文献速递治疗药物监测与药物基因组学特殊人群用药抗菌药物管理

新生儿万古霉素贝叶斯剂量调整的模拟TDM研究

Evaluating Bayesian Vancomycin Dosage Adjustment in Neonates: A Simulation-Based Virtual TDM Study.
PubMed 文献 · Clin Pharmacokinet · 2026-05-02

该虚拟TDM研究基于已发表群体PK模型在1000名虚拟新生儿中模拟万古霉素浓度,比较四种TDM策略达标率(AUC24,SS/MIC 360-540 h)。结果显示贝叶斯MIPD策略在1个TDM周期后超过74%患者达标、4周期后100%达标,且单样本即可准确预测AUC24,SS。

要点

  • 初始方案未调整时不足49%患者达标,常见万古霉素暴露过量。
  • 贝叶斯MIPD策略1周期后超74%达标、4周期后100%达标,不受初始方案影响。
  • 单样本即可准确预测AUC24,SS,可简化新生儿采样策略。

核心结论

  • 贝叶斯MIPD可优化新生儿万古霉素暴露并简化采样策略,值得整合至常规新生儿监护并验证。

研究设计与样本

  • 虚拟TDM研究:基于已发表群体PK模型在NONMEM中为1000名虚拟新生儿模拟参考PK参数与药物浓度。
  • 比较4种TDM策略:不调整初始方案、按标准规则调整至Ctrough 10-15 mg/L、Tucuxi贝叶斯优化(稳态Ctrough)、Tucuxi贝叶斯优化(首剂后两浓度)。

主要发现

  • 策略1不足49%患者达标,导致频繁万古霉素暴露过量。
  • 策略2在每8小时方案中超过75%患者AUC24,SS/MIC达360-540 h,但更长给药间隔仅31%达标。
  • 策略3和4在1个TDM周期后超过74%患者达标,4周期后100%达标,与初始给药方案无关。
  • 单样本足以准确预测AUC24,SS。

关键数值与推荐

  • 目标为AUC24,SS/MIC在360-540 h。
  • 策略2目标Ctrough为10-15 mg/L。

临床怎么用

  • 新生儿万古霉素TDM可采用贝叶斯MIPD进行剂量优化,以提高AUC24,SS/MIC达标率。
  • 贝叶斯MIPD可仅用单样本预测AUC24,SS,减少新生儿采血负担。

局限性

  • 本研究为虚拟TDM模拟研究,结论需在常规新生儿临床实践中整合并验证。

以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。

原文

BACKGROUND AND OBJECTIVES: Therapeutic drug monitoring (TDM) of vancomycin, ideally based on the ratio of the daily area under the curve to the minimal inhibitory concentration at steady state (AUC24,SS/MIC), is hindered by sampling limitations in neonates, so trough concentration (Ctrough) is often used as a surrogate. This virtual TDM study aimed to evaluate the performance of Bayesian model-informed precision dosing (MIPD) for vancomycin in neonates. METHODS: Reference pharmacokinetic (PK) parameters and drug concentrations were simulated in NONMEM for 1000 virtual neonates using a published population PK model. Four TDM strategies were compared based on the percentage of patients achieving an AUC24,SS/MIC between 360 and 540 h. Strategy 1 maintained the initial regimen unchanged; strategy 2 adjusted doses using a standard rule of three to target a Ctrough of 10-15 mg/L; strategies 3 and 4 used the MIPD software Tucuxi with a steady-state Ctrough and two concentrations after the first dose, respectively, for Bayesian dosage optimization. Individual dosages were adjusted following each strategy, recalculating AUC24,SS to determine the percentage of target attainment. Iterative adjustments were performed by resampling Ctrough. A sub-study evaluated optimal sampling time by comparing Tucuxi-estimated AUC24,SS with the reference value. RESULTS: Less than 49% of patients reached the target with strategy 1, leading to frequent vancomycin overexposure. Strategy 2 enabled over 75% of patients to achieve AUC24,SS/MIC between 360 and 540 h with every-8-h regimens, but only 31% with longer dosing intervals. Strategies 3 and 4 allowed over 74% of patients to reach the target after one TDM cycle and 100% after four cycles, regardless of the initial dosing regimen. A single sample sufficed for accurate AUC24,SS prediction. CONCLUSION: This virtual TDM study indicates that Bayesian MIPD optimizes vancomycin exposure in neonates, while simplifying the sampling strategy. This approach deserves to be integrated and validated in routine neonatal care.

来源:Clin Pharmacokinet,查看原文。版权归原作者所有。

引用

  1. Ravix A, Thoma Y, Cathignol A, et al. Evaluating Bayesian Vancomycin Dosage Adjustment in Neonates: A Simulation-Based Virtual TDM Study. Clin Pharmacokinet. 2026. PMID: 42068514. doi:10.1007/s40262-026-01642-4

供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。