False diagnosis of small for gestational age and macrosomia - clinical and sonographic predictors.

Purpose: To investigate clinical and sonographic features associated with sonographic accuracy for the prediction of small for gestational age (SGA) and macrosomia at birth.

Methods: The database of a tertiary medical center was retrospectively searched for women who gave birth at term to a singleton healthy neonate in 2007-2014 and underwent sonographic estimated fetal weight (sEFW) evaluation within 3 d before delivery. Fetal growth restriction (FGR) and SGA were defined as sEFW or birth weight <10th percentile for gestational age; macrosomia was defined as birth weight >4000 grams. Data on maternal age, parity, gestational age, fetal gender, presentation, placental location, diabetes, hypertension, and oligo/polyhydramnios were compared between pregnancies with a false-negative and false-positive diagnosis of SGA or macrosomia.

Results: Of the 5425 fetal weight evaluations, 254 (4.7%) deviated by >15% from the actual birth weight. Nulliparity, absence of diabetes, neonatal female gender, anterior placenta, lower birth weight, and oligohydramnios were associated with a high deviation. We identified 482 SGA neonates (8.9%) and 633 macrosomic neonates (11.7%). A false-positive diagnosis of FGR was associated with oligohydramnios, absence of diabetes, and posterior placenta, and a false-negative diagnosis, with older maternal age, nulliparity, and male gender. A false-positive diagnosis of macrosomia was associated with older maternal age, multiparity, polyhydramnios, anterior placenta, and lack of hypertensive complications, and a false-negative diagnosis, with diabetes, hypertension, oligohydramnios, and vertex presentation.

Conclusion: The accuracy of sEFW is affected by clinical and sonographic pregnancy characteristics. Further analyses should focus on improving accuracy especially at the fetal weight extremes.