Can repeated courses of antenatal corticosteroids for threatened preterm birth be justified?

Preterm birth (birth before 37 weeks gestation) remains the leading cause of perinatal mortality in developed countries. Antenatal corticosteroids have been shown to reduce perinatal mortality, specifically by reducing respiratory distress syndrome (RDS) and intraventricular hemorrhage among preterm babies [1]. The optimal effect is achieved when corticosteroids are administered at least 24 h prior to delivery and the beneficial effect persists for at least 7 days [2]. The long-term effects of antenatal corticosteroids have also been addressed, and follow-up of children from a number of randomized, controlled trials for up to 20 years postdelivery have confirmed no apparent adverse effects in relation to growth, cognitive ability or behavior when mothers were administered a single course over a 48-h period [3,4]. The question we wish to address is whether or not repeated courses of antenatal corticosteroids are of benefit or, more importantly, are they of harm?

Historical context

It was Graham (Mont) Liggins, in 1969, who first noted that lambs delivered prematurely after corticosteroid administration had partial lung aeration. This led him to suggest that administration of glucocorticoids may result in accelerated surfactant activity [5]. Further studies confirmed this, and speculation mounted as to the potential benefits to humans. In the years that followed, numerous randomized, controlled trials studied the effects of antenatal corticosteroid administration on pregnant women at risk of preterm delivery [2,6]. Despite the published evidence, the obstetric world was slow to respond and took its time adopting the use of this treatment. It was not until 1994, some 25 years after Liggins’s pioneering discovery and 4 years following Crowley’s seminal systematic review [1], that the NIH Consensus Development Panel on the Effect of Corticosteroids for Fetal Maturation on Perinatal Outcomes recommended widespread use of antenatal corticosteroids for all women at increased risk of preterm labor [7].

Repeated courses of antenatal corticosteroids

The initial delays in implementing the use of single courses of antenatal corticosteroids were not to be repeated when it came to use of multiple courses. On the contrary, the practice of administration of repeated courses of antenatal corticosteroids crept into clinical practice without any evidence confirming benefit or excluding harm. This was driven by the finding that the lung maturational effects appeared to diminish when preterm birth occurred more than 7 days after the initial treatment [6]. If delivery did not occur, and an ongoing threat of preterm delivery continued, an additional benefit of a repeated dose of corticosteroids was perceived. At that time, in excess of 18 individual trials had evaluated the use of a single-course treatment [8]. There were no reports, however, from randomized, controlled trials to inform the practice of repeated courses, nor had extensive animal studies been designed to specifically address this question.

Surveys of practice in both the UK [9] and Australia [10] confirmed widespread variation in the use of antenatal corticosteroids. In the UK, the majority of units were prescribing repeated courses of antenatal corticosteroids, but with virtually no standardization of practice among units. The Australian survey reported that 97% of obstetricians prescribed antenatal corticosteroids, with 83% of respondents using repeated courses; up to 50% of these were prescribing weekly courses. These variations demonstrated a lack of robust evidence and the urgent need for randomized, controlled trials.

Animal studies

Evidence emerged from animal studies describing potential detrimental effects on the fetus of repeated courses of corticosteroids. A systematic review of carefully controlled experiments performed in animals suggested that repeated doses have beneficial effects in terms of lung function but also carry potential adverse effects [11]. Ikegami and colleagues, while demonstrating improved lung function in lambs exposed to repeated courses of antenatal corticosteroids, also revealed a reduction in birthweight [12]. Studies targeting the effects on the sheep fetal CNS revealed reductions in brain weight and delayed myelination of the optic nerve [13]. Sheep were not the only animals targeted; monkeys, rats and rabbits have also been studied. Earlier observational studies in monkeys showed evidence of adrenal suppression [14] and, later, controlled experiments in rabbits confirmed these findings [15]. The authors of numerous animal studies, highlighting growth delay, neurodevelopmental delay and adrenal suppression, recommended extreme caution to clinicians regarding the use of repeated courses of antenatal corticosteroids in human pregnancy.

Observational studies

Most of the initial published evidence addressing single versus repeated courses of antenatal corticosteroids in humans was from retrospective or prospective observational studies. These studies were evaluated within a systematic review in 2001 [16]. Multiple courses of antenatal corticosteroids were associated with a decreased risk of RDS (odds ratio [OR]: 0.79; 95% CI: 0.64–0.98) and patent ductus arteriosus but an increased risk of endometritis. While there was no consistent or adverse effect on other maternal or neonatal outcomes, the studies were limited by the potential effect of confounding variables. A study of antenatal corticosteroids in twin pregnancies reported that prophylactic corticosteroids were not associated with a significant reduction in RDS (adjusted OR: 0.7; 95% CI: 0.2–2.0), and that repeated courses were associated with a reduction in mean birthweight in term babies of 129 g (range: -218 to -33; p = 0.008) [17].

A number of randomized, controlled trials in humans were subsequently established to investigate the safety and efficacy of repeated doses of antenatal corticosteroids. The Royal College of Obstetricians and Gynaecologists (London, UK) produced a clinical guideline in 2004 and recommended that where use of repeated courses of antenatal corticosteroids was being considered, this should occur within the context of recruitment to ongoing, randomized, clinical trials [18].

