Temporal trends in estimated fetal weight accuracy following national guideline implementation: a six-year cohort study at a tertiary center in Southern China (2014-2019).

INTRODUCTION

In 2012, China released its first national guidelines for prenatal ultrasound screening, marking a major step toward standardized and assessable prenatal ultrasound practice nationwide. However, real-world evidence on the effectiveness of these guidelines in clinical practice remains limited, particularly from large-scale, long-term datasets. To address this gap, we aimed to examine temporal trends in the accuracy of ultrasound-based estimated fetal weight (EFW) following the guidelines’ implementation in China. MATERIALS AND

METHODS

This retrospective cohort study analyzed real-world data from 31,681 singleton pregnancies between 2014 and 2019, drawn from the integrated ultrasound and perinatal databases at a tertiary maternity hospital in southern China. Temporal trends in EFW accuracy, fetal biometric parameters (biparietal diameter, head circumference, abdominal circumference, femur length), and diagnostic performance for fetal growth abnormalities were analysed by calendar quarter from January 2014 to December 2019. RESULT: EFW accuracy improved progressively over time, with the mean percentage error nearing zero and the proportion of estimates within 10% and 15% of birthweight rising from 78.0% to 93.2% in 2014 to 86.2% and 97.3% in 2019 (OR = 1.032 and 1.045; P < 0.001). Accuracy of all biometric parameters increased, particularly for femur length (β = 0.002). Detection of extreme fetal growth abnormalities based on EFW became more consistent, with detection rates and false-positive rates (FPRs) for severe large-for-gestational-age (EFW > 97th percentile) decreasing from 2.06% to 0.67% (FPRs 1.59% to 0.22%), while fetal growth restriction (EFW <3rd percentile) detection slightly increased from 0.37% to 0.50% (FPRs 0.19% to 0.28%), and false-negative rates remained stable. Similar improvements were observed using absolute birthweight thresholds, with reduced overestimation of macrosomia and enhanced detection of low birthweight, indicating a better balance in EFW’s diagnostic performance across the fetal growth spectrum.

CONCLUSION

This large-scale longitudinal study revealed sustained improvements in EFW accuracy, biometric measurement consistency, and fetal growth abnormality detection following China’s 2012 national guidelines. These advances likely reflect standardized protocols, training, quality assurance, and accumulating operator experience. Future multicenter studies are needed to assess their broader clinical impact.