乳腺癌病人静脉血栓栓塞症风险预测模型的构建  

Construction of risk prediction model of venous thromboembolism in breast cancer patients

作　　者：李静[1] 强万敏[1] 张方圆 李笑阳 任佳蕾 LI Jing;QIANG Wanmin;ZHANG Fangyuan;LI Xiaoyang;REN Jialei(Tianjin Medical University Cancer Institute&Hospital,Tianjin 300060 China)

机构地区：[1]天津医科大学肿瘤医院,天津300060 [2]大连大学 [3]天津中医药大学

出　　处：《护理研究》2020年第16期2852-2857,共6页Chinese Nursing Research

摘　　要：[目的]构建乳腺癌病人静脉血栓栓塞症风险预测模型,为临床预防血栓发生提供科学、客观及量化的参考工具。[方法]通过Meta分析获得乳腺癌病人静脉血栓栓塞症的危险因素及相应的合并风险值,构建Logistic回归风险预测模型,通过逻辑评分法转换为风险评分模型并进行危险层级划分。[结果]最终构建了包含体质指数(BMI)≥30 kg/m2、高血压、糖尿病、冠心病、既往血栓史、浸润性导管癌、乳腺手术、化疗、他莫昔芬类药物和中心静脉置管10个危险因素在内的个体化风险预测模型,总分为0~21分,≥13分判定为高危,9~<13分为中危,<9分为低危。[结论]基于Meta分析建立的具有循证基础的乳腺癌病人静脉血栓栓塞症的风险预测模型的稳定性和适用性有待临床实践进行外部验证。Objective:To construct a risk prediction model of venous thromboembolism(VTE)in breast cancer patients,so as to provide a scientific,objective,and quantitative reference tool for clinical prevention of thrombosis.Methods:Risk factors of breast cancer patients complicated with VTE and corresponding merge risk values were obtained by meta analysis. Logistic regression analysis was used to construct the risk prediction model,which was converted into a risk scoring model by logical scoring method. And the risk hierarchy was determined.Results:The individualized risk prediction model included 10 risk factors,as follows:BMI≥30 kg/m2,hypertension,diabetes,coronary disease,history of thrombosis,invasive ductal carcinoma,mastectomy,chemotherapy,drugs like tamoxifen,and central vena catheterization. The overall score of model was from 0 to 21:high risk with the score≥13;middle risk with score<13 and ≥9,low risk with score<9.Conclusions:The robustness and generalizability of evidence-based risk prediction model of venous thromboembolism among breast cancer patients that constructed by using Meta-analysis,which still need to be taken an external validation through the clinical practice.

关 键 词：乳腺癌 静脉血栓栓塞症 风险预测模型 危险因素 系统评价 

分 类 号：R473.73[医药卫生—护理学]