Aim In this study, we evaluated the Clinical Risk Index for Babies ‐ revised (CRIB‐II) score as a predictor of long‐term neurodevelopmental outcomes in preterm infants at 36 months’ corrected age.
Method CRIB‐II scores, which include birthweight, gestational age, sex, admission temperature, and base excess, were recorded prospectively on all infants weighing 1250g or less admitted to the neonatal intensive care unit (NICU). The sensitivity and specificity of CRIB‐II scores to predict poor outcomes were examined using receiver operating characteristic curves, and predictive accuracy was assessed using the area under the curve (AUC), based on the observed values entered on a continuous scale. Poor outcomes were defined as death or major neurodevelopmental disability (cerebral palsy, neurosensory hearing loss requiring amplification, legal blindness, severe seizure disorder, or cognitive score >2SD below the mean for adjusted age determined by clinical neurological examination and on the Wechsler Preschool and Primary Scale of Intelligence, Bayley Scales of Infant Development, or revised Leiter International Performance Scale).
Results Of the 180 infants admitted to the NICU, 155 survived. Complete follow‐up data were available for 107 children. The male:female ratio was 50:57 (47–53%), median birthweight was 930g (range 511–1250g), and median gestational age was 27 weeks (range 23–32wks). Major neurodevelopmental impairment was observed in 11.2% of participants. In a regression model, the CRIB‐II score was significantly correlated with long‐term neurodevelopmental outcomes. It predicted major neurodevelopmental impairment (odds ratio [OR] 1.57, bootstrap 95% confidence interval [CI] 1.26–3.01; AUC 0.84) and poor outcome (OR 1.46; bootstrap 95% CI 1.31–1.71, AUC 0.82) at 36 months’ corrected age.
Interpretation CRIB‐II scores of 13 or more in the first hour of life can reliably predict major neurodevelopmental impairment at 36 months’ corrected age (sensitivity 83%; specificity 84%).