NGHIÊN CỨU SÀNG LỌC TIỀN SẢN GIẬT VÀ THAI CHẬM PHÁT TRIỂN TRONG TỬ CUNG TẠI BỆNH VIỆN PHỤ SẢN HẢI PHÒNG

Abstract

Objective: To evaluate the predictive value of maternal serum PAPP-A and PLGF in the first trimester for preeclampsia (PE) and fetal growth restriction (FGR).

Methods: A retrospective descriptive study of 1,297 singleton pregnancies screened for PAPP-A and PLGF in the first trimester, with complete outcome follow-up, at Hai Phong Obstetrics Hospital (2024-2025). ROC curves and AUC (bootstrap 95% CI, 2,000 replications) were built for four models: PAPP-A MoM, PLGF MoM, combined PAPP-A+PLGF, and the full Fetal Medicine Foundation (FMF) risk model (maternal factors, mean arterial pressure, uterine artery pulsatility index, PAPP-A and PLGF). Univariable and age-adjusted multivariable logistic regression assessed each marker’s independent predictive role.

Results: PE occurred in 1.54% (20/1,297; 4 early-onset <34 weeks) and FGR in 9.33% (121/1,297). PLGF MoM was significantly lower in both PE and FGR (p<0.001); PAPP-A MoM was significantly lower only in FGR (p<0.0001). The full FMF model had the highest AUC: 0.835 for PE and 0.753 for FGR. PLGF MoM independently predicted both outcomes (adjusted OR=0.140 and 0.269); PAPP-A MoM was independently significant only for FGR (OR=0.413). Among 32 high-risk women given prophylactic aspirin, 62.5% did not develop PE, consistent with a “treatment paradox” whereby observed screening performance underestimates the model’s true predictive value.

Conclusion: The full FMF model showed the highest predictive value for both PE and FGR; PLGF was the dominant biomarker, while PAPP-A added limited value once PLGF was included. PE-related findings should be regarded as exploratory given the limited event count, and observed screening performance may underestimate the model’s true value because a subset of high-risk women received prophylactic aspirin before outcomes were recorded.