Model Re-Estimation: An Alternative for Poor Predictive Performance during External Evaluations? Example of Gentamicin in Critically Ill Patients.

dosing nomogram external evaluation gentamicin model re-estimation population pharmacokinetic modeling

Journal

Pharmaceutics
ISSN: 1999-4923
Titre abrégé: Pharmaceutics
Pays: Switzerland
ID NLM: 101534003

Informations de publication

Date de publication:
07 Jul 2022
Historique:
received: 22 04 2022
revised: 03 06 2022
accepted: 06 07 2022
entrez: 27 7 2022
pubmed: 28 7 2022
medline: 28 7 2022
Statut: epublish

Résumé

An external evaluation is crucial before clinical applications; however, only a few gentamicin population pharmacokinetic (PopPK) models for critically ill patients included it in the model development. In this study, we aimed to evaluate gentamicin PopPK models developed for critically ill patients. The evaluated models were selected following a literature review on aminoglycoside PopPK models for critically ill patients. The data of patients were retrospectively collected from two Quebec hospitals, the external evaluation and model re-estimation were performed with NONMEM From the datasets of 39 and 48 patients from the two Quebec hospitals, none of the evaluated models presented acceptable values for bias and imprecision. Following model re-estimations, all models showed an acceptable predictive performance. An a priori dosing nomogram was developed with the best performing re-estimated model and was consistent based on recommended dosing regimens. Due to the poor predictive performance during the external evaluations, the latter must be prioritized during model development. Model re-estimation may be an alternative to developing a new model, especially when most known models display similar covariates.

Sections du résumé

BACKGROUND BACKGROUND
An external evaluation is crucial before clinical applications; however, only a few gentamicin population pharmacokinetic (PopPK) models for critically ill patients included it in the model development. In this study, we aimed to evaluate gentamicin PopPK models developed for critically ill patients.
METHODS METHODS
The evaluated models were selected following a literature review on aminoglycoside PopPK models for critically ill patients. The data of patients were retrospectively collected from two Quebec hospitals, the external evaluation and model re-estimation were performed with NONMEM
RESULTS RESULTS
From the datasets of 39 and 48 patients from the two Quebec hospitals, none of the evaluated models presented acceptable values for bias and imprecision. Following model re-estimations, all models showed an acceptable predictive performance. An a priori dosing nomogram was developed with the best performing re-estimated model and was consistent based on recommended dosing regimens.
CONCLUSION CONCLUSIONS
Due to the poor predictive performance during the external evaluations, the latter must be prioritized during model development. Model re-estimation may be an alternative to developing a new model, especially when most known models display similar covariates.

Identifiants

pubmed: 35890322
pii: pharmaceutics14071426
doi: 10.3390/pharmaceutics14071426
pmc: PMC9315759
pii:
doi:

Types de publication

Journal Article

Langues

eng

Références

Intensive Care Med. 2017 Jul;43(7):1021-1032
pubmed: 28409203
Ther Drug Monit. 2017 Oct;39(5):522-530
pubmed: 28682925
Pharmacol Res. 1995 Oct;32(4):241-8
pubmed: 8866841
Int J Antimicrob Agents. 2017 Feb;49(2):204-211
pubmed: 28038961
Br J Anaesth. 2017 Mar 1;118(3):415-423
pubmed: 28186264
Ther Drug Monit. 2008 Dec;30(6):674-81
pubmed: 19057371
Antimicrob Agents Chemother. 1999 Apr;43(4):727-37
pubmed: 10103173
Nephron. 1976;16(1):31-41
pubmed: 1244564
Antimicrob Agents Chemother. 2019 Mar 27;63(4):
pubmed: 30917981
Infect Dis Clin North Am. 2009 Dec;23(4):791-815, vii
pubmed: 19909885
Antibiotics (Basel). 2021 Apr 29;10(5):
pubmed: 33946905
BMC Nephrol. 2016 Aug 02;17(1):109
pubmed: 27484681
Postgrad Med. 2001 Dec;110(6):55-62; quiz 11
pubmed: 11787409
Cold Spring Harb Perspect Med. 2016 Jun 01;6(6):
pubmed: 27252397
Clin Pharmacokinet. 1999 Feb;36(2):89-98
pubmed: 10092956
Am J Health Syst Pharm. 2015 Aug 15;72(16):1357-64
pubmed: 26246292
Front Pharmacol. 2020 Mar 03;11:172
pubmed: 32194411
J Pharmacokinet Pharmacodyn. 2010 Feb;37(1):49-65
pubmed: 20033477

Auteurs

Alexandre Duong (A)

Faculté de Pharmacie, Université de Montréal, Montreal, QC H3T 1J4, Canada.
Laboratoire de Suivi Thérapeutique Pharmacologique et Pharmacocinétique, Faculté de Pharmacie, Université de Montréal, Montreal, QC H3T 1J4, Canada.

Chantale Simard (C)

Institut Universitaire de Cardiologie et Pneumologie de Québec, Quebec, QC G1V 4G5, Canada.
Faculté de Pharmacie, Université Laval, Quebec, QC G1V 0A6, Canada.

David Williamson (D)

Faculté de Pharmacie, Université de Montréal, Montreal, QC H3T 1J4, Canada.
Hôpital Sacré-Cœur de Montréal, Université de Montréal, Montreal, QC H4J 1C5, Canada.

Amélie Marsot (A)

Faculté de Pharmacie, Université de Montréal, Montreal, QC H3T 1J4, Canada.
Laboratoire de Suivi Thérapeutique Pharmacologique et Pharmacocinétique, Faculté de Pharmacie, Université de Montréal, Montreal, QC H3T 1J4, Canada.
Centre de Recherche, CHU Sainte Justine, Montreal, QC H3T 1C5, Canada.

Classifications MeSH