Expanding the Estimated Fetal Weight Definition of Growth Restriction by Adding Small Abdominal Circumference: Prediction of Neonatal Morbidity.

Objective: The Society for Maternal-Fetal Medicine's (SMFM) diagnostic criteria for fetal growth restriction (FGR) recently added abdominal circumference (AC) <10th percentile to estimated fetal weight (EFW) <10th percentile; however, its prediction of neonatal morbidity is unknown. Our objective was to compare the two definitions for their prediction of composite neonatal morbidity.

Methods: Secondary analysis of the Fetal Growth Study-Singletons, 2009-2013. The last ultrasound (mean 36.9 ± 2.3 weeks) was included from non-anomalous fetuses. Composite neonatal morbidity was the primary outcome: metabolic acidosis, neonatal intensive care unit stay >3 days, significant respiratory morbidities, seizures, hyperbilirubinemia requiring exchange transfusion, intrapartum aspiration, necrotizing enterocolitis, hypoglycemia, hypoxic ischemic encephalopathy, periventricular leukomalacia, sepsis, retinopathy of prematurity, or neonatal death. The secondary outcome was small for gestational age (SGA). Logistic regression modeled the association of each FGR definition with outcomes, and receiver operating characteristic area under the curve (AUC) assessed predictive ability.

Results: Of 2400 eligible individuals, 135 (5.6%) neonates had composite neonatal morbidity, and 245 (10%) were SGA. At the last ultrasound, 181 (7.5%) had FGR based on EFW alone (original definition) and 215 (9.0%) had FGR based on a small EFW or AC (expanded definition) (P < .0001). Both definitions had poor discrimination for composite neonatal morbidity (original: AUC 0.52, 95% confidence interval [CI] 0.49-0.54; expanded: AUC 0.51, 95% CI, 0.48-0.54). Both had acceptable discrimination of SGA (original: AUC 0.70, 95% CI 0.67-0.73; expanded: AUC 0.71, 95% CI 0.68-0.75).

Conclusions: Adding AC <10th percentile to the EFW <10th percentile definition of FGR significantly increased the incidence of FGR but did not improve the prediction of neonatal morbidity in a low-risk population. The SMFM guideline for FGR should be adopted with caution.