ARTICLE |
a Departments of Pediatrics
g Obstetrics and Gynecology, University of Oulu, Oulu, Finland
b Hospital for Children and Adolescents, Helsinki University Central Hospital, Helsinki, Finland
c Department of Pediatrics, University of Tampere, Tampere, Finland
d Department of Pediatrics, University of Turku, Turku, Finland
e Department of Pediatrics, Seinäjoki Central Hospital, Seinäjoki, Finland
f Department of Obstetrics and Gynecology, University of Helsinki, Helsinki, Finland
| ABSTRACT |
|---|
|
|
|---|
METHODS. Women with imminent delivery before 34.0 gestational weeks were eligible if they remained without delivery for >7 days after a single course of betamethasone. After stratification, a single repeat dose of betamethasone (12 mg) or placebo was administered. The primary outcome was survival without respiratory distress syndrome or severe intraventricular hemorrhage (grade 3 or 4).
RESULTS. A total of 249 mothers had been enrolled by the time the study was discontinued. All of the 159 infants in the betamethasone group and 167 in the placebo group were born before 36 weeks of gestation. The intact survival rate was unaffected and was lower than anticipated, because the gestational age-adjusted incidence of respiratory distress syndrome was higher than the population incidence. The requirement for surfactant therapy in respiratory distress syndrome was increased in the betamethasone group. According to posthoc analysis of the data for 206 infants who were delivered within 1 to 24 hours, the betamethasone booster tended to increase the risk of respiratory distress syndrome and to decrease intact survival rates.
CONCLUSIONS. According to this study, a single booster dose of betamethasone just before preterm birth may perturb respiratory adaptation. These results caution against uncontrolled use of a repeat dose of glucocorticoid in high-risk pregnancies.
Key Words: prenatal corticosteroid prenatal glucocorticoid respiratory distress syndrome premature infant premature birth intraventricular hemorrhage
Abbreviations: RDSrespiratory distress syndrome IVHintraventricular hemorrhage CIconfidence interval ORodds ratio
The benefits of a single course of prenatal corticosteroid treatment in decreasing the morbidity of preterm infants have been evident since the 1970s, when Liggins and Howie1 published the first controlled trial of antepartum glucocorticoid treatment. According to a meta-analysis, a single course of prenatal corticosteroid treatment improves the survival rates of preterm infants born at 24 to 34 weeks of gestation and decreases the incidence of respiratory distress syndrome (RDS) and intraventricular hemorrhage (IVH),2 which demonstrates that the treatment is indicated.3 Reliable evidence exists for neonatal benefits from a complete course, starting at 24 hours and lasting for up to 7 days after treatment.1,2 Prenatal glucocorticoid treatment is also effective after preterm rupture of fetal membranes.2 According to experimental studies and clinical observations, the glucocorticoid effect may manifest even before 24 hours46 and last >7 days.711
A single course of glucocorticoid treatment did not decrease the risk of RDS or IVH in extremely preterm infants.2,12,13 However, those studies might not have had adequate power. In an attempt to improve the response, the practice of repeating the course of corticosteroid treatment in threatened preterm delivery has become common, although the efficacy and safety of repeated courses have not been confirmed. According to previous randomized studies, weekly courses of prenatal corticosteroid treatment neither reduced neonatal morbidity nor had any beneficial effect on functional residual capacity or pulmonary compliance in preterm infants.14,15 In contrast, 1 additional dose of prenatal betamethasone treatment (12 mg) after a single course of steroid treatment decreased the risk of RDS, according to an earlier retrospective cohort study.16
We investigated whether the therapeutic effect of a single course of prenatal glucocorticoid treatment could be improved in a high-risk population by avoiding unnecessary treatment that is potentially harmful.9,10,17,18 Accordingly, the aim of the present randomized, blinded, placebo-controlled trial was to investigate whether a single additional dose of betamethasone given in imminent preterm birth before 34 weeks of pregnancy,
7 days after a full treatment course of betamethasone, would improve the proportion of infants surviving without RDS or severe IVH.
| METHODS |
|---|
|
|
|---|
7 days before trial entry, when preterm birth before 34.0 weeks of pregnancy was imminent. Specific qualifying criteria for the second course of corticosteroid/placebo included elective delivery within 48 hours, as indicated on the basis of the clinical status of the mother and/or the fetus, or a very high risk of spontaneous delivery within 48 hours, as indicated by cervical opening of
3 cm and regular contractions at 5- to 10-minute intervals. Exclusion criteria were maternal long-term systemic corticosteroid therapy, clinical chorioamnionitis (maternal fever, uterine tenderness, foul-smelling amniotic fluid, and leukocytosis), or lethal disease of the fetus. Prolonged rupture of fetal membranes without clinical chorioamnionitis was not an exclusion criterion. Gestational age was calculated from the mother's last menstrual period and was confirmed with ultrasonography before 20 weeks of gestation.
