Toward Model-Informed Precision Dosing of Imipenem: Multicenter External Validation of Population Pharmacokinetic Models in Adult Critically Ill Patients.
Аннотация
BACKGROUND
Model-informed individualized dosing of imipenem is required for critically ill patients due to high infection mortality and large pharmacokinetic (PK) variability. This study aims to externally evaluate the predictive performance of available imipenem pharmacokinetic (popPK) models to facilitate clinical application.
METHODS
A multicenter dataset of 152 ICU patients (201 concentrations) was used for the external validation of 9 popPK models. Model performance was investigated for prediction- and simulation-based diagnostics and Bayesian forecasting. The median relative prediction error (rPEmedian), median absolute relative prediction error (rAPEmedian), and NPDE were calculated to quantify accuracy and precision.
RESULTS
Model predictive performance exhibited heterogeneity. Priori prediction showed wide rPEmedian ranges, and only 3 models were within ± 15%. Moreover, the priori predictions of all models failed to demonstrate satisfactory performance in terms of both F20 and F50. Notably, Bayesian forecasting incorporating TDM data significantly improved accuracy, with four models demonstrating superior performance in terms of rPEmedian and F50. Predictive performance was poorer in patients with renal function impairments. Simulation diagnostics revealed systematic bias across all models. In general, the Truong et al (2025) model performed better than the other models.
CONCLUSION
Available imipenem popPK models for adult critically ill patients were unsatisfactory in predictive performance, and the Truong et al (2025) model performed best among all models. The integration of Bayesian forecasting with therapeutic drug monitoring (TDM) data significantly improved predictive accuracy, suggesting that the synergistic use of the popPK model and TDM may optimize clinical dosing decisions and potentially improve clinical outcomes.
Предметные рубрики MeSH
Подготовка и перевод публикации
Обработка статьи · этапы, ошибки и управление
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Тип публикации указан по данным источника. Уровень доказательности автоматически не присваивается.