PAEDIATRIC RESPIRATORY REVIEWS (2004) 5(Suppl A), S321–S327 Use of steroids in the perinatal period 1 Henry L. Halliday ° Regional Neonatal Unit, Royal Maternity Hospital, and Department of Child Health, Queen’s University of Belfast, Belfast, Northern Ireland Summary Introduction: Corticosteroids can be used prenatally to mature the fetal lungs and postnatally to treat or prevent chronic lung disease (CLD). Randomised controlled trials have been performed to evaluate the benefits and risks of perinatal corticosteroid therapy. Methods: Systematic reviews of randomised controlled trials of prenatal and postnatal corticosteroids in the Cochrane Library were examined to determine the cost-benefit ratios of treatment. Outcomes are given as numbers needed to treat (NNT) or numbers needed to harm (NNH) with 95% confidence intervals (CI). Results: Prenatal corticosteroids reduce the risk of RDS (NNT 11; 95% CI 9–16), surfactant use (NNT 9; 95% CI 5–62), intraventricular haemorrhage (NNT 9; 95% CI 6–19) and neonatal mortality (NNT 23; 95% CI 16–42). There are no short-term or long-term adverse effects of a single course of prenatal betamethasone. However, repeated courses of prenatal steroids could be harmful and should be avoided outside of a randomised controlled trial. Postnatal corticosteroids can be used to prevent CLD (early use) or to treat it (late use). Beneficial effects include earlier extubation (typical NNT 5; 95% CI 4–10), reduced CLD (typical NNT 10; 95% CI 8–17) and avoidance of late steroids (NNT 7; 95% CI 6–10). However, there are significant adverse short-term effects such as hyperglycaemia (typical NNH 8; 95% CI 6–10), hypertension (typical NNH 10; 95% CI 8–14). Hy- pertrophic cardiomyopathy (typical NNH 5; 95% CI 4–11), gastrointestinal bleeding (typical NNH 17; 95% CI 11–33) and growth failure (NNH 2; 95% CI 1–2). More important are long-term adverse effects of cerebral palsy (typical NNH 8; 95% CI 6–17), developmental delay (typical NNH 7; 95% CI 4–33) and abnormal neurological examination (typical NNH 4; 95% CI 2–14). These adverse effects are more pronounced with early (<96 h) treatment but probably also occur when steroids are given later in the postnatal period. Conclusions: A single course of prenatal betamethasone has clear benefits for the fetus who is likely to be born preterm but repeated courses may be harmful. Postnatal steroids should be avoided if at all possible. They might be indicated in very low doses for ventilator-dependent infants who might otherwise die without them. © 2004 Elsevier Science Ltd. INTRODUCTION The two main uses of prenatal steroids are: 1. to mature the fetal lungs when used prenatally, and 2. to treat or prevent chronic lung disease (CLD) when administered postnatally. 1 For prenatal use the *Tel.: +44-(28)-90894687; Fax: +44-(28)-90236203; E-mail: h.halliday@qub.ac.uk Correspondence address: Regional Neonatal Unit, Royal Maternity Hospital, Grosvenor Road, Belfast BT12 6BB, Northern Ireland steroid of choice is usually betamethasone whereas postnatally dexamethasone is most commonly used. 2 Different doses have been used depending upon the indication for treatment and concerns have been raised about possible harmful effects in addition to the desired effects. 1,2 In a seminal study in 1972, Liggins and Howie demonstrated that prenatal corticosteroid treatment reduced the incidence of respiratory distress syndrome (RDS). 3 Since then at least 1 Supported by grants from Action Research (UK). 1526-0542/$ – see front matter © 2004 Elsevier Science Ltd. All rights reserved.