|
Outcomes
The primary outcome was survival without RDS or severe IVH during the first hospitalization (ie, intact survival). RDS was defined on the basis of typical chest radiograph findings, requirement for continuous distending airway pressure and supplemental oxygen for
48 hours, or requirement for surfactant therapy in cases of established respiratory failure. The administration of surfactant in the delivery room was recorded but was not included in the diagnosis of RDS. Severe IVH was defined as IVH with ventricular dilation (grade 3) or parenchymal hemorrhage (grade 4).19 Cranial ultrasonography was performed for all infants at least at the age of 4 to 8 days and at postmenstrual age of 36 weeks or before discharge. The most severe grade of IVH was recorded.
Secondary outcomes included cystic periventricular leukomalacia20 and necrotizing enterocolitis of grade
2.21 Bronchopulmonary dysplasia was defined as a requirement for supplemental oxygen or any form of ventilation with continuous distending pressures at postmenstrual age of 36 weeks or at postnatal age of 4 weeks for those born after postmenstrual age of 31 weeks. Patent ductus arteriosus was defined as a requirement for prostaglandin inhibitor therapy or surgery for closure. Additional data were collected prospectively throughout the hospital course.
Statistical Analyses
Sample size analysis was based on Finnish perinatal morbidity and mortality statistics (from the National Research and Development Centre for Welfare and Health: www.stakes.info/2/index.asp). The baseline rate of survival without RDS or severe IVH was 50%. The present trial was designed to detect an improvement in the intact survival rate from 50% to 62.5%, which would have required a sample size of 220 women in each arm to reject the null hypothesis with
of .05 and power of 80% (2-sided). The data safety monitoring committee had planned to analyze the adverse outcomes by the time when one half of the mothers had been recruited.
The incidence of RDS in the study population was compared with the national incidence figures for preterm infants from the Finnish Medical Birth Register for the years 1996 to 2000. During that period, generally surfactant was not administered in the delivery room.
In outcome analyses, we calculated the odds ratios (ORs) with 95% confidence intervals (95% CIs) and also compared the differences between the treatment and placebo groups with
2 tests for categorical data. The single continuous variables were compared between the 2 groups with unpaired t tests, and the repeated measurements were analyzed with repeated-measures analysis of variance. The Mann-Whitney U test was used when the data were not distributed normally. Statistical analyses were performed with SPSS 12.0.1 for Windows (SPSS, Chicago, IL).
| RESULTS |
|---|
|
|
|---|
30 weeks of pregnancy. The safety committee recommended performing an efficacy analysis and dividing the population into the following strata: gestational age at birth (<30 weeks and
30 weeks) and interval from treatment to delivery (<24 hours and
24 hours). According to interim analysis, the short exposure to betamethasone was associated with a decrease in the intact survival rate. The short exposure to betamethasone was also associated with the decreased intact survival rate among the more-mature infants. In addition, recruitment was slower than anticipated, and increasing numbers of infants received surfactant treatment in the delivery room. There were additional concerns about the long-term adverse effects of the glucocorticoid. As proposed by the safety committee, the steering committee decided to terminate recruitment prematurely, primarily on the basis of the safety concerns. A total of 249 mothers were enrolled between May 2001 and March 2005, 125 in the betamethasone group and 124 in the placebo group. A total of 159 infants were born in the betamethasone group and 167 in the placebo group. Within the same period, 2591 infants with gestational age of <34 weeks were born in these 8 centers. In Oulu University Central Hospital, with 14% of all preterm births, 48 mothers participated, accounting for 35% of the 138 mothers who delivered at <34 weeks of gestation and >7 days after the initial betamethasone course.
The maternal and pregnancy characteristics of the study groups, including the length of gestation at the intervention and the time from intervention to birth, are shown in Table 1. With the exception of maternal age, the populations seemed similar. The infants in the betamethasone group were delivered at a mean gestational age of 30.7 weeks, compared with 31.0 weeks in the placebo group. All mothers gave birth before 36 weeks of gestation. The birth characteristics of the infants were similar in the betamethasone and placebo groups (Table 2).
|
|
28 weeks, the intact survival rates were 55% in the betamethasone group and 62% in the placebo group. For those born 1 to 24 hours after prenatal betamethasone or placebo treatment, the intact survival rates were 41% and 55%, respectively. For infants born
24 hours after the intervention, the intact survival rates after betamethasone and placebo treatment were 61% and 46%, respectively. The betamethasone- and placebo-treated subjects born 1 to 24 hours or
24 hours after the intervention had otherwise similar pregnancy events and infant characteristics at birth, compared with the rest of the study population (Table 3 and data not shown).
