探讨肝移植受者住院期间万古霉素谷浓度达标情况及其影响因素，评价群体药代动力学软件JPKD在肝移植受者中的应用价值。

回顾性分析2016年1月至2019年12月在复旦大学附属中山医院行肝移植住院期间使用万古霉素并接受治疗药物监测的受者临床资料。根据万古霉素初始稳态谷浓度是否达到10～20 mg/L，将受者分为达标组和未达标组，分析肝移植受者万古霉素初始稳态谷浓度的影响因素。对具有稳态谷浓度并进行剂量调整的肝移植受者，使用JPKD预测调整给药后的谷浓度。使用预测谷浓度和实际谷浓度之间的权重偏差(WRES)评价JPKD对万古霉素谷浓度的预测能力。计量资料比较采用两独立样本t检验或Mann-Whitney U检验，计数资料比较采用卡方检验或Fisher确切概率法。将P<0.20的变量纳入二分类Logistic回归模型，采用后退法分析万古霉素初始稳态谷浓度的影响因素。P<0.05为差异有统计学意义。

最终纳入58例肝移植受者，其中44例受者在治疗过程中调整给药方案。初始和调整方案万古霉素日剂量分别为(1.79±0.37)和(1.49±0.61)g/d，稳态谷浓度分别为(18±8)和(15±7)mg/L。初始给药后万古霉素谷浓度达标受者比例为46.6%(27/58)；未达标组31例受者中11例(19.0%)为低暴露(<10 mg/L)，20例(34.5%)为高暴露(>20 mg/L)。Logistic回归多因素分析显示，万古霉素日剂量(OR＝15.807，P<0.05)和给药前血清尿素氮(OR＝1.101，P<0.05)是肝移植受者万古霉素初始稳态谷浓度能否达标的影响因素。44例接受剂量调整的肝移植受者调整给药方案后，万古霉素谷浓度达标受者比例为68.2%(30/44)，实测谷浓度和软件预测谷浓度分别为(15±7)mg/L和(14±5)mg/L，WRES≤30%的受者比例为81.8%(36/44)。

肝移植受者万古霉素初始谷浓度达标比例有待提高，万古霉素日剂量和给药前血清尿素氮作为初始稳态谷浓度的影响因素值得引起重视。JPKD对肝移植受者万古霉素谷浓度具有良好的预测能力，可用于协助临床制订个体化给药方案。

To investigate trough concentration of vancomycin in liver transplant recipients during hospitalization and its influencing factors, and to evaluate the value of population pharmacokinetics called software Java PK for Desktop (JPKD) in liver transplant recipients.

A retrospective analysis was conducted on the clinical data of liver transplant recipients who treated with vancomycin and received therapeutic drug monitoring in Zhongshan Hospital of Fudan University from January 2016 to December 2019. The recipients were divided into reach the standard group and fail to reach the standard group according to whether the initial steady-state trough concentration of vancomycin reached 10-20 mg/L. The influencing factors of the initial steady-state trough concentration of vancomycin in liver transplant recipients were also analyzed. For liver transplant recipients with the steady-state trough concentration and received dose adjustment, JPKD software was used to predict the trough concentration of the adjusted regimen. The weight deviation (WRES) between the predicted and the measured concentration was used to evaluate the predictive ability of JPKD software for trough concentration of vancomycin. Measurement data were compared by two independent samples t-test or Mann-Whitney U test, and enumeration data were compared by Chi-square test or Fisher′s exact test. Variables with P<0.20 were included in the binomial logistic regression model, backward LR method to analyze the influencing factors of initial steady-state trough concentrations of vancomycin. A P<0.05 was considered statistically significant.

A total of 58 liver transplant recipients were finally enrolled, 44 of them got dosing regimen adjustment of vancomycin during treatment. The initial and adjusted daily doses of vancomycin were (1.79±0.37) and (1.49±0.61) g/d, respectively, and the steady-state trough concentrations were (18±8) and (15±7) mg/L, respectively. After initial administration, the proportion of recipients with vancomycin trough levels reaching the target was 46.6% (27/58); 11 (19.0%) of the 31 recipients in the substandard group had low exposure (<10 mg/L) and 20 (34.5%) had high exposure (>20 mg/L). Logistic regression analysis revealed that vancomycin daily dose (OR=15.807, P<0.05) and serum urea nitrogen before administration (OR=1.101, P<0.05) were the independent influencing factors of initial trough concentration of vancomycin in liver transplant recipients. Among 44 liver transplant recipients who had adjusted dosing regimen during the treatment, 68.2% (30/44) achieved target trough concentration. The measured concentration and predictive concentration were (15±7) mg/L and (14±5) mg/L respectively. The proportion of liver transplant recipients with WRES ≤30% was 81.8% (36/44).

The proportion of vancomycin initial trough concentration reaching the target needs to be improved in perioperative liver transplant recipients. As the influencing factors of initial trough concentration, vancomycin daily dose and serum urea nitrogen before administration should be paid more attention. JPKD software has a good predictive ability to liver transplant recipients′ vancomycin trough concentration, and may be used to assist clinicians in developing individualized drug protocols.