The ability of antenatal corticosteroid therapy to decrease the frequency of respiratory distress syndrome (RDS) and intraventricular hemorrhage (IVH) in pre-term neonates has been well described.1–3 Less obvious, however, is the necessity for repeated weekly doses after initial corticosteroid treatment because of the potential for diminished steroid effect after 7 days.2,3 Animal studies showed conflicting results regarding the duration of beneficial corticosteroid effect, specifically accelerated lung maturity.4–7 Therefore, we investigated whether frequencies of selected perinatal outcomes, including RDS, IVH, and perinatal infections, were influenced by length of interval between antenatal betamethasone administration and delivery.
Materials and Methods
We did a retrospective cohort study of singleton live-born neonates delivered between 28 and 34 weeks' gestation after a single course of betamethasone, defined as two 12-mg doses over 24 hours. Data were prospective and entered in a validated perinatal database at the Medical University of South Carolina between January 1996 and January 1999. Women who received a single course of betamethasone were categorized in three groups based on length of interval between first injection of betamethasone and delivery (1–2 days, 3–7 days, and 8–14 days). Subjects were admitted for preterm labor tocolysis with intact membranes and received parenteral magnesium sulfate. Estimated gestational age was confirmed by ultrasound examination on admission. Rectovaginal cultures were collected from all women for group B streptococci, and they were also tested for chlamydia and gonorrhea. Subjects received antibiotic prophylaxis against group B streptococci of ampicillin or clindamycin. Women with group B streptococci colonization subsequently were treated during the intrapartum period. Any women who received more than two doses of betamethasone (single course) during pregnancy or those who delivered before receiving the second dose were excluded. Subjects with ruptured membranes for more than 24 hours before delivery, who required insulin for diabetes, or had fetuses with known anomalies also were excluded. Maintenance tocolysis was not used routinely, but some women in each group received short courses of indomethacin (for less than 48 hours) or terbutaline therapy at the discretion of attending physicians. Indications for delivery included failed tocolysis, nonreassuring fetal testing, and clinical chorioamnionitis. Amniocentesis for fetal lung maturity estimation was not done on any subjects.
Groups were compared for selected demographic characteristics, gestational age at delivery, additional tocolytic exposure, mode of delivery, birth weight, clinical chorioamnionitis, maternal group B streptococci colonization status, and postpartum endometritis. Neonatal outcomes included RDS, grade 3 or 4 IVH, days of ventilator use, surfactant therapy, and early-onset neonatal sepsis. Diagnoses of RDS were made by neonataologists on the basis of accepted clinical manifestations and standard radiographic evidence. Neonates had cranial ultrasound evaluations within the first 7 days of life and repeated examinations within 30 days of life to detect IVH, which was graded using criteria established by Papile and colleagues.8 Clinical chorioamnionitis was defined as three of four of the following criteria without other causative factors: Maternal fever (temperature above 100.4F), uterine tenderness, maternal tachycardia (more than 120 bpm), and sustained fetal tachycardia (more than 160 bpm). Early-onset neonatal sepsis was defined as positive blood or cerebrospinal fluid cultures collected within the first 48 hours of life. Diagnoses of postpartum endometritis were made clinically on the basis of maternal fever (above 100.4F) and uterine tenderness on two occasions at least 6 hours apart, and after an initial 24-hour observation.
Continuous data were analyzed for normal distribution and tested for significance using Student t test. Categorical data were tested using χ2 and Fisher exact test. To find the confounding effect of multiple variables, two multiple logistic regression analyses were done using RDS and grades 3–4 IVH as dependent variables in separate analyses. Variables considered risk factors for RDS and IVH were included. Two-tailed P < .05 was considered statistically significant. For sample computations we assumed that approximately 80% of subjects delivered within 7 days. We also assumed a 40% incidence of RDS in women who delivered more than 7 days after therapy, which was expected to be similar to women not exposed to steroids in that gestational age range. Therefore, a sample of 200 neonates (40 who delivered after 7 days compared with 160 who delivered within 7 days) would provide more than 80% power to detect a 50% reduction in the incidence of RDS (ie, 40% compared with 20%) for a two-sided test of significance at a critical level of .05.
