Raykov Sasho, Yordanova-Ignatova Teodora, Ignatov Petar N
Acibadem City Clinic Tokuda, Department of Assisted Reproduction, Sofia, Bulgaria.
Orthogyn Medical Center, Sofia, Bulgaria.
Eur J Obstet Gynecol Reprod Biol. 2025 Jul;311:114032. doi: 10.1016/j.ejogrb.2025.114032. Epub 2025 May 7.
To evaluate the effectiveness of combining sociodemographic factors, circulating endothelial growth factor-1 (sFlt-1), cancer antigen 125 (CA125), placental growth factor (PLGF), along with ultrasound measurements of crown-rump length (CRL), and fetal heart rate (FHR) in assessing the risk of spontaneous abortion, and to develop a predictive model for miscarriage risk.
Between 2020 and 2023, we conducted a prospective observational study involving 235 pregnancies. Participants were divided into 2 groups - viable pregnancies and miscarriage before the 14th week of gestation. We recruited all patients at 6.0 to 6.6 weeks of gestation. At that point, demographic data was collected, and the first measurements of CRL and FHR were performed. Blood samples were taken for sFlt-1, PLGF, and CA125 analysis. Consequent visits were appointed every 2 weeks, which included a US scan and repeated blood serum tests. The outcome variable was early pregnancy loss, defined by the occurrence of spontaneous abortion (SA) before 14 weeks of gestation.
Our study has shown that in cases with first-trimester pregnancy loss, the FHR and sFlt-1 were consistently decreased, while MA and CA125 were increased. We observed that a combination of MA, FHR, CRL, CA125, and sFlt-1 can reliably identify singleton pregnancies at risk of spontaneous abortion before 14 weeks of gestation. Each additional variable substantially improved the prediction characteristics, suggesting a good stepwise discriminatory performance of the overall model. The combination of all variables achieved a DR of 73,53 %, and more importantly - clinically feasible Positive Predictive Value (PPV) of 75,81 %, and high Negative Predictive Value (NPV = 94,75 %) levels.
Combining maternal age, FHR, CA125, and sFlt-1 can reliably identify singleton pregnancies at risk of spontaneous abortion before the 14th week of gestation. To our best knowledge, this is the first SA prediction model combining CA125 and sFlt-1 serum markers with well-established US modalities and demographic characteristics. External validation would be required before adopting the proposed prediction model in clinical practice.
评估将社会人口统计学因素、循环内皮生长因子-1（sFlt-1）、癌抗原125（CA125）、胎盘生长因子（PLGF），以及头臀长度（CRL）和胎儿心率（FHR）的超声测量结果相结合，在评估自然流产风险方面的有效性，并建立流产风险预测模型。
在2020年至2023年期间，我们进行了一项前瞻性观察研究，涉及235例妊娠。参与者被分为两组——可存活妊娠和妊娠14周前流产。我们在妊娠6.0至6.6周时招募了所有患者。在那时，收集了人口统计学数据，并首次测量了CRL和FHR。采集血样进行sFlt-1、PLGF和CA125分析。随后每2周安排一次随访，包括超声扫描和重复血清检测。结局变量为早期妊娠丢失，定义为妊娠14周前发生自然流产（SA）。
我们的研究表明，在孕早期流产的病例中，FHR和sFlt-1持续下降，而平均动脉压（MA）和CA125升高。我们观察到，MA、FHR、CRL、CA125和sFlt-1的组合能够可靠地识别妊娠14周前有自然流产风险的单胎妊娠。每增加一个变量都显著改善了预测特征，表明整体模型具有良好的逐步判别性能。所有变量的组合实现了73.53%的诊断率，更重要的是——临床上可行的阳性预测值（PPV）为75.81%，阴性预测值（NPV = 94.75%）较高。
结合孕妇年龄、FHR、CA125和sFlt-1能够可靠地识别妊娠14周前有自然流产风险的单胎妊娠。据我们所知，这是首个将CA125和sFlt-1血清标志物与成熟的超声检查方法及人口统计学特征相结合的SA预测模型。在临床实践中采用所提出的预测模型之前，需要进行外部验证。