|
|
|
A posthoc analysis of the morbidity of the infants delivered within 1 to 24 hours and of the infants born
24 hours after the intervention was performed. In the subgroup of 206 infants born 1 to 24 hours after the intervention, betamethasone tended to increase the risk of RDS and increased the surfactant requirements in RDS. In the subgroup of 75 infants born
24 hours after the intervention, betamethasone had no detectable effect on the risk of RDS (betamethasone group: 40%; placebo group: 54%) (Table 6).
|
| DISCUSSION |
|---|
|
|
|---|
According to previous studies, a <24-hour interval between the first dose of glucocorticoid and delivery has been associated with a trend toward decreased morbidity rates, because the drug may protect against cerebral complications.1,5,6,22 At the onset, we proposed that the beneficial effect of the single repeat betamethasone dose would be detectable at least as soon as after the first dose and certainly within 48 hours. The aim was to treat women who would very likely deliver prematurely. This might have contributed to the shortening of the prenatal exposure period, so that a total of 79% of all deliveries took place within 24 hours after the intervention.
Posthoc analysis of the present data revealed that the infants who were born 1 to 24 hours after the repeat dose of betamethasone had a lower intact survival rate and a trend toward a higher incidence of RDS than those born shortly after placebo treatment. This finding corresponds to the previous result showing a higher risk of RDS in the betamethasone group among preterm infants who received the treatment <12 hours before delivery23 but not to results from another study.22 In contrast, among infants born
24 hours after the intervention in the present study, betamethasone tended to increase the intact survival rate (OR: 1.80; 95% CI: 0.724.51) and to decrease the risk of RDS (OR: 0.55; 95% CI: 0.221.39), consistent with experimental evidence.24 These results were not significant, and the number of cases was clearly inadequate. The interval between the dose of betamethasone and delivery is critical, and possibly the previous prenatal course of corticosteroid influenced the outcome.
Glucocorticoids alter the pathways regulating inflammatory gene transcription25,26 and influence the hypothalamus-pituitary-adrenal axis.2729 Fetal serum cortisol levels begin to decrease at 6 hours and are lowest 10 to 15 hours after a single maternal injection of 12 mg of betamethasone.30 The suppressed cortisol levels begin to recover by 24 hours after treatment, and suppression disappears by 7 days after prenatal betamethasone treatment.27,30,31 One hour after administration of betamethasone to the mother, the drug was detectable in cord serum. The levels of betamethasone peaked at 6 hours. Levels decreased thereafter and became barely detectable 20 hours after a single maternal dose of betamethasone.31 Preterm infants are able to respond to stress (ie, asphyxia and RDS), and serum cortisol levels increase shortly after preterm birth.32,33 However, corticosteroid given shortly before birth may transiently suppress the hypothalamus-pituitary-adrenal axis during early adaptation.30,34
Corticosteroids have important antiinflammatory and cardiopulmonary effects during early adaptation.33,35 However, betamethasone given 1 week before preterm birth augmented the endotoxin-induced inflammation in newborn ovine lung.26 The endotoxin-induced inflammatory response was strikingly increased in fetal monocytes isolated 7 days after a single dose of maternal corticosteroid treatment.36 In the present study, the first dose of glucocorticoid given >1 week before the booster dose of betamethasone might have increased the inflammatory responsiveness. It is also possible that the betamethasone booster shortly before the birth inhibited the antiinflammatory endocrine response that takes place shortly after birth.32,33 Endotoxins and proinflammatory cytokines, which are common during the high-risk perinatal transition, have interactive effects with corticosteroids on differentiation of noninflammatory cells.37 Glucocorticoid decreases the inhibition of surfactant protein B expression that is caused by proinflammatory cytokines.38 Surfactant protein B and other surfactant components are deficient in RDS, and the lack of surfactant protein B has been shown to result in respiratory failure.37
In contrast, suppression of the endotoxin-induced inflammatory response by fetal monocytes is evident at least 12 to 48 hours after the first prenatal dose of betamethasone.36 This acute inhibition of the proinflammatory cytokine response may promote neonatal hemodynamic adaptation, protecting against cerebral complications after very preterm birth.22
Very recently, a randomized trial of weekly repeated prenatal betamethasone treatment (12 mg) given to women at risk of preterm birth was reported.39 In that study, betamethasone was associated with a decrease in the incidence of RDS (33% after repeated betamethasone treatment and 41% after placebo) and a decrease in the requirement for exogenous surfactant, whereas there were no detectable differences in other neonatal morbidity rates or in the length of initial hospitalization.39 The causes of the different outcome results between these 2 somewhat similar trials remain speculative. Gestational ages at birth were not much different (mean gestational age at birth of 32.4 weeks, compared with 30.8 weeks in our study). The length of fetal exposure after the last dose of betamethasone was not given.39 However, it was considerably longer in the study by Crowther et al39 than in the present study, despite a lower gestational age allowance for betamethasone (<32 weeks, compared with <34 weeks in the present study). In the study by Crowther et al,39 the gestational age at birth was
34 weeks in 36% of all cases (ie, in those cases, the fetal exposure to the last intervention exceeded 2 weeks). A total of 18% of the infants were born at term, whereas in the present study none was born at term and only 23% of the fetuses were exposed to the intervention for
24 hours. It is quite likely that, in the study by Crowther et al,39 preterm birth was not often quite imminent when the repeat dose of betamethasone was given. This might have been a successful treatment strategy for prevention of RDS. Interestingly, in the present study, the few fetuses exposed to betamethasone for >24 hours tended to have an increased intact survival rate (Table 3). Other factors, such as the length of gestation and the number of repeat doses of steroid, might also influence the responsiveness.