Two hundred sixteen subjects were included, 97 of whom delivered in 1–2 days, 78 in 3–7 days, and 41 in 8–14 days after a single course of betamethasone. Groups were similar in selected demographic characteristics, additional tocolytic exposures, mean gestational ages at delivery, mean birth weights, frequencies of vaginal deliveries, and frequencies of maternal group B streptococci colonization (Table 1). Intervals between first betamethasone dose and delivery differed significantly among the groups. The groups had similar frequencies of RDS, ventilator use, surfactant therapy, IVH (grades 3 or 4), clinical chorioamnionitis, postpartum endometritis, and early-onset neonatal sepsis (Table 2). Multiple logistic regression analysis confirmed that delivery more than 7 days after the first dose of corticosteroid therapy was not statistically significantly associated with RDS or IVH (grades 3–4) (Table 3). Younger gestational age was the only statistically significant risk factor independently associated with RDS and advanced grades of IVH.
We found that delivery between 8 and 14 days after a single course of antenatal betamethasone was not associated with increased perinatal morbidity compared with delivery at shorter intervals. However, those findings are limited to pregnancies without premature rupture of membranes (PROM) that delivered between 28 and 34 weeks' gestation. We chose to limit the sample to that gestational age range to identify the greatest possible effect on incidence of RDS because before 28 weeks there is little effect theoretically on the developing fetal type II pneumocytes.
Much recent debate on repeated doses of antenatal corticosteroids is based on a potentially diminished effect of corticosteroids after 1 week. A meta-analysis2 of seven randomized controlled trials that specifically evaluated the effect of corticosteroids on frequency of RDS when delivery was more than 7 days after therapy found a reduced trend in frequency of RDS in those treated with corticosteroids who delivered more than 7 days later, compared with controls, but results were not statistically significant. We found that frequencies of RDS were not different between women who delivered within 7 days of therapy and those who delivered more than 7 days later, which suggests a continued and potentially undiminished effect of antenatal betamethasone beyond 7 days. Our findings might conflict with those of the meta-analysis, but the population samples differed between the studies in that our sample was limited to deliveries between 28 and 34 weeks' gestation and the pooled studies in the meta-analysis had deliveries at much older gestational ages. Unlike the studies in the meta-analysis, we excluded women with PROM, which might have affected outcomes.
Previous human studies found minimal beneficial effect of antenatal corticosteroid administration if delivery was within 24 hours of initial therapy compared with controls.2,3 We included only women who completed a single course of betamethasone and who delivered more than 24 hours after the initial dose. We also analyzed separately the effect of delivery within 48 hours of completion of a single course to ensure that there was a similar effect throughout the pooled group who delivered less than or equal to 7 days after therapy. We were unable to detect a difference in frequencies of selected outcomes between subgroups and between the primary groups.
We recognize the limitations of the nonrandomized design of our study; however, we tried to limit potential confounding factors by strictly confining our population to a limited gestational age range and by enrolling women without PROM.
1. NIH Consensus Development Panel. Effect of corticosteroids for fetal maturation on perinatal outcomes. JAMA 1995;273:413–7.
2. Crowley PA. Antenatal corticosteroid therapy: A meta-analysis of the randomized trials, 1972–1994. Am J Obstet Gynecol 1995;173:322–35.
3. Liggins GC, Howie RN. A controlled trial of antepartum glucocorticoid treatment for prevention of respiratory distress syndrome in premature infants. Pediatrics 1972;50:515–25.
4. Ballard PL, Ballard RA. Scientific basis and therapeutic regimens for use of antenatal glucocorticoids. Am J Obstet Gynecol 1995;173:254–62.
5. Polk DH, Ikegami M, Jobe AH, Sly P, Kohan R, Newnham J. Preterm lung function after retreatment with antenatal betamethasone in preterm lambs. Am J Obstet Gynecol 1997;176:308–15.
6. Ikegami M, Polk DH, Jobe AH. Minimum interval from fetal betamethasone treatment to postnatal lung responses in preterm lambs. Am J Obstet Gynecol 1996;174:1408–13.
7. Ikegami M, Polk DH, Jobe AH, Newnham J, Sly P, Kohan R, et al. Effect of interval from fetal corticosteroid treatment to delivery on postnatal lung function of preterm lambs. J Appl Physiol 1996;80:591–7.
8. Papile LA, Burnstein J, Burnstein R, Koffler H. Incidence and evolution of subependymal and intraventricular hemorrhage: A study of infants with birth weights less than 1500 grams. J Pediatr 1978;92:9–33.