替考拉宁腹腔给药治疗腹透相关性腹膜炎患者治疗失败风险预测模型  

Predictive modeling of the risk of treatment failure in patients with peritoneal dialysis-associated peritonitis treated with teicoplanin administered intraperitoneally

作　　者：雷雯朴 范蒋青 王路路[1] 张晋萍 LEI Wenpu;FAN Jiangqing;WANG Lulu;ZHANG Jinping(Department of Pharmacy,China Pharmaceutical University Nanjing Drum Tower Hospital Jiangsu,Jiangsu Nanjing 210008,China;School of Basic Medicine and Clinical Pharmacy,China Pharmaceutical University,Jiangsu Nanjing 210009,China)

机构地区：[1]中国药科大学南京鼓楼医院药学部,江苏南京210008 [2]中国药科大学基础医学与临床药学学院,江苏南京210009

出　　处：《中国医院药学杂志》2023年第18期2071-2075,2099,共6页Chinese Journal of Hospital Pharmacy

摘　　要：目的:建立并验证替考拉宁腹腔给药治疗腹膜透析相关性腹膜炎患者治疗失败的风险预测模型。方法:回顾性分析2018年8月至2022年8月于南京鼓楼医院使用替考拉宁腹腔给药治疗腹透相关性腹膜炎(peritoneal dialysis-associated peritonitis,PDAP)的患者人口学特征及临床资料,按照约3∶1的比例将患者分为训练集与验证集。Logistic回归分析确定导致治疗失败的影响因素并建立风险预测模型,用ROC曲线评估模型区分度,校准曲线(Hosmer-Lemeshow检验)评估模型准确度,通过验证集患者临床资料再次验证实际的准确度。结果:共纳入144例患者,训练集112例次,其中有33例治疗失败,腹膜透析龄和纤维蛋白原是治疗失败的独立危险因素。据此建立logistic回归方程Z=-3.455+0.018X_(腹膜透析龄)+0.825X_(纤维蛋白原),ROC曲线下面积为0.784(95%CI:0.667~0.902,P<0.01),当Z值的Youden指数最大(0.482)时,其在ROC曲线上最佳截断值为0.2711,验证集33例患者验证该模型准确率为87.5%。结论:基于腹膜透析龄、纤维蛋白原构建的风险预测模型性能良好,准确率为87.5%,对腹腔使用替考拉宁的PDAP患者临床结局有一定预测价值。OBJECTIVE To establish and validate a risk prediction model for treatment failure in patients with peritoneal dialysis-associated peritonitis treated with ticoranine administered intraperitoneally.METHODS Demographic characteristics and clinical data of patients treated with ticoranine intraperitoneally administered for PDAP at Nanjing Gulou Hospital from August 2018 to August 2022 were retrospectively analyzed,and patients were divided into a training set and a validation set according to a ratio of approximately 3:1.Logistic regression analysis was used to determine the independent risk factors leading to treatment failure and to establish a risk prediction model.The ROC curve was used to assess the model discrimination,the calibration curve(Hosmer-Lemeshow test)to assess the model accuracy,and the actual accuracy of the model was again verified by the validation set patient data.RESULTS A total of 144 patients were included,with 112 training episodes,of which 33 treatment failures were observed,with peritoneal dialysis age and fibrinogen being independent risk factors for treatment failure.Accordingly,a logistic regression equation Z=-3.455+0.018X_(peritoneal dialysis age)+0.825X_(fibrinogen)was established,and the area under the ROC curve was 0.784(95%CI:0.667~0.902,P<0.01),and its optimal cut-off value on the ROC curve when the Youden index of Z value was the largest(0.482)was 0.2711,and the accuracy of the model was 87.5%in the validation set of 33 patients.CONCLUSION The risk prediction model constructed based on the age of peritoneal dialysis and fibrinogen performed well with an accuracy of 87.5%and had some predictive value for clinical outcomes in PDAP patients on intraperitoneal teicoplanin.

关 键 词：替考拉宁 腹透相关性腹膜炎 危险因素 预测模型 

分 类 号：R969[医药卫生—药理学]