[1]牛敬荣,张丽金,杨文爽.绝经期非肿瘤女性发生静脉血栓栓塞症的危险因素分析及预测模型构建[J].陆军军医大学学报(原第三军医大学学报),2020,42(22):2224-2230.
 NIU Jingrong,ZHANG Lijin,YANG Wenshuang.Risk factors of venous thromboembolism in postmenopausal women without neoplasms and establishment of the prediction model[J].J Amry Med Univ (J Third Mil Med Univ),2020,42(22):2224-2230.
点击复制

绝经期非肿瘤女性发生静脉血栓栓塞症的危险因素分析及预测模型构建(/HTML )
分享到:

陆军军医大学学报(原第三军医大学学报)[ISSN:1000-5404/CN:51-1095/R]

卷:
42卷
期数:
2020年第22期
页码:
2224-2230
栏目:
临床医学
出版日期:
2020-11-30

文章信息/Info

Title:
Risk factors of venous thromboembolism in postmenopausal women without neoplasms and establishment of the prediction model
作者:
牛敬荣张丽金杨文爽
首都医科大学附属北京朝阳医院血管外科
 
Author(s):
NIU Jingrong ZHANG Lijin YANG Wenshuang

Department of Vascular Surgery, Beijing Chaoyang Hospital, Capital Medical University, Beijing, 100020, China

关键词:
绝经期静脉血栓栓塞症危险因素预测模型
Keywords:
menopause venous thromboembolism risk factors prediction model
分类号:
R181.23; R364.15; R711.51
文献标志码:
A
摘要:

目的探究绝经期非肿瘤女性发生静脉血栓栓塞症(venous thromboembolism, VTE)的危险因素,并构建预测模型。方法选取2018年5月至2019年12月我科收治的268例绝经期非肿瘤女性疑似VTE的患者作为研究对象。以彩色多普勒超声确诊是否发生VTE分为VTE组(n=42)和非VTE组(n=226)。对两组的临床资料进行单因素分析和多因素Logistic回归分析,并建立预测模型,用灵敏度、特异度、ROC曲线下面积及预测正确率评价模型的预测效果。结果经多因素Logistic 回归分析发现合并糖尿病(OR值=15.318,95%CI=6.563~74.275,P=0.036)、Caprini评分≥5分(OR值=7.737,95%CI=2.459~13.168,P=0.012)、D二聚体>0.5 mg/L(OR值=10.602,95%CI=3.767~17.706,P=0.000)、三酰甘油(Triglyceride,TG)>1.7 mmol/L(OR值=2.651,95%CI=1.238~5.357,P=0.002)、黄体生成素(Luteinizing Hormone,LH)>52.3 U/L(OR值=1.539,95%CI=1.127~4.673,P=0.031)、雌二醇(Estrodiol,E2)<30 nmol/L(OR=1.164,95%CI=1.064~2.058,P=0.047)为绝经期非肿瘤女性发生VTE的危险因素。根据危险因素得出预测模型:Prob=1/(e^Y),Y=36.5782.729×合并糖尿病2.046×Caprini评分2.361×血清D二聚体0.975×血清TG0.431×血清LH0.152×血清E2。模型经ROC曲线分析得AUC为0.915,灵敏度为90.30%,特异度为94.10%。通过验证得出模型预测的正确率为88.43%。结论合并糖尿病、Caprini评分≥5分、D二聚体>0.5 mg/L、TG>1.7 mmol/L、LH>52.3 U/L、E2<30 nmol/L是绝经期非肿瘤女性发生VTE的危险因素,根据危险因素建立预测模型能有效评估绝经期非肿瘤女性VTE的发生风险。

Abstract:

ObjectiveTo explore the risk factors and establish a prediction model of venous thromboembolism (VTE) in postmenopausal women without neoplasms. MethodsA total of 268 postmenopausal women with suspected VTE without neoplasms admitted to our department from May 2018 to December 2019 were divided into VTE group (n=42) and nonVTE group (n=226) based on color Doppler ultrasound findings. The clinical data of the 2 groups were analyzed by univariate and multivariate logistic regression analysis to identify the risk factors of VTE. The prediction model was established based on the identified risk factors and evaluated for its sensitivity, specificity, area under the receiveroperating characteristic (ROC) curve, and the prediction accuracy. ResultsMultivariate logistic regression analysis identified diabetes mellitus (OR=15.318, 95%CI: 6.56374.275,P=0.036), a Caprini score ≥5 (OR=7.737, 95%CI: 2.45913.168, P=0.012), Ddimer >0.5 mg/L (OR=10.602, 95%CI: 3.76717.706, P=0.000), triglycerides (TG)>1.7 mmol/L (OR=2.651, 95%CI: 1.2385.357, P=0.002), luteinizing hormone (LH)>52.3 IU/L (OR=1.539, 95%CI: 1.1274.673, P=0.031), and estradiol (E2)<30 nmol/L (OR=1.164, 95%CI: 1.0642.058, P=0.047) as the risk factors for VTE in postmenopausal women without neoplasms. Based on these risk factors, the prediction model for VTS was derived: Prob=1/(e^Y), where Y=36.5782.729×combined diabetes2.046×Caprini score2.361×serum Ddimer0.975×serum TG0.431×serum LH0.152×serum E2. ROC curve analysis showed that this prediction model had an AUC of 0.915 with a sensitivity of 90.30% and a specificity of 94.10%. The verification results showed that this model had an accuracy of 88.43% for predicting VTE in postmenopausal women without neoplasms. ConclusionDiabetes, a Caprini score ≥5, a Ddimer level >0.5 mg/L, a TG level >1.7 mmol/L, a LH level >52.3 U/L, and an E2 level <30 nmol/L are all risk factors of VTE in postmenopausal women without neoplasms. The predictive model constructed based on these risk factors can effectively assess the risk of VTE in these women.

更新日期/Last Update: 2020-11-19