Department of Pharmacy, the First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, China.
Ther Drug Monit. 2020 Aug;42(4):600-609. doi: 10.1097/FTD.0000000000000750.
Vancomycin is a critical antibiotic used in important infections, and therapeutic drug monitoring (TDM) is recommended. Bayesian forecasting is demonstrated to provide an approach that can improve trough concentration monitoring for dose adjustment. The objective of this study was to determine whether TDM coupled with a Bayesian approach could increase trough concentration target attainment and prevent vancomycin-associated nephrotoxicity in patients with renal insufficiency.
A prospective study was performed using propensity score matching to provide covariate balance in renal insufficiency patients with gram-positive bacterial infections treated with vancomycin. Patients were divided into non-TDM (84 cases) and TDM (84 cases) groups, and their clinical outcomes were compared. The primary endpoints were probability of trough concentration target attainment and incidence of vancomycin-associated nephrotoxicity. A decision-tree model was developed to assess the cost effectiveness of TDM to prevent vancomycin-associated nephrotoxicity.
Of the 168 eligible patients, 69 from each group (non-TDM and TDM) were matched based on propensity scores. In the matched cohort, trough concentration target attainment was higher with TDM (P = 0.003). Furthermore, reaching toxic trough concentrations was avoided (P = 0.027) in the TDM group. Multivariate logistic regression analysis confirmed that TDM practice independently reduced the incidence of vancomycin-associated nephrotoxicity in renal insufficiency patients (P = 0.021). According to this reduced nephrotoxicity, the incremental cost-effectiveness ratios of ¥22,638 per nephrotoxic episode prevented was found for vancomycin TDM.
TDM coupled with Bayesian forecasting led to an increase in trough concentration target attainment and a decrease in the incidence of vancomycin-associated nephrotoxicity in renal insufficiency patients. In this high-risk population, TDM was demonstrated to be a cost-effective procedure.
万古霉素是一种用于重要感染的关键抗生素,推荐进行治疗药物监测(TDM)。贝叶斯预测被证明是一种可以改善谷浓度监测以调整剂量的方法。本研究的目的是确定 TDM 联合贝叶斯方法是否可以提高谷浓度目标达标率并预防肾功能不全患者的万古霉素相关性肾毒性。
采用倾向评分匹配进行前瞻性研究,为接受万古霉素治疗的革兰阳性菌感染肾功能不全患者提供协变量平衡。将患者分为非 TDM(84 例)和 TDM(84 例)组,并比较其临床结局。主要终点是谷浓度目标达标率和万古霉素相关性肾毒性发生率。建立决策树模型评估 TDM 预防万古霉素相关性肾毒性的成本效益。
在 168 名合格患者中,每组(非 TDM 和 TDM)各有 69 例基于倾向评分进行匹配。在匹配队列中,TDM 组谷浓度目标达标率更高(P = 0.003)。此外,TDM 组避免了达到毒性谷浓度(P = 0.027)。多变量逻辑回归分析证实,TDM 实践独立降低了肾功能不全患者万古霉素相关性肾毒性的发生率(P = 0.021)。根据这种降低的肾毒性,发现预防每例肾毒性事件的万古霉素 TDM 的增量成本效益比为 ¥22638。
TDM 联合贝叶斯预测可提高肾功能不全患者的谷浓度目标达标率并降低万古霉素相关性肾毒性的发生率。在这个高危人群中,TDM 被证明是一种具有成本效益的方法。