Comparison of the INTERGROWTH-21st, National Institute of Child Health and Human Development, and WHO fetal growth standards.

Abstract

To compare the ability of the INTERGROWTH-21st, National Institute of Child Health and Human Development, and WHO fetal growth standards to identify fetuses at risk of adverse perinatal outcomes. A retrospective analysis was performed among women enrolled in a multicenter randomized controlled trial (Routine Antenatal Diagnostic Imaging with Ultrasound) that was conducted in six states in the USA between November 1987 and May 1991. The predictive capability of various biometric indicators (biparietal diameter, femur length, abdominal circumference, head circumference, estimated fetal weight, and birthweight) was evaluated. Adverse outcomes included severe morbidity and perinatal death. There were 9409 women included. Biometric indicators measured at a gestational age of 18-24 weeks had insufficient predictive sensitivity (range, 4%-47%). By contrast, measurements taken at 28-34 weeks predicted statistically significant relative risk (range 1.5-10.2; P<0.05 for 27/30 relative risk values) and area under the receiver operating characteristic curve (range, 0.50-0.59; P<0.05 for 8/15 curves). Nonetheless, differences in accuracy between standards for predicting adverse perinatal outcomes were subtle (P>0.05 for differences in area under the curve values). Although useful to monitor fetal growth trajectory and the level of risk, all three fetal growth standards provided limited accuracy for identifying fetuses at risk of adverse perinatal outcomes.