 2 of 178 [1.1%], P=.25), retinopathy of prematurity (28 of 181 [15.5%] compared with 26 of 178 [14.6%], P=.92), patent ductus arteriosus (12 of 181 [6.7%] compared with 14 of 178 [7.9%], P=.81), neonatal sepsis (16 of 181 [8.9%] compared with 18 of 178 [10.1%], P=.83), bronchopulmonary dysplasia (27 of 181 [15.0%] compared with 18 of 178 [10.1%], P=.22), need for vasopressor (14 of 181 [7.8%] compared with 6 of 178 [3.4%], P=.07), and neonatal mortality (5 of 181 [2.8%] compared with 6 of 178 [3.4%], P=.98) (Table 3). However, the rates of intraventricular hemorrhage (6 of 105 [5.7%] compared with 17 of 100 [17.0%], relative risk 2.97, 95% confidence interval [CI] 1.22–7.24, P=.02) and any brain lesion (7 of 105 [6.7%] compared with 18 of 100 [18.0%], relative risk 2.7, 95% CI 1.18–6.19, P=.02) were significantly lower in neonates exposed to dexamethasone compared with betamethasone (Table 3).
The rate of grades III and IV intraventricular hemorrhage was 7% (7 of 100) in neonates exposed to betamethasone compared with 1.9% (2 of 105) in neonates exposed to dexamethasone, P=.09 (Table 3). Furthermore, the rate of periventricular leukomalacia was 4.0% (4 of 100) in neonates exposed to betamethasone compared with 1.9% (2 of 105) in neonates exposed to dexamethasone, P=.44 (Table 3). The absolute risk reduction in the rate of intraventricular hemorrhage was 11.3% (95% CI 2.7–11.9%), and the number needed to treat was 9 (95% CI 5–37) in favor of dexamethasone.
A single course of antenatal steroids before preterm delivery results in a significant decrease in neonatal morbidity and mortality as well as substantial savings in health care costs. Betamethasone and dexamethasone continue to be the preferred steroids because of their unique biologic properties, ability to readily cross the placenta, and because of their weak immunosuppressive and mineralocorticoid activities. The choice between betamethasone and dexamethasone is currently based on ease of administration, cost, availability, and results from conflicting observational studies.
Our study found that both betamethasone and dexamethasone are largely comparable in reducing most morbidities and mortality among preterm neonates. This finding is consistent with the reports of Baud et al5 and Bar-lev and colleaques,6 who did not demonstrate any differences in the rate of respiratory distress syndrome, bronchopulmonary dysplasia, intraventricular hemorrhage, neonatal mortality, and other complications of prematurity among neonates exposed to betamethasone and dexamethasone. However, contrary to the finding of a higher rate of periventricular leukomalacia among neonates exposed to dexamethasone compared with betamethasone reported by Baud et al,5 our study did not find any differences in the rate periventricular leukomalacia between the groups. In fact, the rate of periventricular leukomalacia in the dexamethasone-exposed neonates was lower among neonates exposed to betamethasone, although our study was underpowered to detect differences in periventricular leukomalacia. Three other groups of investigators6,9,10 have reported no differences in the rates of periventricular leukomalacia among neonates exposed to betamethasone and dexamethasone.
The differences in intraventricular hemorrhage after steroid exposure seen in our study may be related to potency, duration of exposure, and fetal response after exposure. Dexamethasone is five times more potent than betamethasone in nongenomic effects in addition to the stereoisomer difference of a methyl group at position 16 of ring D in their molecular structure.11,12 Derks et al12 described a more rapid fall in fetal sheep progesterone levels and a sharper rise in serum cortisol levels after betamethasone administration compared with dexamethasone, and the cortisol levels continued to rise at 72 hours postexposure in the dexamethasone-treated fetal sheep. Ballard and Liggins13,14 showed umbilical cord cortisol levels increased more rapidly and with a higher zenith after each 12 mg of betamethasone compared with 6 mg of betamethasone, with no reported increased benefit to 24 mg betamethasone dosing. The exact difference in physiologic responses of these stereoisomer molecules is not known and requires further investigation.
Our study largely supports the continuing use of both betamethasone and dexamethasone in the treatment of women at risk of preterm delivery. However, we found dexamethasone to be superior to betamethasone in reducing the rate of intraventricular hemorrhage. Other potential advantages of dexamethasone include lower cost, widespread availability, and less effect on fetal biophysical variables that might lead to premature intervention or delivery. On the other hand, betamethasone requires two injections as opposed to the four injections of dexamethasone.
1. National Institutes of Health. Report of the Consensus Development Conference on the Effect of Corticosteroids for Fetal Maturation on Perinatal Outcomes (NIH Publication No. 95–3784). Bethesda (MD): National Institutes of Health, National Institute of Child Health and Human Development; 1994
2. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of distress syndrome in premature infants. Pediatrics 1972;50:515–25.
3. Crowley PA. Antenatal corticosteroid therapy: a meta-analysis of the randomized trials, 1972 to 1994. Am J Obstet Gynecol 1995;173:322–35.
4. Ballard PL, Ballard RA. Scientific basis and therapeutic regimens for use of antenatal glucocorticoids. Am J Obstet Gynecol 1995;173:254–62.
5. Baud O, Foix-L'Helias L, Kaminski M, Audibert F, Jarreau PH, Papiernik E, et al Antenatal glucocorticoid treatment and cystic periventricular leukomalacia in very premature infants. N Engl J Med 1999;341:1190–6.
6. Bar-Lev MR, Maayan-Metzger A, Matok I, Heyman Z, Sivan E, Kuint J. Short-term outcomes in low birth weight infants following antenatal exposure to betamethasone versus dexamethasone. Obstet Gynecol 2004;104:484–8.
7. Papile LA, Burstein J, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: a study of infants with birth weights less than 1,500 gm. J Pediatr 1978;92:529–34.
8. Blackwelder WC. “Proving the null hypothesis” in clinical trials. Control Clin Trials 1982;3:345–53.
9. LeFlore JL, Salhab WA, Broyles RS, Engle WD. Association of antenatal and postnatal dexamethasone exposure with outcomes in extremely low birth weight neonates. Pediatrics 2002;110:275–9.
10. Lee BH, Stoll BJ, McDonald SA, Higgins RD, National Institute of Child Health and Human Development Neonatal Research Network. Adverse neonatal outcomes associated with antenatal dexamethasone versus antenatal betamethasone. Pediatrics 2006;117:1503–10.
11. Derks JB, Giussani DA, Van Dam LM, Jenkins SL, Winter JA, Zhao XF, et al. Differential effects of betamethasone and dexamethasone fetal administration of parturition in sheep. J Soc Gynecol Investig 1996;3:336–41.
12. Buttgereit F, Brand MD, Burmester GR. Equivalent doses and relative drug potencies for non-genomic glucocorticoid effects: a novel glucocorticoid hierarchy. Biochem Pharmacol 1999;58:363–8.
13. Ballard PL, Liggins GC. Glucocorticoid activity in cord serum: comparison of hydrocortisone and betamethasone regimens. J Pediatr 1982;101:468–70.
14. Ballard PL. Scientific rationale for the use of antenatal glucocorticoids to promote fetal development. Pediatr Rev 2000;1:E83–90.