Antenatal Corticosteroids for Bronchopulmonary Dysplasia—Mediation Analysis and Causal Reasoning

Antenatal corticosteroids (ACS), administered to pregnant women at risk of preterm birth, represent one of the most successful stories in modern medicine.They have also deservedly inspired the logo of the Cochrane Collaboration, which illustrates the forest plot of a systematic review on this topic.While every doctor knows the importance of ACS, it is nonetheless useful to remind that preterm neonates whose mothers receive ACS have a significantly lower risk not only of death but also of respiratory distress syndrome (RDS) and mechanical ventilation, intracranial hemorrhage, necrotizing enterocolitis, and other conditions, compared with neonates whose mothers do not receive ACS. 1 Although ACS reduce RDS and need for ventilator support, 1 both of which are known risk factors for bronchopulmonary dysplasia (BPD), 2 their effects on BPD have not been proven in randomized clinical trials (RCTs).A Cochrane review published in 2020 1 cites only 5 RCTs assessing BPD as an outcome; the most recent of these was published in 1999.In total, only 745 infants were evaluated, and there was high heterogeneity between studies.The estimated effect size was a relative risk of 0.86 (95% CI, 0.41-1.79).This result represents what philosophers would call a scientific anomaly in the current paradigm; in fact, the concept of risk factor itself implies that by reducing it, we will reduce the outcome, and if RDS and ventilation are true risk factors for BPD, reducing them should reduce BPD.The (apparent) lack of effect of ACS on BPD is neglected or justified because the trials were conducted a long time ago, at a time when prenatal, perinatal, and neonatal care were vastly different, mortality rates were much higher, ventilatory techniques and ventilators were inferior, and even BPD was a different disease with a different definition. 3early, it will not be possible to conduct new RCTs comparing ACS with placebo.Therefore, to