Application of pharmacokinetic modeling to individualized antibiotic therapy in critically ill septic patients
(2012)
Files
THESE_I-Delattre.pdf
Restricted Access - Adobe PDF
- 1.96 MB
Details
- Authors
- Supervisors
- Wallemacq, Pierre
- Abstract
- (en) During severe sepsis, many physiological modifications take place with a significant impact on the antibiotic pharmacokinetics (PK). The challenge for clinicians is to timely administer the appropriate antibiotic dosage given the pathophysiology and the high interindividual variations in the drug PK. Any efficient therapeutic individualization is required, including therapeutic drug monitoring (TDM). This thesis aimed to develop and evaluate population PK models that can be used as tools to individualize dosage strategies for five antibiotics (one aminoglycoside and four β-lactams) in critically ill septic patients. Our research focused on the ability to predict PK of β-lactams, which are not routinely monitored, using data from aminoglycoside TDM. In the first part of our research, PK models were developed using a standard two-stage approach. Hence, antibiotic PK was described, and multivariate linear regression analyses were performed to test the influence of the aminoglycoside PK parameters, together with other pathophysiological covariates, on the β-lactam PK. Aminoglycoside data from TDM were found to be the major patient-specific characteristics predicting the β-lactam PK. In the second part of our investigations, nonlinear mixed-effects models were developed for the aminoglycoside and for each β-lactam. With this approach, creatinine clearance was found to be the major factor contributing to the variability in the aminoglycoside elimination, and therefore the important patient characteristic to consider for improving dosing regimens. The developed population models for β-lactams were able to account for PK variability using renal biomarkers or aminoglycoside data from TDM. Results supported the superiority of the aminoglycoside concentrations, over renal function and aminoglycoside PK parameters, to predict the dosage requirements of β-lactams. This thesis demonstrates that optimization of the β-lactam dosage can be reached without any β-lactam measurement, just by using aminoglycoside data from TDM. Overall, the developed PK models provide practical tools that meet the current clinical constraints and can guide clinicians to effectively dose the antibiotics in critically ill septic patients.
- Affiliations
Citations
Delattre, I. (2012). Application of pharmacokinetic modeling to individualized antibiotic therapy in critically ill septic patients. https://hdl.handle.net/2078.5/162870