摘要
目的:采用Logistic回归分析联合受试者操作特征(receiver operating characteristic,ROC)曲线分析头孢哌酮-舒巴坦钠导致患者凝血功能异常的影响因素,并建立模型预测引起凝血功能异常的发生风险。方法:收集2019年1月1日—2021年8月31日梧州市红十字会医院使用头孢哌酮-舒巴坦钠的139例患者的病历资料,采用Logistic回归分析联合ROC曲线分析头孢哌酮-舒巴坦钠导致凝血功能异常的影响因素,并建立模型进行凝血功能异常的预测分析。结果:139例患者中62例(44.60%)发生了凝血功能异常;Logistic回归分析显示,头孢哌酮日剂量、患者肌酐值和禁食状态是发生凝血功能异常的独立危险因素;分别用上述独立危险因素建立Logistic回归方程,经变换后得到联合预测因子计算公式,Y_(联合)=45.15 X_(禁食)+27.23 X_(日剂量)+0.1 X_(Cr);联合预测因子的ROC曲线下面积最大为0.760(95%CI为0.679~0.841,P<0.05);约登指数0.459对应的切点为ROC曲线上的最佳临界值(91.89)。结论:当患者接受头孢哌酮-舒巴坦钠治疗时,可将头孢哌酮日剂量、患者肌酐值和禁食状态代入联合预测因子计算公式,预测患者发生凝血功能异常的风险,以保证患者的用药安全。
Objective:To analyze the influencing factors of coagulation disorders caused by cefoperazone/sulbactam with Logistic regression analysis combined with receiver operating characteristic(ROC)curve,and establish a model to predict the risks of coagulation disorders.Methods:The medical records of 139 patients treated with cefoperazone/sulbactam in Wuzhou Red Cross Hospital from January 1,2019 to August 31,2021 were collected,and the influencing factors of coagulation disorders caused by cefoperazone/sulbactam were analyzed by Logistic regression analysis combined with ROC curve.The model was established to predict the coagulation disorders.Results:Among 139 patients,62 cases(44.60%)had the coagulation disorders.Logistic regression analysis showed that daily dose of cefoperazone,patient’s creatinine level and fasting status were the independent risk factors for coagulation disorders.The Logistic regression equation was established with the above independent risk factors,and the calculation formula of combined predictors was obtained after transformation,Y_(combined)=45.15 X_(fasting)+27.23 X_(daily dose)+0.1 X_(Cr).The maximum area under the ROC curve of the combined predictors was 0.760(95%CI 0.679~0.841,P<0.05).The cut-off point was the optimal value(91.89)on the ROC curve when Youden index was 0.459.Conclusion:When patients receive cefoperazone/sulbactam treatment,the daily dose of cefoperazone,patient’s creatinine value and fasting status can be substituted into the combined predictor formula to predict the risk of coagulation disorders in patients so as to ensure the safety of medication.
作者
谭皓文
TAN Hao-wen(Pharmacy Department.Wuzhou Red Cross Hospital,Wuzhou Guangxi 543002,China)
出处
《抗感染药学》
2022年第2期300-304,共5页
Anti-infection Pharmacy