Presented at the Neonatal Society 2005 Autumn Meeting.

Doyle LW

Division of Paediatrics, Royal Women’s Hospital, Melbourne and the Departments of Obstetrics and Gynaecology, and Paediatrics, University of Melbourne.

Background: Systemic postnatal corticosteroid therapy in ventilator-dependent infants has short-term benefits. However, controversy exists about its effects on long-term survival free of neurosensory impairments and disabilities.

Aim: To review available data on the long-term effects of systemic postnatal corticosteroid treatment in very preterm infants.

Methods: Randomised controlled trials (RCTs) of postnatal corticosteroid therapy for prevention or treatment of CLD in preterm infants reporting rates of both mortality and long-term neurological outcomes were reviewed and their data synthesised. Including the DART study, there were twenty-one studies with data on 1791 randomised infants which met eligibility criteria. In addition to usual meta-analyses for overall and subgroups effects, the relationship between the corticosteroid effect on the combined outcome, death or cerebral palsy, and the risk of CLD in control groups was analysed by weighted meta-regression.

Results: Among all infants randomised a significantly higher rate of cerebral palsy after corticosteroid treatment (typical risk difference [RD] 0.04, 95% CI 0.01, 0.07; P=0.005) was partly offset by a nonsignificant reduction in mortality (typical RD –0.02, 95% CI –0.06, 0.02; P=0.33). Consequently there was no significant effect of corticosteroid treatment on the combined rate of mortality or cerebral palsy (typical RD 0.02, 95% CI –0.02, 0.07; P=0.28). Adverse effects of steroids on cerebral palsy were mostly limited to studies where treatment started in the first week of life. On meta-regression there was a significant negative relationship between the treatment effect on death or cerebral palsy and the risk of CLD in control groups (P=0.002) (Figure 1). Addition of the DART study data did not change this relationship.

Based on the 95% CI, with risks of CLD below 35% corticosteroid treatment significantly increased the chance of death or cerebral palsy, whereas with risks of CLD exceeding 65% it reduced this chance.

Conclusions: Long-term adverse effects of postnatal corticosteroids may be limited to treatment started early in the first week, when the level of risk for CLD is low.