欢迎访问《中国临床药理学与治疗学》杂志官方网站,今天是

中国临床药理学与治疗学 ›› 2005, Vol. 10 ›› Issue (11): 1279-1289.

• 研究原著 • 上一篇    下一篇

用 NONMEM 法建立中国癫痫儿童丙戊酸钠的群体药动学/药效学结合模型

姜德春, 王丽, 卢炜1   

  1. 北京大学第一医院儿科;
    1北京大学药学院,北京 100034
  • 收稿日期:2005-07-01 修回日期:2005-10-14 出版日期:2005-11-26 发布日期:2020-11-12
  • 通讯作者: 丽,女, 博士, 教授, 博士生导师, 从事儿科学和神经药理学研究。Tel:010-66551122-5303 E-mail:wangli5000@sina.com
  • 基金资助:
    国家自然科学基金资助项目(No30440076)

Setting uPpopulation pharmacokinetics/pharmacodynamics model of VPA n children with epilepsy by NONMEM software

JIANG De-chun, WANG Li, LU Wei1   

  1. Department of Pediatricsof First Hospital;
    1Pharmacy School,Peking University,Beijing 100034,China
  • Received:2005-07-01 Revised:2005-10-14 Online:2005-11-26 Published:2020-11-12

摘要: 目的: 建立中国癫痫儿童应用丙戊酸钠(VPA)的群体药动学/药效学(PPK/PD)结合模型,为设计个体化用药方案奠定基础。方法: 回顾性收集246例癫痫患儿应用VPA的临床数据。血药浓度是常规监测的稳态浓度。用246例患儿的数据,通过NONMEM法已经自行成功建立PPK模型。现将246例中单用VPA的69例的数据与已经建立的PPK模型结合,建立PPK/PD模型。药效指标用癫痫发作次数减少百分比,分为5级。应用Logistic回归分析,拟合线性药效模型,用NONMEM法建立PPK/PD模型,求算血药浓度获得某一级疗效的概率。结果: 应用Logistic回归分析,拟合线性药效模型,求算出血药浓度获得某一级疗效的概率:血药浓度超过23μg·ml-1时,5级的概率小于50%,获得4、3、2级的最大概率及浓度为(30μg·ml-1,32.3%)、(50μg·ml-1,26.3%)、(65μg·ml-1,36.5%);血药浓度超过78μg·ml-1时,1级的概率大于50%;浓度为100μg·ml-1时,1级的概率约84.2%。结论: 用NON-MEM法成功地建立了中国癫痫儿童应用VPA的PPK/PD模型,定量地求出某一血药浓度获得不同疗效等级的概率。

关键词: 癫痫, 儿童, 丙戊酸钠, 群体药动学, 群体药效学, NONMEM法

Abstract: AIM: To set uPpopulation pharmacoki-netics/pharmacodynamics(PPK/PD)model of valproate(VPA)in children with epilepsy in China,and promote easonable use of antiepileptic drugs(AEDs)in clinical ractice.METHODS: Sparse data of VPA serum con-centrations from 246 pediatric children were collected.These patientswere divided into three groups:PPK-Model rouP(n =146),to calculate PPK parametervalues of PA and set uPa PPK model;PPK-Valid grouP(n =100),to valid the PPK model;and PPD grouP(n =69),to set uPPPK PD model.Based on the data of PK-Model grouPand PPK-Valid group,a PPK model of PA in children with epilepsy in China was successfully et uPby using NONMEM software by ourselves.Now,using the data of 69 patients in PPD grouPwho were on PA monotherapy and this PPK model,we set uPPPK/PD model by NONMEM software.Efficacy of epilepsy treatment was divided into 5 grades according to the per-centage of seizure frequency decreased(PSFD%):grade 1:PSFD%was 100%;grade 2:PSFD%was 75%-100%;grade 3:PSFD% was 50%-75%;grade 4:PSFD% was 25%-50%;grade 5:PSFD%was less than 25%.The quantitive relationshiPbetween the VPA serum concentrations and the probability forits efficacy score was characterized by Logistic regression analysis with NONMEM.RESULTS: Logistic regression analysis showed that,VPA serum concentrations and the proba-bility forits efficacy grades 5,4,3,2,and 1 were(23μg·ml-1,5,50%),(30μg·ml-1,4,32.3%),(50μg·ml-1,3,26.3%),(65μg·ml-1,2,36.5%),(78μg·ml-1,1,50%),and(100μg·ml-1,1,84.2%)respectively.CONCLUSION: A PPK/PD mod-el of VPA in children with epilepsy in China is success-fully established by using NONMEM software,and the probability of efficacy grade forany concentration can be calculated.It will be valuable to facilitate individualized dosage regimen.

Key words: epilepsy, children, population pharmaco-kinetics, population pharmacodynamics, valproate, NON-MEM software

中图分类号: