文献速递产科母胎医学

双胎妊娠11-13周母体因素与生物标志物预测子痫前期:EVENTS试验数据

Prediction of pre-eclampsia in twin pregnancy by maternal factors and biomarkers at 11-13 weeks' gestation: data from EVENTS trial.
PubMed 文献 · Ultrasound Obstet Gynecol · 2021-01-01

基于EVENTS试验等两个前瞻性多中心数据集,验证双胎妊娠子痫前期筛查模型并开发新模型。EVENTS数据集1798例中168例(9.3%)发生子痫前期,原模型校准良好;联合数据3938例中339例(8.6%)发生子痫前期。联合母体因素、MAP、UtA-PI和PlGF筛查效果最佳,10%假阳性率下对<32周分娩子痫前期检出率86.4%。

要点

  • 双胎妊娠可沿用基于母体特征和病史的原有先验模型评估子痫前期风险。
  • 计算后验风险时需采用双胎妊娠UtA-PI、MAP和PlGF的log10 MoM新分布。
  • 联合母体因素、MAP、UtA-PI和PlGF筛查效果最佳,优于单独母体因素。

核心结论

  • 双胎妊娠子痫前期筛查可沿用原母体因素先验模型,但后验风险需使用双胎特有的UtA-PI、MAP和PlGF的log10 MoM分布。

研究设计与样本

  • 两个前瞻性非干预多中心筛查数据集,EVENTS试验数据集1798例双胎妊娠,合并数据集3938例。
  • 采用竞争风险模型估计<32周和<37周分娩子痫前期的个体风险,以检出率和ROC曲线下面积评估筛查性能。

主要发现

  • EVENTS数据集1798例中168例(9.3%)发生子痫前期,原模型校准良好:<32周校准斜率0.827、截距0.009,<37周斜率0.942、截距-0.207。
  • 合并数据3938例中339例(8.6%)发生子痫前期,253例(6.4%)在<37周分娩子痫前期。
  • 双胎妊娠中MAP、UtA-PI和PlGF在较早孕龄区分度高于单胎,较晚孕龄低于单胎;PAPP-A在子痫前期与未患病妊娠间差异小。

关键数值与推荐

  • 10%假阳性率下,仅母体因素筛查对<32周分娩子痫前期检出率30.6%,联合试验升至86.4%。
  • 10%假阳性率下,仅母体因素筛查对<37周分娩子痫前期检出率24.9%,联合试验为41.1%。

临床怎么用

  • 双胎妊娠11-13周筛查子痫前期时,可先用原母体因素模型计算先验风险。
  • 联合MAP、UtA-PI和PlGF并采用双胎特异性分布计算后验风险,以提高检出率。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

OBJECTIVES: First, to validate a previously developed model for screening for pre-eclampsia (PE) by maternal characteristics and medical history in twin pregnancies; second, to compare the distributions of mean arterial pressure (MAP), uterine artery pulsatility index (UtA-PI), serum placental growth factor (PlGF) and serum pregnancy-associated plasma protein-A (PAPP-A) in twin pregnancies that delivered with PE to those in singleton pregnancies and to develop new models based on these results; and, third, to examine the predictive performance of these models in screening for PE with delivery at < 32 and < 37 weeks' gestation. METHODS: Two datasets of prospective non-intervention multicenter screening studies for PE in twin pregnancies at 11 + 0 to 13 + 6 weeks' gestation were used. The first dataset was from the EVENTS (Early vaginal progesterone for the preVention of spontaneous prEterm birth iN TwinS) trial and the second was from a previously reported study that examined the distributions of biomarkers in twin pregnancies. Maternal demographic characteristics and medical history from the EVENTS-trial dataset were used to assess the validity of risks from our previously developed model. The combined data from the first and second datasets were used to compare the distributional properties of log10 multiples of the median (MoM) values of UtA-PI, MAP, PlGF and PAPP-A in twin pregnancies that delivered with PE to those in singleton pregnancies and develop new models based on these results. The competing-risks model was used to estimate the individual patient-specific risks of delivery with PE at < 32 and < 37 weeks' gestation. Screening performance was measured by detection rates (DR) and areas under the receiver-operating-characteristics curve. RESULTS: The EVENTS-trial dataset comprised 1798 pregnancies, including 168 (9.3%) that developed PE. In the validation of the prior model based on maternal characteristics and medical history, calibration plots demonstrated very good agreement between the predicted risks and the observed incidence of PE (calibration slope and intercept for PE < 32 weeks were 0.827 and 0.009, respectively, and for PE < 37 weeks they were 0.942 and -0.207, respectively). In the combined data, there were 3938 pregnancies, including 339 (8.6%) that developed PE and 253 (6.4%) that delivered with PE at < 37 weeks' gestation. In twin pregnancies that delivered with PE, MAP, UtA-PI and PlGF were, at earlier gestational ages, more discriminative than in singleton pregnancies and at later gestational ages they were less so. For PAPP-A, there was little difference between PE and unaffected pregnancies. The best performance of screening for PE was achieved by a combination of maternal factors, MAP, UtA-PI and PlGF. In screening by maternal factors alone, the DR, at a 10% false-positive rate, was 30.6% for delivery with PE at < 32 weeks' gestation and this increased to 86.4% when screening by the combined test; the respective values for PE < 37 weeks were 24.9% and 41.1%. CONCLUSIONS: In the assessment of risk for PE in twin pregnancy, we can use the same prior model based on maternal characteristics and medical history as reported previously, but in the calculation of posterior risks it is necessary to use the new distributions of log10 MoM values of UtA-PI, MAP and PlGF according to gestational age at delivery with PE. © 2020 International Society of Ultrasound in Obstetrics and Gynecology.

来源:Ultrasound Obstet Gynecol,查看原文。版权归原作者所有。

引用

  1. Benkő Z, Wright A, Rehal A, et al. Prediction of pre-eclampsia in twin pregnancy by maternal factors and biomarkers at 11-13 weeks' gestation: data from EVENTS trial. Ultrasound Obstet Gynecol. 2021. PMID: 33142361. doi:10.1002/uog.23531

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。