Martin-Alonso Raquel, de Paco Matallana Catalina, Valiño Nuria, Chaveeva Petya, Dagklis Themistoklis, Siargkas Antonios, Wright Alan, Camacho Mario, Rolle Valeria, Santacruz Belén, Gil Maria M
Department of Obstetrics and Gynecology, Hospital Universitario de Torrejón, 28850 Torrejón de Ardoz, Madrid, Spain.
School of Medicine, Universidad Francisco de Vitoria, 28223 Pozuelo de Alarcón, Madrid, Spain.
Medicina (Kaunas). 2025 Sep 8;61(9):1626. doi: 10.3390/medicina61091626.
: Third-trimester screening is widely used to identify small for gestational age (SGA) and fetal growth restriction (FGR), but optimal models and timing remain under investigation. This study aimed to assess the performance of combined maternal factors and biomarkers, including ultrasound estimated fetal weight (EFW), Doppler indices, mean arterial pressure (MAP), and angiogenic biomarkers, for predicting SGA neonates after a routine 35-36 weeks' scan in an unselected population. We conducted a retrospective cohort study in three Spanish centers offering universal third-trimester ultrasound. Logistic regression analyses were carried out to predict birthweight < 10th and <5th percentile using maternal characteristics and medical history, EFW, MAP, Doppler indices, and the angiogenic biomarkers placental growth factor (PlGF) and soluble fms-like tyrosine kinase-1 (sFlt-1). Using a 10-fold cross-validation, we estimated the area under the receiver operating characteristic curve (AUC), detection rates (DRs), false-positive rates (FPRs), and their corresponding screen-positive rates (SPRs). External validation was performed using an independent cohort. Among 3992 pregnancies, the DR of ultrasound alone for birthweight <10th percentile was 47.9% (95% CI: 44.0 to 51.9), with an FPR of 7.3%. Adding maternal factors increased DR to 57.0% (95% CI: 53.0 to 60.9) at 10% FPR and to 83.0% (95% CI: 79.9 to 85.9) at 30% FPR. Similarly, the DR of ultrasound alone for birthweight < 5th percentile was 48.4% (95% CI: 43.1 to 53.6), with an FPR of 4.5%. Adding maternal factors increased DR to 65.7 (95% CI: 60.5 to 70.5) at 10% FPR and to 88.2 (95% CI: 84.4 to 91.3) at 30% FPR. The inclusion of MAP, Doppler, and biomarkers provided marginal additional gains, particularly for <5th percentile prediction. To achieve a DR > 80%, an SPR of approximately 40% was required. Performance improved when focusing on neonates born before 38 weeks, with a DR of 77.5 (95% CI: 68.6 to 84.9) at 10% FPR for SGA < 10th percentile. However, less than 40% of screen-positive women remained undelivered by 40 weeks, limiting the number requiring further surveillance. : A third-trimester screening at 35-36 weeks using maternal characteristics and EFW identifies most SGA neonates, particularly those delivering before 38 weeks. Even including other biomarkers, an SPR of about 40% should be necessary to achieve a high DR. However, less than 40% of the women would remain undelivered before a subsequent follow-up is required.
孕晚期筛查被广泛用于识别小于胎龄儿（SGA）和胎儿生长受限（FGR），但最佳模型和时机仍在研究中。本研究旨在评估包括超声估计胎儿体重（EFW）、多普勒指数、平均动脉压（MAP）和血管生成生物标志物在内的母体因素和生物标志物组合，在未选择人群中进行常规35 - 36周扫描后预测SGA新生儿的性能。我们在西班牙的三个提供普遍孕晚期超声检查的中心进行了一项回顾性队列研究。使用母体特征和病史、EFW、MAP、多普勒指数以及血管生成生物标志物胎盘生长因子（PlGF）和可溶性fms样酪氨酸激酶-1（sFlt-1），进行逻辑回归分析以预测出生体重低于第10百分位数和第5百分位数的情况。使用10倍交叉验证，我们估计了受试者操作特征曲线下面积（AUC）、检测率（DRs）、假阳性率（FPRs）及其相应的筛查阳性率（SPRs）。使用独立队列进行外部验证。在3992例妊娠中，仅超声对出生体重低于第10百分位数的检测率为47.9%（95%CI：44.0至51.9），假阳性率为7.3%。添加母体因素后，在10%假阳性率时检测率提高到57.0%（95%CI：53.0至60.9），在30%假阳性率时提高到83.0%（95%CI：79.9至85.9）。同样，仅超声对出生体重低于第5百分位数的检测率为48.4%（95%CI：43.1至53.6），假阳性率为4.5%。添加母体因素后，在10%假阳性率时检测率提高到65.7（95%CI：60.5至70.5），在30%假阳性率时提高到88.2（95%CI：84.4至91.3）。纳入MAP、多普勒和生物标志物带来的额外收益有限，特别是对于低于第5百分位数的预测。要实现检测率>80%，大约需要40%的筛查阳性率。当关注38周前出生的新生儿时性能有所改善，对于低于第10百分位数的SGA，在10%假阳性率时检测率为77.5（95%CI：68.6至84.9）。然而，不到40%的筛查阳性女性在40周前仍未分娩，限制了需要进一步监测的人数。：在35 - 36周使用母体特征和EFW进行孕晚期筛查可识别出大多数SGA新生儿，特别是那些在38周前分娩的新生儿。即使包括其他生物标志物，要实现高检测率也需要约40%的筛查阳性率。然而，在需要后续随访之前，不到40%的女性仍未分娩。
