Article Text

Neurological adverse effects of early postnatal dexamethasone in very preterm infants
  1. Service de Réanimation et Pédiatrie Néonatales
  2. Hôpital Antoine Béclère
  3. 157 rue de la porte de Trivaux
  4. 92141 Clamart Cedex
  5. France

    Statistics from

    Request Permissions

    If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

    Editor—Administration of dexamethasone very early after birth has been explored as a prophylactic treatment of bronchopulmonary dysplasia (BPD). The results have been conflictual. Saunders et al 1 and the recent meta-analysis by Tushar Bhuta and Arne Ohlsson2 concluded a reduction in the odds of BPD. However, Tapia et al 3 recently failed to show a reduction in its incidence. None of these studies has considered the impact of the treatment on the rate of periventricular leucomalacia (PVL).

    We used a protocol similar to Saunder’s study design (0.25 mg/kg intravenously twice before 24 hours of life) in neonates born before 29 completed weeks of gestation with respiratory distress syndrome. This open study conducted over 20 months was interrupted after an important increase in cystic PVL rate had been observed. Indeed, the incidence of PVL during this period (15%) was significantly higher than that observed during both the preceeding 20 months (7%) and the following 20 months (7%) (p < 0.02). Moreover, during the same period (73 treated and 99 untreated infants), PVL rate was higher in group treated with dexamethasone (23% vs 9%; p < 0.05, χ2 test after controlling for gestational age). Perinatal morbidity in the treated group was greater, but did not account for the higher PVL rate observed during the treatment period. Indeed, early neonatal mortality (< 7 days), the rates of respiratory distress syndrome, haemodynamic failure and congenital sepsis, as well as the circumstances of delivery known to be associated with high rate of PVL (premature rupture of the membranes, intrauterine infection, vaginal bleeding) were not statistically different before, during, and after the treatment period.

    Adverse effects on neuromotor functions have already been reported by Yeh et al in an early prolonged trial of dexamethasone.4 As far as the conflicting data of controlled trials on its benefits are concerned, early postnatal administration of dexamethasone should be considered with caution, as long as follow up results of other randomised studies are not yet available.