Dexamethasone (DEX) is associated with decreased duration of ventilator dependency in very preterm infants at risk for chronic lung disease. Observational studies have provided conflicting conclusions about the effect of DEX on the risk of retinopathy of prematurity (ROP). We studied this issue during a randomized placebo-controlled trial of DEX in preterm infants with birth weight < 1500 grams, who were 15 to 25 days old at study entry and who were ventilator dependent and not weaning. Treatment with DEX or placebo was continued for 42 days and the dosage was tapered as described by Cummings et al [NEJM 1989;320:1505-1510]. 57 infants were randomized to DEX; 61 to placebo. Infants who survived to 6 weeks of age were examined by a single opthalmologist and followed thereafter according to his recommendations. Criteria for surgery were stage three plus ROP involving five contiguous or eight total clock hours of the retina. 3 DEX and 7 placebo infants died before an eye examination. The 54 DEX and 54 placebo infants who were examined at least once were comparable in terms of birth weight [expressed as median(range): DEX - 747 (420-1362); placebo - 761 (440-1324)], gestational age [DEX- 25 (23-29); placebo - 26 (23-31)], gender [DEX - 48% male; placebo - 54% male], race [DEX - 63% white; placebo - 54% white], and the percentage with a major abnormality on cranial ultrasonography [DEX - 17%; placebo - 13%]. More DEX infants were treated with antenatal steroids [DEX - 35%; placebo - 20%; p=0.09]. 29 DEX infants (56%) and 23 placebo infants (45%) had grade 3 ROP. 19 DEX infants (35%) and 11 placebo infants (20%) underwent surgery for ROP [odds ratio = 2.1; 95% confidence limits = 0.9, 5.0; p = 0.09]. 47 DEX and 44 placebo infants were followed through one year adjusted age. 3 DEX infants(6%) and 3 placebo infants (7%) were diagnosed with severe myopia and 6 DEX(13%) and 2 placebo (5%) infants were diagnosed with strabismus. Two DEX infants (4%) were blind in one eye; 3 placebo infants (7%) were blind in both eyes. In conclusion, the risk of severe ROP is high among infants treated with a 42-day tapering course of DEX. Further investigation is needed to investigate whether this risk can be decreased by modifying the dosage or timing of DEX treatment.

Author information

Rights and permissions

Reprints and Permissions

About this article

Further reading