神经重症患者经外周静脉穿刺中心静脉导管相关性静脉血栓分割线列线图构建与验证  

Construction and verification of risk nomogram model of central venous catheter-related venous thrombosis in patients with severe neurological diseases

作　　者：刘琴 杜更胜[1] LIU Qin;DU Gengsheng(Department of Neurology,Central Hospital of Xinyu Iron and Steel Group Co.,Ltd.,Jiangxi Province,Xinyu 338000,China)

机构地区：[1]江西省新余钢铁集团有限公司中心医院神经内科,江西新余338000

出　　处：《中国当代医药》2024年第30期13-18,共6页China Modern Medicine

基　　金：江西省卫生健康委科技计划项目(202212244)。

摘　　要：目的 分析神经重症患者经外周静脉穿刺置入中心静脉导管(PICC)后发生相关性静脉血栓的影响因素,并构建风险列线图模型进行验证。方法 选取2020年10月至2023年10月在新余钢铁集团有限公司中心医院留置PICC的144例神经重症患者作为研究对象,根据置管1个月后是否发生相关性静脉血栓进行分组,分为发生相关性静脉血栓组(n=35)和未发生相关性静脉血栓组(n=109)。采用单因素及多因素logistic回归分析筛选相关因素,构建列阵图风险模型。使用受试者工作曲线(ROC)、Homster-Lemeshow拟合优度检验分别对模型进行性能评价。并另选30例患者的临床资料进行模型外部验证。结果 144例神经重症患者PICC置管后发生相关性静脉血栓的概率为24.31%。单因素分析结果提示,发生相关性静脉血栓组和未发生相关性静脉血栓组的急性生理和慢性健康状况评分(APACHE-Ⅱ)、颈内静脉置管、凝血酶原时间(PT)、D-二聚体、纤维蛋白原(FIB)比较,差异有统计学意义(P<0.05)。多因素分析结果提示,高APACHE-Ⅱ评分(β=0.272,OR=1.313,P<0.001,95%CI:1.127~1.530)、颈内静脉置管(β=1.729,OR=5.637,P=0.001,95%CI:2.072~15.341)、PT(β=0.170,OR=1.185,P=0.009,95%CI:1.043~1.348)、D-二聚体(β=2.457,OR=11.666,P=0.039,95%CI:1.130~120.416)、FIB(β=0.803,OR=2.232,P=0.001,95%CI:1.374~3.626)均是神经重症患者PICC置管后发生相关性静脉血栓的危险因素。内部验证,Hosmer-Lemeshow拟合优度检验结果显示,χ2=5.153,P>0.05,模型拟合优度良好;ROC曲线下面积为0.850(P<0.001,95%CI:0.777~0.923),灵敏度为80.00%,特异性为75.20%。外部验证,Hosmer-Lemeshow拟合优度检验结果显示,χ2=6.158,P>0.05,模型拟合优度较好;ROC曲线下面积为0.821(P<0.001,95%CI:0.668~0.974),灵敏度为81.30%,特异性为78.60%。模型预测神经重症患者PICC置管后发生相关性静脉血栓的校正曲线总体趋势与理想曲线基本吻合,表明该模型拟合优度良�Objective To analyze the influencing factors of venous thrombosis after peripherally inserted central catheter(PICC)in critically ill patients with neurological diseases,and to construct a risk nomogram model for verification.Methods From October 2020 to October 2023,144 critically ill patients with PICC in the Central Hospital of Xinyu Iron and Steel Group Co.,Ltd.were selected as the research objects.According to whether there was related venous thrombosis after 1 month of catheterization,they were divided into related venous thrombosis group(n=35)and non-related venous thrombo-sis group(n=109).Univariate and multivariate logistic regression analysis were used to screen relevant factors and construct a risk model of array diagram.The receiver operating curve(ROC)and Homster-Lemeshow goodness-of-fit test were used to evaluate the performance of the model.The clinical data of 30 patients were selected for external validation of the model.Results The probability of related venous thrombosis after PICC catheterization in 144 patients with severe neurological dis-eases was 24.31%.Univariate analysis showed that there were significant differences in acute physiology and chronic health status score(APACHE-I),internal jugular vein catheterization,prothrombin time(PT),D-dimer and fibrinogen(FIB)be-tween the related venous thrombosis group and the non-related venous thrombosis group(P<0.05).The results of multivari-ate analysis suggested.High APACHE-II score(β=0.272,OR=1.313,P<0.001,95%CI:1.127-1.530),internal jugular vein catheterization(β=1.729,0R=5.637,P-0.001,95%CI:2.072-15.341),PT(β=0.170,0R=1.185,P=0.009,95%CI:1.043-1.348),D-dimer(β=2.457,0R=11.666,P=0.039,95%CI:1.130-120.416),FIB(β=0.803,0R=2.232,P=0.001,95%CI:1.374-3.626)were risk factors for venous thrombosis after PICC catheterization in critically neurological patients.Inter-nal validation,Hosmer-Lemeshow goodness of fit test results showed that χ^(2)=5.153,P>0.05,the goodness of fit of the model was good.The area under the R0C curve was 0.850(P<0.001,95

关 键 词：神经重症 经外周静脉穿刺置入中心静脉导管 相关性静脉血栓 风险列线图模型 

分 类 号：R322.123[医药卫生—人体解剖和组织胚胎学]