妊娠相关静脉血栓栓塞症评估工具预测价值的比较

Comparison of predictive performance of different risk assessment tools forpregnancyrelated venous thromboembolism

目的探讨4种不同评估工具对妊娠相关静脉血栓栓塞症发生风险的预测性能的比较。方法采用回顾性病例对照研究,选取我院2016—2020年分娩并确诊发生静脉血栓栓塞症(Venous thromboembolism,VTE)的35例患者作为研究组,并按1∶2配比选取同期同分娩方式的70例非VTE患者作为对照组;采集患者一般资料及VTE风险因素,应用Caprini、Autar、昆士兰及RCOG风险评估工具进行回顾性评估,并根据妊娠相关VTE发生情况绘制ROC曲线,以AUC、灵敏度、特异度、约登指数来综合评价各评估工具的预测性能。结果本研究人群VTE显著的风险因素为紧急剖宫产及产后输血,差异具有统计学意义(P<0.05)。各评估工具中,昆士兰评估工具测得风险程度在中危及以上水平时对妊娠相关VTE的预测具有一定的准确性,AUC为0.764,P<0.05,灵敏度80.0%,特异度72.9%,预测性能优于Caprini、Autar、RCOG等风险评估工具。结论昆士兰评估工具在妊娠相关VTE患者中预测性能更佳,易于临床使用,但在国内孕产妇人群中应用的可行性及有效性有待前瞻性、多中心大样本的临床验证。

Objective To compare the performance of four different VTE risk assessment tools.Methods A retrospective case-control study was used in this study.35 women with VTE who delivered at our institution from 2016 to 2020 were selected as the disease group.According to the matching of 1∶2,70 non-thrombotic women with the same delivery mode during the same period were selected as the control group.Antepartum,intrapartum and postpartum records were reviewed.Data were collected on known risk factors for VTEs in order to calculate scores for four risk assessment tools:Caprini,Autar,Queensland and RCOG.ROC curve was drawn according to the occurrence of VTE to evaluate the predictive performance of each assessment tool.Results The significant risk factors were emergency caesarean section and blood transfusion(P<0.05).Of the assessment tools,Queensland assessment tool showed a degree of accuracy in predicting VTE at risk levels above moderate risk,with an AUC of 0.764,P<0.05.The sensitivity and specificity were 80.0%and 72.9%respectively.Conclusion The Queensland assessment tool has better predictive performance for VTE and is easy to clinically use.However,the feasibility and effectiveness of this tool in domestic obstetric population needs to be prospectively verified by multi-center and large samples.