Article Text
Abstract
Background: Weekly repeated antenatal corticosteroid treatment improves respiratory outcome but decreases fetal growth and may impair neurodevelopmental outcome. We have previously reported that a single repeat betamethasone (BM) dose neither decreased fetal growth nor improved the outcome of preterm infants during the first hospitalisation.
Objective: To study prospectively whether a single repeat dose of BM influences neurodevelopment and growth within 2 years.
Design: Women with imminent delivery before 34.0 gestational weeks were eligible if they remained undelivered for >7 days after a single course of antenatal BM. After stratification, a single repeat dose of BM (12 mg) or placebo was given. The children underwent neurological and psychometric examinations and a speech evaluation at a corrected age of 2 years.
Setting: Prospective, blinded evaluation following the randomised multicentre trial.
Patients: 259 (82%) surviving infants completed the 2-year follow-up, 120 in the BM group and 139 in the placebo group.
Results: The rate of survival without severe neurodevelopmental impairment was similar in both groups (BM 98%, placebo 99%). The risk of cerebral palsy (BM 2%, placebo 1%), growth or re-hospitalisation rates (BM 60%, placebo 50%) did not differ between the groups.
Conclusions: A single repeat dose of antenatal BM tended not to influence physical growth or neurodevelopment at 2 years of age.
Statistics from Altmetric.com
A single course of antenatal corticosteroid given to pregnant women before preterm birth reduces mortality, risk of respiratory distress syndrome (RDS) and intraventricular haemorrhage (IVH).1 2 The treatment is most beneficial in reducing the incidence of RDS among preterm infants of mean (SD) gestational age between 26 and 34 (6) weeks.2 The timing of administration influences the outcome, showing decreased risk of RDS when the steroid is given 24 h to 7 days before birth, whereas <24 h or >7 days of exposure does not influence the risk.1 2
The use of weekly repeated corticosteroids became common in the late 1990s for the treatment of pregnant women who remained undelivered after a single corticosteroid course. According to non-randomised studies multiple courses or a single repeat dose of antenatal corticosteroid may reduce the risk of RDS, but additionally may increase the risk of intrauterine growth retardation.3 4 5 Cohort studies have shown that multiple courses may be associated with an increased risk of cerebral palsy (CP) and behavioural disturbances.6 7 Recent randomised clinical trials have shown a beneficial effect of a weekly repeated dose of corticosteroid on pulmonary morbidity,8 9 10 but increased risk of fetal growth retardation.9 10 According to a retrospective cohort study, a single rescue dose of antenatal betamethasone (BM) at 28–34 weeks of gestation decreased the risk of RDS.5
We conducted a randomised trial involving pregnancies in imminent preterm birth. The aim was to study whether a single antenatal repeat dose of BM that was given at least 7 days after the first open course of antenatal glucocorticoid, increases the survival of children without acute neonatal morbidity.11 The preterm infants exposed to the repeat dose of BM had an increased need for surfactant therapy if they were born within <24 h of antenatal BM treatment, whereas the risk of RDS tended to decrease when they were born later.11 The fetal growth was unaffected and no significant difference was seen in either neonatal morbidity or mortality.
Antenatal steroid use has been associated with a decrease in intrauterine growth and with abnormal neurodevelopment. These adverse effects may depend on the dosage and duration of the drug. Our aim was to determine whether the single repeat dose of BM influences later neurodevelopment, that is, neuromotor, neurosensory and cognitive development, as well as respiratory morbidity and growth during the first 2 years of life.
Methods
Description of the original trial
The children were born to mothers who had participated in a randomised, placebo-controlled trial employing a single repeat dose of antenatal BM for imminent preterm birth. The primary outcome was survival of infants without RDS or severe IVH. Pregnant women with <34.0 weeks of gestation were eligible if they had an imminent delivery and had received the standard course of antenatal BM (12 mg twice at an interval of 24 h) <7 days before. Details of the trial have been reported previously.11 The protocol was approved by the ethics committee of Oulu University Hospital and by the National Agency for Medicines. Written informed consent was obtained from the parents.
