Objective: To identify independent predictors of adverse neonatal outcome in cases of fetal growth restriction (FGR) at > or = 34 weeks.
Methods: From a cohort of 481 FGR cases delivered at > or = 34 weeks, demographic and obstetric variables, fetal biometry and Doppler indices of the uterine, umbilical and fetal middle cerebral arteries available within 2 weeks of delivery, were related to adverse neonatal outcome, defined as admission to the neonatal intensive care unit for indications other than low birth weight alone.
Results: Logistic regression analysis showed that gestational age (GA) at delivery (odds ratio (OR) = 0.59; 95% CI, 0.50-0.70), abdominal circumference (AC) centile (OR = 0.69; 95% CI, 0.59-0.81) and umbilical artery (UA) pulsatility index (PI) centile (OR = 1.02; 95% CI, 1.01-1.04) significantly correlated with adverse neonatal outcome. From this model we calculated a score of adverse neonatal outcome expressed by the formula: (UA-PI centile/3) - (10 x AC centile) + (10 x (40 - GA at delivery in weeks)). Receiver-operating characteristics curve analysis demonstrated that a score of > or = 25 optimally predicted adverse neonatal outcome (sensitivity of 75%, false-positive rate of 18%). Beyond 37.5 weeks, gestational age no longer had an independent impact on outcome.
Conclusions: In late preterm or term FGR, GA at delivery is the most important predictor of adverse neonatal outcome. At > 37.5 weeks, delivery may be the best option to minimize adverse outcome in all FGR cases. At 34-37 weeks, a score based on GA at delivery, UA-PI centile and AC centile optimally predicts adverse neonatal outcome.