乳腺癌根治术化疗患者PICC相关性感染风险评估及模型构建  

Risk assessment of PICC-associated infection and model construction in patients undergoing radical mastectomy and chemotherapy

作　　者：姬粉芝[1] 张志伟 徐艳[3] JI Fen-zhi;ZHANG Zhi-wei;XU Yan(Department of Infection,the First Affiliated Hospital of Henan University of Science and Technology,Luoyang,Henan 471000;Department of Internal Medicine,Xuchang No.5 People's Hospital,Xuchang,Henan 461000;Department of Obstetrics and Gynecology,Luoyang 202 Hospital,China Air-to-Air Missile Research Institute,Luoyang,Henan 471003,China)

机构地区：[1]河南科技大学第一附属医院感染科,河南洛阳471000 [2]许昌市第五人民医院内四科,河南许昌461000 [3]中国空空导弹研究院洛阳二〇二医院妇产科,河南洛阳471003

出　　处：《热带医学杂志》2021年第11期1433-1436,1448,共5页Journal of Tropical Medicine

基　　金：河南省医学科技攻关计划项目(201701027)。

摘　　要：目的探讨乳腺癌根治术化疗患者经外周静脉留置中心静脉导管(PICC)相关性感染风险及模型构建。方法回顾分析2014年1月-2018年4月于河南科技大学第一附属医院感染科行乳腺癌根治术化疗PICC置管80例患者临床资料,根据患者有无PICC相关性感染分为感染组(n=19)与非感染组(n=61),采用Cox比例风险模型分析乳腺癌患者PICC相关性感染的危险因素并建立风险评估模型,根据风险评估模型描绘出受试者工作特征曲线(ROC),评价PICC相关性感染风险模型的预测价值。结果 80例患者中出现PICC相关性感染19例,感染率为23.75%;PICC相关性感染单因素分析结果显示两组患者年龄、高血压比率、糖尿病比率、化疗次数、敷料更换频率、导管留置时间、WBC水平比较,差异有统计学意义(χ^(2)=4.484、4.373、5.302、4.935、5.864、5.535、5.825,均P<0.05);Cox多因素回归分析结果显示,导管留置时间≥6个月(HR=1.846)、化疗次数≥5次(HR=1.817)、敷料更换频率≥4 d/次(HR=1.655)、WBC<4×109/L(HR=1.631)、糖尿病(HR=1.718)是乳腺癌根治术化疗患者PICC相关性感染的危险因素(P<0.05);PICC相关性感染风险评分模型为1.846χ(导管留置时间)+1.817χ(化疗次数)+1.655χ(敷料更换频率)+1.631χ(WBC)+1.718χ(糖尿病),风险评分的ROC曲线面积为0.815,95%CI为0.713~0.893,差异有统计学意义(P<0.001)。结论乳腺癌根治术化疗患者PICC相关性感染的危险因素为导管留置时间、化疗次数、敷料更换频率、糖尿病以及WBC,PICC相关性感染风险评估模型具有较好预测价值。Objective To explore the risks of peripherally inserted central catheter(PICC)-associated infection and model construction in patients undergoing radical mastectomy and chemotherapy.Methods The clinical data of 80 PICC catheterization patients who underwent radical mastectomy and chemotherapy in the department from January 2014 to April2018 were retrospectively analyzed.According to presence or absence of PICC-associated infection,they were divided into infection group(n=19)and non-infection group(n=61).Cox proportional hazards model was applied to analyze risk factors of PICC-associated infection in breast cancer patients.And risk assessment model was established.According to risk assessment model,receiver operating characteristic curves(ROC curves) were drawn.The predictive value of PICC-associated infection risk model was evaluated.Results Of the 80 patients,there were 19 cases with PICC-associated infection,with infection rate of 23.75%.The univariate analysis of PICC-associated infection showed that there were significant differences in age,hypertension rate,diabetes rate,times of chemotherapy,frequency of dressing changing,catheter indwelling time and WBC level between the two groups(χ^(2)=4.484,4.373,5.302,4.935,5.864,5.535,5.825,P<0.05).The results of Cox multivariate regression analysis showed that catheter indwelling time ≥6 months(HR=1.846),times of chemotherapy ≥5 times(HR=1.817),frequency of dressing changing ≥4 d/once(HR=1.655),WBC <4×109/L(HR=1.631) and diabetes(HR=1.718) were risk factors of PICC-associated infection in patients undergoing radical mastectomy and chemotherapy(P<0.05).The risk score of PICC-associated infection was as follows:1.846χ(catheter indwelling time)+ 1.817χ(times of chemotherapy)+ 1.655χ(frequency of dressing changing)+ 1.631χ(WBC)+ 1.718χ(diabetes).The area under ROC curve and 95%CI of the risk score were 0.815 and 0.713-0.893,respectively(P<0.001).Conclusions The risk factors of PICC-associated infection in patients undergoing radical mastectomy and chemother

关 键 词：乳腺癌 经外周静脉留置中心静脉导管 感染 风险模型 

分 类 号：R737.9[医药卫生—肿瘤]