预测肿瘤患者PICC相关性上肢静脉血栓风险诺谟图模型的外部验证  

External validation of nomogram for predicting risk of PICC-related upper extremity venous thrombosis for cancer patients

作　　者：杨方英[1] 吴红娟[1] 华荣誉 傅林娟[1] 江子芳[1] 应丽[1] 陈丽羽[1] 刘丽华[1] Yang Fangying;Wu Hongjuan;Hua Rongyu;Fu Linjuan;Jiang Zifang;Ying Li;Chen Liyu;Liu Lihua(Zhejiang Cancer Hospital,Hangzhou Zhejiang 310022,China)

机构地区：[1]浙江省肿瘤医院,浙江杭州310022

出　　处：《护理与康复》2021年第1期8-12,共5页Journal of Nursing and Rehabilitation

基　　金：浙江省医药卫生平台计划重点资助A类项目,编号2016ZDA003。

摘　　要：目的验证预测肿瘤患者PICC相关性上肢静脉血栓风险诺谟图模型的准确性及临床效益,为临床应用提供理论支撑。方法采用前瞻性研究收集数据,分析行PICC置管的790例肿瘤患者临床资料,应用构建的诺谟图对患者进行上肢静脉血栓风险预测,模型预测准确性采用受试者工作特征曲线下面积检验,模型临床应用价值采用决策曲线分析。结果诺谟图模型预测准确度为63.67%,诺谟图的受试者工作特征曲线下面积为0.645,95%CI为0.573~0.717,显示模型具有一定的预测准确度;决策曲线分析显示模型有一定的临床获益。结论预测肿瘤患者PICC相关性上肢静脉血栓风险模型具有一定的准确性和临床应用价值,可为医护人员或患者提供决策参考。Objective To validate the accuracy and clinical value of nomogram for predicting risk of PICC-related upper extremity venous thrombosis for cancer patients to provide reference for clinical application.Method Collect the information by prospective study and analyze the clinical information of 790 cancer patients undergoing PICC catheterization.Predict the risk of upper extremity venous thrombosis by nomogram constructed.The accuracy is tested by area under receiver operating characteristic curve.The clinical value of model is analyzed by decision curve.Result The prediction accuracy of nomogram model is 63.67%.The area under receiver operating characteristic curve of nomogram is 0.645.The 95%CI is 0.573~0.717,which shows that the accuracy of model can be confirmed.The decision curve proves the clinical value of the model.Conclusion The nomogram model for predicting risk of PICC-related upper extremity venous thrombosis has a certain accuracy and clinical value,which can provide reference for medical staff and patients.

关 键 词：肿瘤 经外周静脉置入中心静脉导管 血栓 诺谟图 

分 类 号：R473.73[医药卫生—护理学]