儿科患者万古霉素群体药动学的模型验证与确认

Validation of Vancomycin Population Pharmacokinetic Model for Pediatric Patients

目的选择适合国内患儿的万古霉素群体药动学(PPK)模型,用以指导其给药方案调整。方法在Pub Med和CNKI数据库中系统检索万古霉素儿科PPK文献,提取基本信息,将收集的国内患儿万古霉素治疗药物监测数据代入文献报道的最终模型中,根据模型拟合图形、可视化预测检验(VPC)及正态化预测分布误差(NPDE)等,判断已有模型对国内患儿血药浓度数据的拟合效果。结果纳入12个万古霉素儿科PPK文献,除模型4、9的拟合效果较好外,其余模型对患儿数据的拟合效果均不佳。结论国内患儿数据与国外文献报道的大多数儿科万古霉素PPK模型拟合效果不佳;模型4和9拟合效果相对较好,可以参考用于国内患儿PPK参数的计算等;为研究国内患儿万古霉素的药动学特征,有必要建立针对国内患儿的万古霉素PPK模型。

OBJECTIVE To choose a PPK model of vancomycin most suitable for Chinese pediatric patients, in order to guide the dosage adjustment. METHODS Based on the database of PubMed and CNKI, all studies regarding vancomyein population pharmaeo- kinetics were investigated and their basic information including PPK models was extracted. The data of Chinese pediatric patients who were administered vancomycin and received therapeutic drug monitoring （TDM） were introduced into the reported final PPK models, and the fitting was conducted by model fitting graphics. RESULTS Twelve vancomycin PPK studies during 1986 -2014 were includ- ed. As judged by the correlation coefficient （R） and R-square（R2） between predicted concentration and measured concentration, the models 4 and 9 presented a relatively better fitting with the data of Chinese pediatric patients we collected. The VPC fitting demonstra- ted that number 1 model achieved the best fitting. However, because the limited data used in this study was based on irregular sampling time, so the VPC test resuhs were difficult to be distinguished and could only be used as a secondary reference. On the other hand, NPDE has corresponding statistical test, and its evaluation ability for the model is not affected by the factors of the experimental design. NPDE analysis showed that one-compartment model was better than two-compartment model, and model 4 and model 9 achieved better fitting to the collected data than others. CONCLUSION The fitting effects of most reported vancomycin PPK models, except individu- al models, were poor for the TDM data of Chinese pediatric patients, therefore it is necessary to establish a vancomycin population pharmacokinetic model particularlly for Chinese pediatric patients, in order to guide dosage adjustment more accurately.