Objective To evaluate the incidence of death or neurodevelopmental impairment (NDI) at 18-22 months corrected age in subjects enrolled in a trial of early dexamethasone treatment to prevent death or chronic lung disease in extremely low birth weight infants. Study design Evaluation of infants at 18-22 months corrected age included anthropomorphic measurements, a standard neurological examination, and the Bayley Scales of Infant Development-II, including the Mental Developmental Index and the Psychomotor Developmental Index. NDI was defined as moderate or severe cerebral palsy, Mental Developmental Index or Psychomotor Developmental Index <70, blindness, or hearing impairment. Results Death or NDI at 18-22 months corrected age was similar in the dexamethasone and placebo groups (65% vs 66%, P =.99 among those with known outcome). The proportion of survivors with NDI was also similar, as were mean values for weight, length, and head circumference and the proportion of infants with poor growth (50% vs 41%, P =.42 for weight less than 10th percentile); 49% of infants in the placebo group received treatment with corticosteroid compared with 32% in the dexamethasone group (P =.02). Conclusion The risk of death or NDI and rate of poor growth were high but similar in the dexamethasone and placebo groups. The lack of a discernible effect of early dexamethasone on neurodevelopmental outcome may be due to frequent clinical corticosteroid use in the placebo group. © Copyright 2014 Mosby Inc. All rights reserved.