超声多模态评分及sFlt-1、D-Dimer对原因不明复发性流产妊娠结局的预测  

Prediction of pregnancy outcome of unexplained recurrent abortion by ultrasonic multimodal score and sFlt-1 and D-Dimer

作　　者：耿玲玲 邱建国 张文华 徐见兴[1] 袁飞燕 刘玉华[1] GENG Lingling;QIU Jianguo;ZHANG Wenhua;XU Jianxing;YUAN Feiyan;LIU Yuhua(Department of Obstetrics and Gynecology,Dongguan Maternal and Child Health Hospital,Guangdong Dongguan 523000,China;Department of Ultrasound,Dongguan Maternal and Child Health Hospital,Guangdong Dongguan 523000,China)

机构地区：[1]东莞市妇幼保健院妇产科,广东东莞523000 [2]东莞市妇幼保健院超声科,广东东莞523000

出　　处：《新疆医科大学学报》2023年第11期1490-1495,共6页Journal of Xinjiang Medical University

基　　金：东莞市社会科技发展一般项目(202050715007858);广东省医学科研基金项目(A2022461)。

摘　　要：目的探讨超声多模态评分及可溶性血管内皮生长因子受体-1(Soluble vascular endothelial growth factor receptor-1,sFlt-1)、D-二聚体(D-dimer,D-D)对原因不明复发性流产(Unexplained recurrent spontaneous abortion,URSA)妊娠结局的预测价值,并构建URSA不良妊娠结局的风险预测模型。方法选取2020年4月至2022年3月东莞市妇幼保健院收治的有URSA史的早孕患者136例,根据随访妊娠结局分为流产组和未流产组,比较两组超声多模态评分及sFlt-1、D-D水平,采用多因素Logistic回归分析法分析影响URSA妊娠再次流产的危险因素,建立风险预测模型并进行模型验证。结果136例孕妇保胎治疗后流产率为38.24%。流产组年龄≥35岁、饮酒史、URSA家族史、自然流产≥4次患者占比及血清sFlt-1、D-D水平均高于未流产组,孕酮及超声多模态评分均低于未流产组(P<0.05)。二元Logistic回归分析显示,自然流产≥4次、sFlt-1高水平、D-D高水平、超声多模态评分低是影响URSA妊娠再流产的独立危险因素(P<0.05)。预测模型方程:Logit(P)=-1.635+0.605×自然流产≥4次+1.332×sFlt-1+0.841×D-D+1.042×超声多模态评分;内部验证结果显示,预测模型的校正曲线与理想曲线拟合良好(Hosmer-Lemeshowχ^(2)=0.322,P=0.113)。受试者工作特征(Receiver operating characteristic,ROC)曲线结果显示,Logistic预测模型的AUC为0.933(95%CI:0.877~0.969),优于sFlt-1、D-D、超声多模态评分单项预测(P<0.05)。结论血清sFlt-1、D-D水平升高、超声多模态评分降低、自然流产≥4次是URSA妊娠再流产的高危因素,构建的风险预测模型区分度、拟合度好,能够直观准确预测URSA不良妊娠结局发生风险。Objective To investigate the predictive value of ultrasound multimodal scoring and soluble vascular endothelial growth factor receptor-1(sFlt-1)and D-dimer(D-D)on pregnancy outcome in unexplained recurrent spontaneous abortion(URSA)and to construct a risk prediction model for adverse pregnancy outcome in URSA.Methods 136 early pregnant patients with a history of URSA admitted to the hospital from April 2020 to March 2022 were selected and divided into miscarriage and non-miscarriage groups according to follow-up pregnancy outcome.Ultrasound multimodal scores and sFlt-1 and D-D levels were compared between the 2 groups.The risk factors for recurrent miscarriage in URSA pregnancies were analysed withusing multifactorial Logistic regression analysis.The risk prediction model was established and validated and evaluated.Results The miscarriage rate was 38.24%in 136 pregnant women after the fertility preservation treatment.The miscarriage group had a higher proportion of the patients aged≥35 years,history of alcohol consumption,family history of URSA,spontaneous abortion≥4 times and serum sFlt-1 and D-D than the non-miscarriage group,and lower progesterone and ultrasound multimodality score than the non-miscarriage group(P<0.05).Prediction model equation:Logit(P)=-1.635+0.605×spontaneous abortion≥4 times+1.332×sFlt-1+0.841×D-D+1.042×ultrasound multimodal score.Internal verification results showed that the calibration curve of the prediction model fitted well with the ideal curve(Hosmer-Lemeshowχ^(2)=0.322,P=0.113).The receiver operating characteristic(ROC)curve results showed that the Logistic prediction model had an AUC of 0.933(95%CI:0.877~0.969),which was better than the single prediction of sFlt-1,D-D and ultrasonic multimodal score(P<0.05).Conclusion Elevated serum sFlt-1 and D-D levels,reduced ultrasound multimodal scores,spontaneous abortion≥4 time are high-risk factors for recurrent abortion in URSA pregnancies,and the risk prediction model constructed has good discrimination and fit and can intuit

关 键 词：原因不明复发性流产 妊娠结局 超声多模态评分 可溶性血管内皮生长因子受体-1 D-二聚体 预测模型 

分 类 号：R714.21[医药卫生—妇产科学]