Related Experiment Video
Updated: Aug 1, 2026
06:09
Intrarenal Injection of Escherichia coli in a Rat Model of Pyelonephritis
Published on: July 18, 2017
8.2K
Model Re-Estimation: An Alternative for Poor Predictive Performance during External Evaluations? Example of
Alexandre Duong1,2, Chantale Simard3,4, David Williamson1,5
1Faculté de Pharmacie, Université de Montréal, Montreal, QC H3T 1J4, Canada.
Pharmaceutics
|July 27, 2022
Summary
External evaluation of gentamicin population pharmacokinetic (PopPK) models for critically ill patients is essential. Re-estimating existing models improved their predictive performance, offering an alternative to new model development.
Area of Science:
- Pharmacokinetics
- Clinical Pharmacology
- Critical Care Medicine
Background:
- External evaluation of gentamicin population pharmacokinetic (PopPK) models is critical for clinical application.
- Few existing PopPK models for critically ill patients incorporate external evaluation in their development.
Purpose of the Study:
- To evaluate existing gentamicin PopPK models developed for critically ill patients.
- To assess the predictive performance of these models using external data.
Main Methods:
- Literature review to identify relevant aminoglycoside PopPK models.
- Retrospective data collection from two Quebec hospitals.
- External evaluation and model re-estimation using NONMEM® (v7.5).
Main Results:
- None of the initially evaluated models demonstrated acceptable bias and imprecision.
- All models showed acceptable predictive performance after re-estimation.
- A dosing nomogram was developed using the best-performing re-estimated model.
Conclusions:
- Prioritizing external evaluation during PopPK model development is crucial.
- Model re-estimation is a viable alternative to de novo model development, particularly when covariates are similar across existing models.
