BACKGROUND: Prenatal diagnosis of an infant suspected of having fetal growth restriction (FGR) is important due to its strong association with perinatal mortality and morbidity. The current Delphi consensus criteria include a decline >50 centiles in fetal growth when diagnosing late FGR; however, the evidence underpinning this criterion is limited.

OBJECTIVES: To analyze the relationships between the magnitude of decline in fetal growth and stillbirth, perinatal mortality and adverse neonatal outcomes.

STUDY DESIGN: This cohort study of 15,861 pregnancies was conducted at the Mater Mother's Hospital in Brisbane, Australia. Decline in fetal growth was calculated as a drop in either estimated fetal weight (EFW) or abdominal circumference (AC) centiles between two ultrasound scans performed after 18+0 weeks of gestation. Relationships between declining fetal growth and the outcomes were, firstly, analyzed as a continuous variable and, if significant, further assessed with the rate of decline and different magnitudes of decline, compared to the referent category (change in growth of +/- 10 centiles between scans). The three categories of growth decline were >10 - <25 centiles, ≤25 - <50 centiles and ≥50 centiles. Associations were analyzed by logistic regressions. The primary study outcomes were stillbirth and perinatal mortality (composite of stillbirth and neonatal death). Secondary outcomes were birth of an SGA infant (birthweight <10th centile for gestation), emergency cesarean section (CS) for non-reassuring fetal status (NRFS), and composite severe neonatal morbidity.

RESULTS: The risks of stillbirth and perinatal mortality increased significantly by 2.6% (0.4%-4.6%) and 2.8% (1.0%-4.5%), respectively, per one centile decline in fetal growth. Additionally, the odds of stillbirth (adjusted odds ratio (aOR) 3.68 (95% CI 1.32-10.24)) and perinatal mortality (aOR 4.44 (1.82-10.84)) compared with the referent group were significantly increased only when the decline was ≥50 centiles, regardless of birth weight. Furthermore, none of the primary outcomes were significantly associated with the rate of growth decline. The risk of a small for gestational age infant increased by 2.4% (2.2%-2.7%) for every centile decline. Conversely, reduced fetal growth was not associated with emergency CS for NRFS or severe neonatal morbidity.

CONCLUSIONS: Our results support the use of a ≥50 centile decline in fetal growth as a criterion for identifying infants at risk of late fetal growth restriction. This cut-off also identifies fetuses at high risk of perinatal mortality regardless of birth weight and rate of growth decline. Our findings may guide obstetric practice by alerting clinicians to the importance of incorporating the magnitude of fetal growth decline into antenatal counselling and decisions regarding timing of birth.