Résumé
OBJECTIVE: To examine the validity of the Neonatal Neurobiologic Risk Score (NBRS) for predicting neurodevelopmental outcome to 3 years in infants born at < 28 weeks gestation. METHOLDOLOGY: The NBRS was retrospectively determined for 56 consecutive infants cared for in our NICU and prospectively followed to 3 years. Neurodevelopmental assessments performed at 3 years were correlated with the NBRS, and the predictive powers of individual items in the NBRS determined. RESULTS: The mean (range) birth weight was 908 (514-1295) g and gestational age was 26 (24-27) weeks. Three-year outcome was abnormal in 12 (21%) infants. A high NBRS at discharge was associated with an increased risk of abnormal 3-year outcome (odds ratio 2.56; 95% C.I. 1.4-4.7, p = 0.002). A modified NBRS using only significantly predictive items (acidosis, hypoxemia, hypotension, intraventricular hemorrhage, infection and hypoglycemia) demonstrated high sensitivity (1.00), specificity (0.98), positive predictive value (0.92) and negative predictive value (1.00) for abnormal 3-year outcome. CONCLUSIONS: This study confirms the validity of the NBRS as a simple and objective means of identifying very premature infants at highest risk of abnormal neurodevelopmental outcome, and of identifying specific events which may contribute to such outcomes.