The women were randomly assigned to receive either a single repeat dose of 12 mg BM given intramuscularly or an apparently identical placebo. The patients were stratified according to gestational age (<28 weeks +0 days or between 28 weeks +0 days and 34 weeks +0 days). Both gestation groups were additionally stratified on the basis of the number of fetuses (singleton or multiple pregnancies).
According to the interim analysis the single repeat dose of BM was associated with decreased intact survival during the first hospitalisation, as the infants born within 24 h of the intervention had a high incidence of RDS. Recruitment was terminated primarily on the grounds of safety concerns after 249 mothers had been enrolled.
The follow-up study
Altogether, 259 of the 315 surviving infants (82%) at five centres were enrolled for the follow-up study at 2 years of age counted from a postmenstrual age of 40 weeks. A paediatrician or paediatric neurologist was assigned for the follow-up at each of the participating centres. The examiners and families were unaware of treatment-group assignment. A questionnaire including information on respiratory problems, infections and medical history during the first 2 years of life was filled out during the examination at two centres (Oulu and Helsinki) and by telephone at the other centres. The socioeconomic status of the family was evaluated according to the information on maternal and paternal education, occupation and marital status. The child’s weight, length and head circumference were recorded, and the children underwent a standardised age-specific neurological examination conducted by the paediatric neurologist or paediatrician. CP was defined as described.12 Psychometric evaluations were performed by a neuropsychologist at the same visit, using the Bayley Scales of Infant Development II (BSID-II).13 Overall development was evaluated by a paediatrician (OP) or a paediatric neurologist (AL) using Griffiths Developmental Score.14 Speech was evaluated by a speech therapist (AY) or a paediatric neurologist (AL). The number of words in the child’s vocabulary was used to evaluate speech development, which was divided into three categories: normal (>25 percentiles), mildly abnormal (10–25 percentiles) or severely abnormal (<10 percentiles). Survival without severe neurological, cognitive, or sensory impairment (severe neurodevelopmental impairment (NDI), ie, survival without CP, Mental Developmental Index (MDI) <70, developmental quotient (DQ) <70, deafness, or blindness) was evaluated for all the children who completed follow-up.
Statistical analysis
The first power analysis was based on the primary outcome.11 Altogether, 315 of 326 infants survived during the first hospitalisation and they were eligible for the prospective follow-up study. Present follow-up trial with 259 participating children has an 80% power to detect a significant difference at an alpha level of 0.05 in the survival without NDI from 99% to 91%.
The baseline data were compared using the unpaired t test for continuous variables and the χ2 test for categorical data. In the outcome analysis, odds ratios (ORs) were calculated with 95% confidence intervals and differences were also compared using χ2 tests for categorical data. Single continuous variables were compared by means of unpaired t tests between the two groups, and repeated measurements were analysed using repeated measures analysis of variance. The Mann–Whitney U test was used when the data were not normally distributed. The statistical analyses were performed using SPSS V.16.0 for Windows (SPSS Inc, Chicago, Illinois, USA).
Results
The study design is illustrated in fig 1. Two children in the BM group and one in the placebo group died after the first hospitalisation. One hundred and twenty children in the BM group and 139 in the control group completed the follow-up, with 179 (75%) altogether undergoing Bayley testing, 178 (69%) Griffiths testing, and 228 (88%) the comprehensive speech evaluation. Fifty-six children (19% BM, 14% placebo) were lost to follow-up because of parental refusal or moving to an unknown address. The group of infants lost to follow-up tended to have a longer gestation time at birth (mean (SD) 31.9 (2.1) weeks) and lower incidences of RDS (38%), bronchopulmonary dysplasia (5%), severe IVH (0%) and periventricular leukomalacia (PVL) (13%) during primary hospitalisation than those examined, as shown in table 1. Maternal age was higher in the BM group (p = 0.005), while the proportion of multiple pregnancies tended to be higher in the placebo group (p = 0.08). All children were born before 36 weeks of gestation, and 33% before 30 weeks of gestation, but only 16% before 28 weeks of gestation.
