A landmark meta-analysis published in BJOG (Crowley et al. 1990;97:11–25) underpins the recommendation that women presenting in threatened preterm labour (TPTL) should receive antenatal corticosteroids (ACS) (Preterm labour and birth, Guideline 25; NICE 2015). This simple intervention has become an icon of evidence-based practice, with the forest plot of the initial trials depicted in the Cochrane Collaboration logo (Roberts D, Dalziel SR. Cochrane Database of Systematic Reviews 2006;3:CD004454). Graham Collingwood Liggins, a senior lecturer in Obstetrics and Gynaecology at the National Women's Hospital in Auckland, New Zealand, is credited for discovering that corticosteroids cross the placenta and accelerate fetal lung maturation. In the 1960s, experimental observations in lambs sparked the groundbreaking randomised placebo-controlled trial in humans (Liggins GC, Howie R. Paediatrics 1972;50:515–25). This trial provided the first evidence that betamethasone reduces the frequency of respiratory distress syndrome (RDS) in infants born <32 weeks of gestation, and secures a five-fold reduction in mortality; however, it was not until Crowley's meta-analysis was published in this journal almost 20 years later that a clear change in clinical practice was appreciable. In addition to reducing RDS, a significant reduction in periventricular haemorrhage and was confirmed (see Figure 1, reproduced from the original publication). By 1999, the administration of ACS to women in TPTL was practiced almost universally. However that the symptoms of TPTL are poor predictors of delivery (over 80% remain undelivered 7 days after presentation), the majority of infants are exposed to steroids unnecessarily. Indeed, 98% of those not delivered within a week were offered a further treatment course (Brocklehurst et al. BJOG 1999;106:979–9). Although the health and economic benefits of appropriate ACS are proven, if treated outside the 7-day window, even with a single course, infants demonstrate lower birthweight (mean difference −147 g, 95% CI −291 to −2 g), head circumference, and length. This is compounded by the potential need for repeat courses, with similar further insult. More concerning still, the 2006 Cochrane Review described a worrying trend towards an increased risk of death for babies who go on to deliver at full term (relative risk 3.25; 95% CI 0.99–10.66). Other human consequences of ACS during this critical developmental period are less certain, but animal models imply adverse neurodevelopmental outcomes and decreased fetal brain growth. Concerns have been raised in emerging long-term follow-up studies regarding hyperactivity in childhood and impact on glucose metabolism, an implied predictor of cardiovascular disease (Kelly et al. Pediatrics 2012;129:1282–90). Although the ubiquitous implementation of ACS is a triumph of evidence-based medicine, it requires continual reappraisal. Accurate prediction of delivery within the desired therapeutic window may help. The benefit of accurate prediction models and thus timely administration is further reinforced by new evidence suggesting important outcomes (death and intraventricular haemorrhage) may be improved if delivery occurs <48 hours after treatment (WHO. Recommendations on Interventions to Improve Preterm Birth Outcomes. 2015). Although simplifying the management quandary, a treat-all strategy (NICE 2015) may expose this vulnerable population to unnecessary risk. To optimise care we must re-evaluate evidence as long-term studies and prediction tools become more sophisticated. Full disclosure of interests available to view online as supporting information. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.
Read full abstract