目的:考察肾移植患者口服环孢菌素A（Cyclosporine A,CsA）常规监测的群体药代动力学特征,为临床调整个体用药提供新途径。方法:回顾性收集我院745例门诊肾移植病人的1468个全血稳态谷浓度样本,用NONMEM程序分析其群体药物动力学模型,估算米-曼氏模型参数Km和Vm,Km,Vm的个体间变异,以及个体自身变异采用指数模型。定量地考察日剂量、术后时间、体重、环孢菌素A剂型和合并用药等固定效应对药物动力学参数的影响。结果:环孢菌素A的监测值为稳态谷浓度时,其药代动力学能用米-曼氏模型较好地拟合,用建模组数据获得的群体参数在验证组中有较理想的拟合优度。最终群体药代动力学参数为:群体标准值Vm为601mg·L-1,Km为2120mg·L-1。Vm的结构模型参数:术后时间和体重用指数模型,系数分别为-0.210和0.139;性别和剂型用乘法模型,男性较女性的校正系数为1.078,胶囊剂较水剂的校正系数为0.927。Km的结构模型参数:剂量和术后时间用指数模型,系数分别为-1.67和-0.651;合并应用雷公藤多甙片、保肾片用乘法模型,系数分别为1.170、1.130。每日剂量预测值与观察值的平均绝对百分误差为7.72±6.60％。结论:剂量和术后时间对米-曼模型的参数拟合有非常明显的影响。本文估算的参数误差在相对较小的范围内,模型拟合情况较好,可 OBJECTIVE: To investigate the population pharmacokinetic characteristics of oral cyclosporine A (CsA) administered routinely in renal transplant recipients and to provide a new approach for the clinical adjustment of individual medication. Methods: We retrospectively collected 1468 whole blood steady-state trough concentration samples from 745 outpatients with renal transplant in our hospital. The population pharmacokinetic model was analyzed by NONMEM program to estimate the parameters of the M-Man model Km and Vm, Km, Vm Inter-individual variation, as well as the individual’s own variation using exponential model. Quantitative study of daily dose, postoperative time, body weight, cyclosporin A dosage form and combination of fixed effects on pharmacokinetic parameters. RESULTS: The cyclosporine A was monitored at steady-state trough concentrations and its pharmacokinetics was better fitted with the Mickman’s model. The population parameters obtained with the model group data were statistically significant among the validation groups Ideal fit goodness. The pharmacokinetic parameters of the final population were as follows: population standard value Vm 601mg · L-1, Km 2120mg · L-1. Vm structural model parameters: postoperative time and body weight with an exponential model, the coefficients were -0.210 and 0.139; gender and dosage form multiplication model, male than female calibration factor was 1.078, capsule correction agent was 0.927. Km’s structural model parameters: dose and postoperative time using an exponential model, the coefficients were -1.67 and -0.651; combined application of Tripterygium polygalae tablets, kidney tablets with a multiplication model, the coefficients were 1.170,1.130. The mean absolute percentage error between the daily dose forecast and the observed value was 7.72 ± 6.60%. CONCLUSIONS: Dose and postoperative time have a very significant effect on the parameter fitting of the Mi-Man model. The error of the parameter estimated in this paper is in a relatively small range, the model fitting is better