The mean (SD) age at follow-up was 24.2 (0.9) months in both groups (table 2). Weight, height and head circumference were similar in both groups (table 2).
The rate of survival without severe NDI was 98% in the BM group and 99% in the placebo group (OR 0.28 with 95% CI 0.03 to 2.71, table 3). We performed subgroup analysis based on gestational age and length of intrauterine study drug exposure (interval between drug injection and birth). Among children born before 30 weeks of gestation, survival without severe NDI was 98% in both groups (p = 1.00) and among children born at 30 weeks or later, 97% in the BM group and 100% in the placebo group (p = 0.19). When the exposure was <24 h, survival without severe NDI was 98% in the BM group and 99% in the placebo group (p = 0.59). The corresponding figures were 96% (BM) and 100% (placebo) when the exposure was ⩾24 h (p = 1.00).
Two children in the BM group and one in the placebo group had CP (OR 2.38; 95% CI 0.21 to 26.57). One child in the BM group had monoplegia and another had hemiplegia. The boy with monoplegia was born after a pregnancy complicated with premature rupture of membranes and chorioamnionitis. He developed RDS, sepsis, grade 2 IVH and PVL. The boy with hemiplegia had severe RDS and grade 2 IVH. Both of them were born at >30 gestational weeks without any signs of birth asphyxia. One child born at 27 weeks of gestation in the placebo group was diagnosed as having hemiplegia and hydrocephalus. She had birth asphyxia, severe RDS and grade 4 IVH during the neonatal period. Two children in the BM group had strabismus versus five children in the placebo group (OR 0.46; 95% CI 0.09 to 2.41). None of the children required a hearing aid. The mean (SD) DQ, evaluated in terms of the Griffiths Developmental Scale, was 97 (12) in the BM group and 95 (9) in the placebo group (p = 0.11, table 3). The mean (SD) MDI was 104 (13) in the BM group and 101 (13) in the placebo group (p = 0.18, table 3). Speech development was severely abnormal in 15% of the children in the BM group versus 17% in the placebo group (OR 1.23; 95% CI 0.62 to 2.41).
The re-hospitalisation rate during the first 2 years of life was 59% in the BM group and 50% in the placebo group (p = 0.16). Altogether, 33% of children in the BM group required more than one re-hospitalisation versus 31% in the placebo group (p = 0.78). The most common indications were tympanostomy (altogether 43 children) and recurrent wheezing or asthma (31 children). Twenty-six children in the BM group underwent tympanostomy for recurrent otitis media, compared with 17 children in the placebo group. Severe lung disease (ie, pneumonia, recurrent wheezing or asthma) was diagnosed in 39% of the children in the BM group and 32% in the placebo group (p = 0.34). Ten children in each group had pneumonia.
Discussion
In the present follow-up study a single repeat dose of antenatal BM had no detectable influence on survival without severe neurodevelopmental disability or on neurological or cognitive development, nor did it influence physical growth, re-hospitalisation rate or the risk of severe infections during the first 2 years. Our results extend the two recently published follow-up studies of randomised trials on weekly repeated BM.15 16
In this study, all the infants in the BM group were exposed to a single repeat dose of BM and were born before term. The ACTORDS trial used a weekly repeated single dose (BM 11.4 mg), 58% of the fetuses were re-exposed at least twice and altogether 18% of the original study group were born at term.9 The US trial used a weekly repeated single course including two doses of BM (12 mg × 2); 63% of the mothers were given ⩾4 courses and 36% of the original study group were born at term.10 A recently published MACS trial used multiple courses of antenatal BM (12 mg × 2); repeated every 14 days, 60% of fetuses were exposed at least twice and 32% of the infants were born at term.17 According to a recently published randomised trial of one rescue course of antenatal corticosteroid, 20% of study infants were born after 36 weeks of gestation.18 The difficulty of optimal timing of weekly repeated antenatal corticosteroids has been acknowledged, as many of the infants have been exposed to unnecessary courses of corticosteroids after the critical prenatal period, that is, near term of the pregnancy.
