Antenatal prediction models for short‐ and medium‐term outcomes in preterm infants

INTRODUCTION In extremely and very preterm infants, predicting individual risks for adverse outcomes antenatally is challenging but necessary risk-stratified perinatal management parents' participation in decision-making about treatment. Our aim was to develop validate prediction models short-term (neonatal period) medium-term (3 years of age) based on antenatal maternal fetal factors alone. MATERIAL AND METHODS A population-based study conducted 31 157 neonates weighing ≤1500 g born between 22 31 weeks gestation registered the Neonatal Research Network Japan during 2006-2015. Short-term were assessed infants 13 751 among infants. The clinical data randomly divided into training validation sets a ratio 2:1. developed by selected using stepwise logistic regression from 12 with set. number incorporated model varied 3 10, basis each outcome. To evaluate predictive performance, area under receiver operating characteristics curve (AUROC) calculated outcome RESULTS Among outcomes, AUROCs in-hospital death, chronic lung disease, intraventricular hemorrhage (grade III or IV) periventricular leukomalacia 0.85 (95% CI 0.83-0.86), 0.80 0.79-0.81), 0.78 0.75-0.80), 0.58 0.55-0.61), respectively. cerebral palsy developmental quotient <70 at 3 years age 0.66 0.63-0.69) 0.72 0.70-0.74), CONCLUSIONS Although performance these outcome, their discriminative ability relatively good. We provided bedside tool calculating likelihood various infant complications use. would be valuable country, risk assessment tools could facilitate shared understanding infants' subsequent risks.