Randomized, controlled trials

In 2006, Wapner and colleagues published the results of a multicenter, double-blind, randomized, controlled trial to evaluate single versus weekly courses of antenatal corticosteroids [19]. A tendency of reduced birthweight in the repeated corticosteroid group, without any benefit in terms of composite neonatal outcome (8.0% weekly group vs 9.1% placebo group; p = 0.67), resulted in this study being stopped prematurely (495 recruited and 2400 planned). Repeated doses did, however, result in better lung function, particularly among infants delivered before 32 weeks gestation. The following year, a study reporting on the long-term follow-up of the infants demonstrated a higher, albeit not statistically significant, rate of cerebral palsy among children who had been exposed to repeated doses of corticosteroids (2.9 vs 0.5%; p = 0.12) [20]. The authors expressed concern over the findings and recommended further research addressing the safety of repeated doses of antenatal corticosteroids.

The Multiple Courses of Antenatal Corticosteroids for Preterm Birth (MACS) trial, another multicenter, double-blind, randomized, controlled trial, recruited 1858 participants at 25–32 weeks’ gestation and published their results in 2008. The authors reported that multiple courses of antenatal corticosteroids every 14 days did not improve preterm birth outcomes (morbidity and mortality 12.9% for multiple courses versus 12.5% for placebo group) [21]. Reductions in birthweight, length and head circumference were also described in the group that received multiple courses. They concluded that multiple courses of antenatal corticosteroids do not improve preterm birth outcomes and are associated with a decreased weight, length and head circumference at birth. Therefore, the study treatment schedule cannot be recommended.

These findings differed from those of The Australasian Collaborative Trial of Repeat Doses of Steroids (ACTORDS), the Australian trial that enrolled 982 women at less than 32 weeks’ gestation. The findings, published in 2006, demonstrated short-term neonatal benefits of repeated courses of antenatal corticosteroids compared with women receiving placebo (RDS 33 vs 41%; relative risk 0.82; 95% CI: 0.71–0.95; p = 0.01) [22]. The ACTORDS trial, which repeated a single injection of corticosteroid every 7 days rather than every 14 days, followed-up the patients at 2 years of age and concluded that there was no difference in survival free of major neurosensory disability or in body size between the two groups [23].

Another trial, which used a booster course or ‘rescue’ dose before delivery, was discontinued early owing to an increase in RDS and need for surfactant therapy in the rescue-treatment group [24]. The different therapeutic regimens used within the individual clinical trials published to date limits the ability to combine the results within a meta-analysis. The most recent Cochrane systematic review (published prior to the MACS trial) concluded that repeat doses of prenatal corticosteroids reduce the occurrence and severity of neonatal lung disease and the risk of serious health problems in the first few weeks of life [25]. These short-term benefits for babies support the use of repeat doses of prenatal corticosteroids for women at risk of preterm birth. However, these benefits are associated with a reduction in some measures of weight and head circumference at birth, and there is still insufficient evidence on the longer-term benefits and risks.

High-risk groups

The decision to administer a single course of antenatal corticosteroids to a woman presenting high risk of preterm delivery is a relatively straightforward one; the difficulty arises if she remains undelivered after an interval of 7 days but at ongoing risk of preterm birth. This occurs in situations where the woman falls within a high-risk group, such as multiple pregnancy (particularly monochorionic twin pregnancy), women with a history of extreme preterm delivery, women with previous cervical surgery and pregnancies complicated by intrauterine growth restriction. For example, twin pregnancy is an easily recognizable risk factor for preterm delivery, with up to 40% of twins delivered at less than 37 weeks’ gestation [26]. Corticosteroids exert their optimal effect if administered more than 24 h before delivery; however, many twins who would potentially benefit from corticosteroids do not receive optimal treatment, often because of insufficient time to complete a 24-h course prior to delivery. It is not currently possible to detect with complete accuracy which twins or high-risk singletons will deliver preterm; therefore, a prophylactic corticosteroid policy requires careful consideration of the optimal timing of a single course of antenatal corticosteroids. If given too early, the potential benefit may be wasted; if given too late, the potential benefit may be missed.

Conclusion

Undoubtedly, the mixed messages emerging from randomized, controlled trials will continue to cause confusion. Where such confusion exists, variations in practice will persist. The simplest approach is to return to the basic tenets of medicine: we must first ‘do no harm’. A wealth of evidence verifies that a single course of antenatal corticosteroids administered more than 24 h before birth saves lives by enhancing fetal lung maturation in pregnancies at high risk of preterm delivery. However, the potential to produce harm occurs with inappropriate and excessive use of antenatal corticosteroids. The possibility of improved respiratory function must be balanced with a potential adverse effect on fetal growth. A small reduction in head circumference may not be of great clinical significance, but what are the potential effects on brain development? There is no doubt that further, careful, long-term follow-up studies of these patients are required. For now, we conclude that antenatal corticosteroid treatment should be restricted to a single course of treatment administered at the optimal time, as judged by an experienced clinician for the given clinical circumstances, and that repeated courses of corticosteroids cannot be justified.

Financial & competing interests disclosure

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

No writing assistance was utilized in the production of this manuscript.