According to the present study, a single repeat dose of prenatal betamethasone treatment >1 week after the first course would be feasible for 35% to 40% of all pregnancies terminating before 34 weeks of gestation. However, the current treatment strategy failed to improve the outcomes. The contrasting results regarding the outcomes after the single first dose22 and the single repeat dose of prenatal glucocorticoid treatment may be the result of delayed programming and activation of endotoxin responsiveness.36 Fetal stress attributable to repeat doses of betamethasone administered to the mother may have additional, adverse, long-term consequences, programming the growth pattern or increasing the risk of metabolic syndrome and the risk of abnormal neurologic outcomes.40,41 The present data are consistent with the proposal that repeating the dose of betamethasone is not indicated when preterm birth is quite imminent >1 week after the single dose of prenatal betamethasone prophylaxis.14,15 In case new prenatal treatment strategies for improving the intact survival rates of premature infants are planned, both the length of the intervention and the target population remain to be considered.
| ACKNOWLEDGMENTS |
|---|
The Antenatal Betamethasone Study Group was as follows: T. Raudaskoski, J. Räsänen, P. Jouppila, O. M. Peltoniemi, T. Saarela, and M. Hallman (Oulu University Hospital, Oulu, Finland); E. Halmesmäki, V. Stefanovic, M. A. Kari, and S. Andersson (Helsinki University Central Hospital, Helsinki, Finland); S. Heinonen, K. Heinonen, and K. Nikolajev (Kuopio University Hospital, Kuopio, Finland); J. Uotila and O. Tammela (University Hospital of Tampere, Tampere, Finland); U. Ekblad and L. Lehtonen (University Hospital of Turku, Turku, Finland); T. Kiviniemi and R. Lounamaa (Central Hospital of Central Finland, Jyväskylä, Finland); M. Heikkilä and R. Marttila (Seinäjoki Central Hospital, Seinäjoki, Finland); and E. Koistinen and M. Dalla Valle (Central Hospital of Northern Karelia, Joensuu, Finland).
| FOOTNOTES |
|---|
Address correspondence to Mikko Hallman, MD, PhD, Department of Pediatrics, Biocenter Oulu, University of Oulu, PO Box 5000, FIN-90014, Oulu, Finland. E-mail: mhallman{at}cc.oulu.fi
The authors have indicated they have no financial relationships relevant to this article to disclose.
| REFERENCES |
|---|
|
|
|---|
1000 grams at birth after preterm rupture of the amnion.
Am J Obstet Gynecol. 1999;180
:677
682[CrossRef][ISI][Medline]This article has been cited by other articles:
![]() |
C. McEvoy, D. Schilling, P. Spitale, D. Peters, J. O'Malley, and M. Durand Decreased Respiratory Compliance in Infants Less Than or Equal to 32 Weeks' Gestation, Delivered More Than 7 Days After Antenatal Steroid Therapy Pediatrics, May 1, 2008; 121(5): e1032 - e1038. [Abstract] [Full Text] [PDF] |
||||
![]() |
C. A. Crowther, L. W. Doyle, R. R. Haslam, J. E. Hiller, J. E. Harding, J. S. Robinson, and the ACTORDS Study Group Outcomes at 2 Years of Age after Repeat Doses of Antenatal Corticosteroids N. Engl. J. Med., September 20, 2007; 357(12): 1179 - 1189. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||