利奈唑胺致重症患者血红蛋白减少的预测模型研究  

A predictive model study of hemoglobin reduction induced by linezolid in critical patients

作　　者：杨万枝[1] 胡章海 居靖[2] YANG Wan-zhi;HU Zhang-hai;JU Jing(Department of Pharmacy,Anqing First People's Hospital of Anhui Medical University,Anqing ANHUI 246003,China;Department of Equipment,Anqing Hospital Affiliated to Anhui Medical University,Anqing ANHUI 246001,China)

机构地区：[1]安徽医科大学附属安庆第一人民医院药学部,安徽安庆246003 [2]安徽医科大学附属安庆医院设备科,安徽安庆246001

出　　处：《中国新药与临床杂志》2022年第3期175-180,共6页Chinese Journal of New Drugs and Clinical Remedies

基　　金：安庆市第一人民医院2021年“三新”立项项目(Y2021030)。

摘　　要：目的分析利奈唑胺致重症患者血红蛋白(Hb)减少的危险因素,建立预测模型并验证准确率。方法选取使用利奈唑胺治疗的93例重症患者,回顾性分析患者的临床资料,包括性别、年龄、用药时间、用药前实验室检查指标、合并疾病与联合用药等,采用单因素方差分析和多因素logistic回归分析确定利奈唑胺相关性Hb减少的危险因素,并建立预测模型。采用受试者工作特征(ROC)曲线分析评价预测模型,另外选取使用利奈唑胺治疗的31例重症患者验证该预测模型的准确率。结果使用利奈唑胺治疗的93例重症患者中有16例(17%)发生Hb减少。前白蛋白降低、丙氨酸转氨酶(ALT)升高和肾小球滤过率(eGFR)降低是发生利奈唑胺相关性Hb减少的独立危险因素。综合三种危险因素建立logistic回归模型logit(P)=1.183+0.004X_(ALT)-0.014X_(前白蛋白)-0.018X_(eGFR),并得到联合预测因子Y_(联合)=X_(ALT)-3.5X_(前白蛋白)-4.5X_(eGFR),构建各危险因素及联合预测因子的ROC曲线,发现Y_(联合)的曲线下面积最大、Youden指数最大,其在ROC曲线对应的最佳截断值为-616.79。将符合纳入标准的31例患者资料对模型进行外部验证,预测准确率为81%。结论前白蛋白降低、ALT升高和e GFR降低是利奈唑胺致重症患者发生Hb减少的独立危险因素,三者拟合建立预测模型有一定价值,准确率较高。AIM To analyze the risk factors of hemoglobin reduction induced by linezolid,then establish a predictive model by the risk factors and verify the accuracy of the model.METHODS Clinical data of 93 critical patients treated with linezolid were selected and analyzed retrospectively,including gender,age,duration of medication,premedication laboratory examination indexes,combined disease and co-medication,etc.Univariate analysis of variance and multivariate logistic regression analysis were used to determine the risk factors of hemoglobin reduction induced by linezolid,and the predictive model was established.Then the receiver operating characteristic(ROC)curve was used to evaluate the model,and clinical data of another 31 critical patients treated with linezolid were selected to verify the accuracy of the predictive model.RESULTS Hemoglobin reduction occurred in 16(17%)of 93 critical patients treated with linezolid.Decreased prealbumin,increased alanine aminotransferase(ALT),and decreased glomerular filtration rate(eGFR)were independent risk factors of the occurrence of hemoglobin reduction.The logistic regression model was established by integrating the three risk factors:logit(P)=1.183+0.004X_(ALT)-0.014X_(prealbumin)-0.018X_(eGFR),and the combined predictor was obtained:Y_(combined)=X_(ALT)-3.5X_(prealbumin)-4.5X_(eGFR).By constructing ROC curves of the three risk factors and Y_(combined),it was found that the area under the ROC curve and Youden index of Y_(combined) were all the largest,and the best cut-off point of the ROC was-616.79.The data of 31 patients who met the inclusion criteria were used to externally verify the model,and the predictive accuracy was 81%.CONCLUSION Decreased prealbumin,increased ALT and decreased eGFR were the independent risk factors of hemoglobin reduction induced by lineczolid in critical patients.The predictive model based on the combination of the three factors has a certain value with high accuracy.

关 键 词：利奈唑胺 危重病人医疗 血红蛋白类 预测模型 

分 类 号：R969[医药卫生—药理学]