In the present study, a single repeat dose of BM had no effect on weight, height and head circumference at birth or at 2 years of age. Similarly, in Garite’s study, a single rescue course of corticosteroid did not influence intrauterine growth.18 In Crowther’s study, repeat doses of corticosteroid were associated with lower weight and reduced head circumference (z scores) at birth,10 but not later.15 In Wapner’s study, a reduction in weight and length was seen particularly in infants exposed to ⩾4 courses of antenatal BM, but at 2 years of age, growth was comparable between the groups.16 However, the proportion of children with anthropometric measurements below the 10th percentile tended to be higher in the BM group.16 In the MACS trial, weight, length and head circumference at birth were significantly affected by repeated BM. The present findings suggest that the effect of antenatal corticosteroids on fetal growth may be dose-dependent as seen in animal studies.19 Normal length or weight at the age of 2 years does not exclude the possibility that an abnormal pattern of growth and weight gain in later life may be acquired from the antenatal period. Intrauterine growth retardation has an established association with an increased risk of excessive weight gain, metabolic disease and increased morbidity from cardiovascular disease.20 21
In the present study neurosensory development and the rate of CP were comparable between the treatment and placebo groups. However, there are limitations in the present study. Our study population is too small to evaluate reliably the long-term outcome. Early interruption of the study results in inadequate power with a risk of type II error. In addition, our initial sample size calculation was based on neonatal outcome and did not include neurodevelopmental outcome, which further complicates the evaluation of long-term results. In the ACTORDS trial, the neurodevelopmental outcome was similar,15 although children in the repeat BM group had slightly more attention problems than in the placebo group.15 Previously we did not find any detectable difference in the temperament profiles between single repeat BM and placebo groups.22 In the US trial, children in the BM and placebo groups had similar Bayley scores. Repeated BM tended to increase the risk of CP in the subgroup of children who were exposed to four or five BM courses. The fact that five of the six children with CP were born later than 34 weeks of gestation and had normal cranial ultrasound findings during the neonatal period arouses further concern about adverse neurological development following multiple courses of antenatal corticosteroids.16
We conclude that a single repeat dose of antenatal BM had no detectable effect on growth or neurodevelopmental outcome at the age of 2 years. One or two repeat doses of antenatal BM appear to be beneficial in reducing acute pulmonary morbidity, when the premature birth can be delayed at least 24 h after the steroid. More follow-up studies are needed before a recommendation is feasible.
What is already known on this topic
Weekly repeated antenatal corticosteroid treatment improves pulmonary outcome, but decreases fetal growth and may have harmful effects on neurodevelopment.
One to two repeat doses of antenatal corticosteroid given 24 h to 7 days before birth may improve the pulmonary outcome without affecting intrauterine growth.
What this study adds
A single repeated dose of antenatal betamethasone (BM) given for imminent preterm birth at least 1 week after standard BM administration did not influence neurodevelopmental outcome and had no effect on growth at the age of 2 years.
REFERENCES
Footnotes
↵*OMP, P Olsen, T Saarela, MH (Department of Paediatrics, Oulu University Hospital, Oulu, Finland); MAK, AL, RP, H Heiskala, S Andersson (Children’s Hospital, Helsinki University Central Hospital, Helsinki, Finland); K Nikolajev (Department of Paediatrics, Kuopio University Hospital, Kuopio, Finland); OT (Department of Paediatrics, University Hospital of Tampere, Tampere, Finland); LL (Department of Paediatrics, University Hospital of Turku, Turku, Finland).
Funding Supported by grants from the Foundation for Paediatric Research, the Alma and K.A. Snellman Foundation (Oulu, Finland), the Arvo and Lea Ylppö Foundation and the Sigrid Juselius Foundation (Finland).
Competing interests None.
Ethics approval The protocol was approved by the ethics committee of Oulu University Hospital and by the National Agency for Medicines.
Patient consent Parental consent obtained.
Provenance and Peer review Not commissioned; externally peer reviewed.
Linked Articles
- Fantoms