ICU sedation, delirium & recovery
Towards model-informed precision dosing of piperacillin: multicenter systematic external evaluation of pharmacokinetic models in critically ill adults with a focus on Bayesian forecasting
Authors' conclusion
The predictive performance of published PIP models in ICU patients varied considerably, highlighting the relevance of appropriate model selection for MIPD.
Study context
- Population
- Critically ill adult intensive care unit (ICU) patients receiving piperacillin (11 centres)11 centers/3654 concentrations
- Participants
- 561 patients
- Comparison
- 24 published piperacillin PopPK models: a priori predictions vs Bayesian forecasting with first (B1) or first and second (B2) therapeutic drug monitoring (TDM) samples
- Outcome
- Model accuracy and precision: median relative prediction error (MPE) [%] and median absolute relative prediction error (MAPE) [%]
Methodological detailsBayesian forecasting
- Bayesian forecasting
Main finding
As reported in the abstract
Predictive performance across 24 models
Large inter-model variability, A: MPE range
- 135.6-78.3%
Large inter-model variability, A: MAPE range
35.7-135.6%
B1/B2 relative improvement vs A, |MPE| median all models
45.1/67.5%
B1/B2 relative improvement vs A, MAPE median all models
29/39%
Further findings
Best model for total dataset (Kim et al.)
- Kim et al. MPE A/B1/B2
- - 9.8/- 5.9/- 0.9%
- Kim et al. MAPE A/B1/B2
- 37/27.3/23.7%
Best model by infusion mode (subgroup)
- Udy et al. performed best with intermittent infusion; Klastrup et al. best predicted continuous infusion
Stratified by sex and RRT (subgroup)
- Further promising models identified
Original publication
Towards model-informed precision dosing of piperacillin: multicenter systematic external evaluation of pharmacokinetic models in critically ill adults with a focus on Bayesian forecasting. Intensive Care Med. Published 13 July 2023. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →