The value of fetal growth biometry velocities to predict large for gestational age (LGA) infants

Objective: The use of growth velocities derived from fetal biometrics have been suggested to improve prediction of large for gestational age (LGA). Our objective was to determine if ultrasonographic growth velocities (GV) for abdominal circumference (AC) and estimated fetal weight (EFW) improve the prediction of LGA infants when compared to Hadlock EFW. Methods: This was a secondary analysis of data from a prospective study of women referred for growth ultrasounds during the 3rd trimester. Growth velocities (GV) for AC (AC − GV) and EFW (EFW − GV) were derived from the difference in Z-scores between measurements at the time of anatomy survey (18–24 week) and third trimester ultrasound (26–36 weeks). Change in AC − GV and EFW − GV >90th %ile alone or in combination with Hadlock EFW >90th%ile were compared for prediction of a LGA neonate. The primary outcome was the sensitivity and specificity of the (1) Hadlock EFW >90%ile, (2) AC − GV, (3) EFW − GV, (4) Hadlock EFW + AC − GV, and (5) Hadlock EFW + EFW − GV for the prediction of neonatal LGA. Test characteristics and area under the ROC curve (AUC) were determined. The association between the ultrasound predicted growth and adverse neonatal outcome was assessed using logistic regression. Results: Of 630 women meeting inclusion criteria, 85 (13.5%) had LGA neonates. Hadlock EFW showed a better NPV (98.0%) and sensitivity (71.1%) when compared to AC − GV (NPV 87.5%, sensitivity 17.7%) and EFW − GV (NPV 88.0%, sensitivity 22.6%). Combining Hadlock EFW and AC-GV or EFW − GV did little to improve the test characteristics for the prediction of LGA (AUC 0.65 and 0.64, respectively). All five measurements were unable to predict a composite of adverse neonatal outcome or need for maternal cesarean delivery. Adjustment of the growth velocities for gestational age at anatomy scan or 3rd trimester growth scan did not change these results. Conclusion: AC and EFW growth velocities do not appear to improve the prediction of LGA infants when compared to using the third trimester Hadlock EFW.