基于群体药动学的拉莫三嗪个体化剂量调整研究  

Study on Individualized Dosage Adjustment of Lamotrigine Based on Population Pharmacokinetics

作　　者：尚德为[1] 朱秀清[1] 陈伟家 李璐[1] 王占璋[1] 邓书华[1] 卢浩扬[1] 胡晋卿[1] 张明[1] 倪晓佳[1] 温预关[1] SHANG Dewei;ZHU Xiuqing;CHEN Weijia;LI Lu;WANG Zhanzhang;DENG Shuhua;LU Haoyang;HU Jinqing;ZHANG Ming;NI Xiaojia;WEN Yuguan(Dept.of Pharmacy,the Affiliated Brain Hospital of Guangzhou Medical University/Guangzhou Hui＇ai Hospital,Guangdong Guangzhou 510370,China)

机构地区：[1]广州医科大学附属脑科医院/广州市惠爱医院药学部,广东广州510370

出　　处：《中国医院用药评价与分析》2018年第7期867-869,共3页Evaluation and Analysis of Drug-use in Hospitals of China

基　　金：国家自然科学基金青年基金(No.81403016);广州市医药卫生科技项目一般引导项目(No.20171A010276);广东省医院药学研究基金(正大天晴基金)(No.2017A12)

摘　　要：目的:基于群体药动学方法,预测患者不同用药方案服用拉莫三嗪后的稳态血药浓度,为临床治疗提供参考。方法:收集58例患者的282个拉莫三嗪血药浓度观测值,并记录联合应用丙戊酸的情况,利用已经建立的拉莫三嗪群体药动力学模型,分别通过第1个、前2个血药浓度值估算个体间变异,并预测后续浓度值,评估两种情况对模型预测准确性的影响。结果:预测结果显示,分别采用第1个、前2个血药浓度值预测产生的误差的中位值分别为-0.32、-0.41μg/ml,均接近0值;得到的平均预测误差分别为-0.66、-0.42μg/ml,均方根误差分别为3.80、3.19μg/ml,说明在上述两种情况下,模型均有较好的预测性。结论:无论从采集1个浓度值还是2个浓度值时开始使用该模型进行预测,均能得到较准确的预测结果。OBJECTIVE： Based on population pharmacokinetics,to predict the steady-state plasma concentrations of lamotrigine（ LTG） in patients with different drug regimens and provide reference for clinical treatment. METHODS：58 patients were enrolled, with 282 LTG concentrations and corresponding comedication of VPA records. An established population pharmacokinetic model was applied to predict the following up concentrations of LTG based on the inter-subject variabilities estimated from the first observation in each patient or the first two observations in each patient. The precision and accuracy of the prediction were evaluated under the two scenarios. RESULTS： According to prediction results,the median prediction errors of the two scenarios were respectively -0. 32 μg/ml and -0. 41 μg/ml,which were close to zero; the mean prediction error（ MPE） were -0. 66 μg/ml and -0. 42 μg/ml,and root mean squared error（ RMSE） were 3. 80 μg/ml and 3. 19 μg/ml,respectively. Thus,the model exhibited good prediction efficiency under both scenarios. CONCLUSIONS： The results demonstrated that no matter based on one or two observations,the model would both show accurate prediction.

关 键 词：拉莫三嗪 治疗药物监测 群体药动学 仿真预测 个体化剂量调整 

分 类 号：R971.6[医药卫生—药品]