Reduced use of postnatal corticosteroids in extremely preterm infants and incidence of chronic lung disease

Chronic lung disease (CLD) is a major cause of respiratory illness in preterm infants. Many strategies have been attempted to prevent CLD, but success has been limited because the aetiology of the disease is multifactorial and multiple interventions may be needed. Postnatal corticosteroids have been used extensively in babies requiring mechanical ventilation, on the basis of the results of several randomised studies conducted in the 1980s and 1990s. However, when the long-term neurodevelopment of infants enrolled in these trials were reported, many showed a trend towards an increased risk of cerebral palsy in those receiving corticosteroids. These concerns were also confirmed in some systematic reviews. Subsequently, the American Academy of Paediatrics (AAP) and Canadian Paediatric Society (CPS) in 2002 issued a consensus statement which recommended that 'outside the context of a randomised, controlled trial, the use of corticosteroids should be limited to exceptional clinical circumstances'. Prior to those publications, a reduction in the use of corticosteroids occurred in many units: postnatal corticosteroid use peaked in 1996-1998 and then declined progressively from 1999 to 2004, with no apparent impact on mortality and short-term morbidity. It has been reported that the decreasing use of postnatal steroids could be associated with increased oxygen dependency.
An Australian group conducted a study to examine the effect of change in practice with the use of postnatal corticosteroids in extremely preterm babies over an 8-year period. The incidence of CLD and of cerebral palsy at 1 year of age were evaluated.
Babies of birth weight < 1,000 g or gestational age < 28 weeks admitted from 1997 to 2004 to the NICU at the Mater Mothers' Hospital in Brisbane were included in this retrospective analysis. The study was split into two different periods: group 1(1997-2000, n=389) and group 2 (2001-2004, n=368).
Results show a significant decrease in the use of corticosteroids from 27% in group 1 to 13% in group 2 (p=0.0001), and total dose-mg/kg (4.5±2.9 vs 2.6±1.6, p=0.0001). The incidence of CLD and need for home oxygen was similar between groups. However, the incidence of cerebral palsy reduced from 10.4% in group 1 to 6.6% in group 2, even if this difference did not reach statistical significance (Odds Ratio=0.63).
Taken together, these results indicate that postnatal corticosteroid use for CLD decreased significantly in extremely preterm infants included in the present study, even before the publication of the joint statement of AAP and CPS. Noteworthy, this decrease appears to have had no impact on the incidence of CLD or the need for home oxygen therapy. A possible benefit of the decrease in corticosteroid use may be a resultant lower incidence of late-onset neonatal infection. The benefit in terms of a trend towards a reduced rate of cerebral palsy requires further investigation.

Top