A multiparameter model combining ultrasound and maternal serum biomarkers accurately predicts late-onset fetal growth restriction.

Abstract

Objectives: Late-onset fetal growth restriction (FGR) in pregnancy is frequently encountered and is linked to adverse perinatal results. This investigation aimed to evaluate the clinical use of a multiparameter model combining ultrasound features and maternal serum biomarkers for late-onset FGR.

Methods: A retrospective investigation was conducted on 350 pregnant women in their third trimester who received standard prenatal care between January 2021 and December 2025. Participants were divided into a late-onset FGR group (n = 175) and a normal fetal development group (n = 175) according to the Chinese expert consensus criteria. Ultrasound findings including fetal biometry and Doppler flow indices were measured. Maternal serum levels of soluble fms-like tyrosine kinase-1 (sFlt-1) and placental growth factor (PLGF) were quantified through enzyme-linked immunosorbent assay. Subsequently, the sFlt-1 to PLGF ratio (sFlt-1/PLGF) was calculated.

Results: Among the 350 enrolled women, the FGR group exhibited markedly reduced abdominal circumference, estimated fetal weight, cerebroplacental ratio, and elevated uterine artery-pulsatility index and sFlt-1/PLGF compared to those in the normal group (all P < 0.05). The multiparameter evaluation model achieved an area under the curve of 0.977, demonstrating a sensitivity of 91.4% and specificity of 92.6% at the optimal threshold.

Conclusions: The multiparameter model integrating ultrasound features and maternal serum biomarkers demonstrated excellent predictive performance for late-onset FGR, supporting its clinical use for early identification and management.