摘要
目的 分析非瓣膜性房颤患者服用达比加群酯后出血风险的相关因素和构建风险预测模型。方法 选取2020年6月至2023年6月九江市第一人民医院收治的187例非瓣膜性房颤患者作为研究对象,根据是否发生出血进行分组,分为出血组(n=50)和非出血组(n=137)。使用Logistic回归构建预测风险模型,受试者工作曲线(ROC)、Homster-Lemeshow拟合优度检验(H-L)对模型进行性能评价。并另选50例非瓣膜性房颤患者的临床资料进行模型外部验证。结果 187例非瓣膜性房颤患者发生出血的概率为26.74%。根据单因素和Logistic多因素分析,高达比加群酯剂量、出血风险评估(HAS-BLED)评分、活化部分凝血活酶时间(APTT)、D-二聚体(D-D)是非瓣膜性房颤患者是否发生出血的危险因素,高水平肌酐清除率(Ccr)是其保护因素。内部验证H-L检验结果显示,χ^(2)=2.23,P>0.05,模型拟合情况良好;ROC曲线下面积为0.811(P<0.001,95%CI 0.738~0.883),灵敏度为76.00%,特异性为78.10%,最大约登指数为0.541。另外选取50例患者的临床资料中,H-L检验结果显示,χ^(2)=4.97,P>0.05,模型拟合情况良好;ROC曲线下面积为0.888(P<0.001,95%CI 0.797~0.980),灵敏度为76.70%,特异性为95.00%,最大约登指数为0.717。结论 高达比加群酯剂量、HAS-BLED评分、APTT、D-D是非瓣膜性房颤患者是否发生出血的危险因素,高水平Ccr是其保护因素。其建立的风险预警模型具有较好的预测效能,可用于非瓣膜性房颤患者服用达比加群酯后发生出血风险的早期预测。
Objective To analyze the related factors of bleeding risk in patients with non-valvular atrial fibrillation after taking dabigatran etexilate and to construct a risk prediction model.Methods A total of 187 patients with non-valvular atrial fibrillation admitted to Jiujiang First People's Hospital from June 2020 to June 2023 were selected as the study subjects.They were divided into bleeding group(n=50)and non-bleeding group(n=137)according to whether bleeding occurred.Logistic regression was used to construct a predictive risk model,and receiver operating curve(ROC)and Homster-Lemeshow goodness-of-fit test(H-L)were used to evaluate the performance of the model.The clinical data of 50 patients with non-valvular atrial fibrillation were selected for external validation of the model.Results The probability of bleeding in 187 patients with non-valvular atrial fibrillation was 26.74%.According to single factor and Logistic multivariate analysis,high dose of dabigatran etexilate,HAS-BLED score,APTT and D-D were risk factors for bleeding in patients with non-valvular atrial fibrillation,and high level of creatinine clearance rate(Ccr)was a protective factor.The results of internal validation H-L test showed thatχ^(2)=2.23,P>0.05,and the model fitting was good.The area under the ROC curve was 0.811(P<0.001,95%CI 0.738-0.883),the sensitivity was 76.00%,the specificity was 78.10%,and the maximum Youden index was 0.541.In addition,the clinical data of 50 patients were selected.The results of H-L test showed thatχ^(2)=4.97,P>0.05,and the model fitting was good.The area under the ROC curve was 0.888(P<0.001,95%CI 0.797-0.980),the sensitivity was 76.70%,the specificity was 95.00%,and the maximum Youden index was 0.717.Conclusion High dose of dabigatran etexilate,HAS-BLED score,APTT and D-D are risk factors for bleeding in patients with non-valvular atrial fibrillation,and high level of Ccr is a protective factor.The established risk early warning model has good predictive efficacy and can be used for early prediction of blee
作者
朱子云
张莲红
刘宇恒
ZHU Ziyun;ZHANG Lianhong;LIU Yuheng(Jiujiang First People's Hospital,Jiujiang Jiangxi 332000,China)
出处
《药品评价》
CAS
2024年第3期348-352,共5页
Drug Evaluation
关键词
非瓣膜性房颤
达比加群酯
出血
模型预测
Non-valvular atrial fibrillation
Dabigatran etexilate
Bleeding
Model prediction
