Department of Epidemiology and Biostatistics, School of Public Health, Xi'an Jiaotong University Health Science Center, Xi'an, Shaanxi, China.
Division of Behavioral Medicine, Department of Pediatrics, Jacobs School of Medicine and Biomedical Sciences, State University of New York at Buffalo, New York, USA.
J Hypertens. 2020 Jul;38(7):1355-1366. doi: 10.1097/HJH.0000000000002395.
The purpose of this study was to parameterize mid-trimester drop in blood pressure (BP) trajectory during pregnancy and to evaluate its utility for predicting preeclampsia.
To develop parametric models for BP trajectory during pregnancy, we used data from 7923 Chinese pregnant women with 24 810 routine antenatal care visits. Then, we evaluated the utility of BP trajectory parameters for predicting clinician-diagnosed preeclampsia in a separate sample of 3524 pregnant women from a randomized controlled trial of prenatal vitamin supplementation conducted in the same area. We focused on parameters related to the mid-trimester BP drop, including the gestational age and BP value at the nadir (lowest point), change in BP, velocity, and area under curve during two periods (from 12 weeks of gestation to the nadir and from the nadir to 33 weeks of gestation).
All participants in our analysis had a mid-pregnancy drop in their SBP, DBP, and mean arterial pressure (MAP) trajectories. There were high correlations (|r| > 0.90) among trajectory parameters of the same BP measure. The final prediction model included selective parameters of SBP, DBP, and MAP trajectories, prepregnancy BMI and gestational age at the first antenatal care visit. The area under the receiver-operating curve for predicting preeclampsia was 0.886 (95% confidence interval 0.846--0.926) in the training dataset and 0.802 (0.708--0.895) in the validation dataset.
Our novel BP trajectory parameters are informative and can predict preeclampsia at a clinically acceptable level.
本研究旨在对妊娠中期血压（BP）下降轨迹进行参数化，并评估其对预测子痫前期的效用。
为了建立妊娠期间 BP 轨迹的参数模型，我们使用了来自 7923 名中国孕妇的 24810 次常规产前保健就诊的数据。然后，我们在同一地区进行的产前维生素补充随机对照试验的 3524 名孕妇的独立样本中评估了 BP 轨迹参数预测临床诊断子痫前期的效用。我们重点关注与中期 BP 下降相关的参数，包括最低点（最低点）时的妊娠年龄和 BP 值、BP 变化、速度和两个时期的曲线下面积（从妊娠 12 周至最低点和从最低点至 33 周）。
我们分析中的所有参与者的 SBP、DBP 和平均动脉压（MAP）轨迹都有中期下降。同一 BP 测量的轨迹参数之间存在高度相关性（|r|＞0.90）。最终预测模型包括 SBP、DBP 和 MAP 轨迹的选择性参数、孕前 BMI 和首次产前检查时的妊娠年龄。在训练数据集中，预测子痫前期的受试者工作特征曲线下面积为 0.886（95%置信区间 0.846-0.926），在验证数据集中为 0.802（0.708-0.895）。
我们的新型 BP 轨迹参数是信息丰富的，可以在可接受的临床水平预测子痫前期。
