深静脉置管患者导管相关性感染的危险因素及其预测模型构建

Risk factors for catheter-related infections in patients with deep venous cannulae and construction of their prediction model

目的为探讨深静脉置管患者导管相关性感染(CRI)的危险因素并构建其预测模型。方法选取宁波大学医学院附属医院2019年9月-2022年9月接受深静脉置管的109例患者的临床资料进行回顾性分析,根据患者CRI情况分为合并CRI组(27例)和无CRI组(82例);统计合并CRI组病原菌情况,归纳深静脉置管患者发生CRI的危险因素,并绘制受试者工作特征(ROC)曲线分析预测模型对深静脉置管患者发生CRI的预测价值,获取其曲线下面积(AUC)、敏感度、特异度。结果27例发生CRI的深静脉置管患者共检出菌株37株,其中革兰阴性菌19株占51.35%,革兰阳性菌14株占37.84%,真菌4株占10.81%;合并糖尿病、血清白蛋白(ALB)是深静脉置管患者CRI的危险因素(OR=5.043、3.873,P均<0.05);构建回归模型:logit(P)=-10.114+合并糖尿病×1.618+ALB×1.354;按照预测概率logit(P)绘制深静脉置管患者发生CRI的ROC曲线,当logit(P)>11.00时,AUC值为0.886,敏感度为92.59%,特异度为69.51%。结论深静脉置管患者发生CRI的危险因素包括合并糖尿病、血清ALB,据此构建回归模型对深静脉置管患者发生CRI具有较好的预测价值,且其感染病原菌以革兰阴性菌为主,临床可对高危人群进行针对性干预或防治以降低CRI发生风险。

OBJECTIVE To explore the risk factors for catheter-related infections(CRI)in patients with deep vein catheterization and construct their prediction model.METHODS The clinical data of 109 patients who underwent deep vein catheterization in the Affiliated Hospital of Ningbo University School of Medicine from Sep.2019 to Sep.2022 were selected for retrospective analysis,and the patients were divided into the combined CRI group(27 cases)and the no CRI group(82 cases)according to their CRI status.The pathogenic bacteria in the combined CRI group were statistically analyzed,the risk factors for the occurrence of CRI in the patients with deep vein catheterization were summarized,and the receiver operating characteristic(ROC)curves were plotted to analyze the predictive value of the prediction model for the occurrence of CRI in patients with deep vein catheterization,obtaining their area under the curve(AUC),sensitivity,and specificity.RESULTS A total of 37 strains of pathogens were detected in 27 patients with deep vein catheterization who developed CRI,of which 19 strains of gram-negative bacteria accounted for 51.35%,14 strains of gram-positive bacteria accounted for 37.84%,and 4 strains of fungi accounted for 10.81%.Comorbid diabetes mellitus and serum albumin(ALB)were risk factors for CRI in patients with deep vein catheterization(OR=5.043,3.873,P<0.05).The regression model was constructed as follows:logit(P)=-10.114+diabetes×1.618+ALB×1.354.The ROC curve for the occurrence of CRI in patients with deep vein catheterization were plotted according to the predictive probability logit(P),with an AUC value of 0.886,a sensitivity of 92.59%,and a specificity of 69.51%when logit(P)>11.00.CONCLUSIONS The risk factors of CRI in patients with deep vein catheterization included comorbid diabetes mellitus and serum ALB,according to which the construction of regression model had a good predictive value for the occurrence of CRI in patients with deep vein catheterization,and the main pathogens of infection were gram-negative bacteria.Clinically,targeted intervention or prevention could be carried out for high-risk groups to reduce the risk of the occurrence of CRI.