External Evaluation of Population Pharmacokinetic Models of Piperacillin in Preterm and Term Patients from Neonatal
Frida S Boer-Pérez1, Victoria Lima-Rogel2, Ana R Mejía-Elizondo2
1Facultad de Ciencias Químicas, Universidad Autónoma de San Luis Potosí, #6, Dr. Manuel Nava Martinez, S.L.P. PO Box 78210, San Luis Potosí, México.
Insights
This study found the Cohen-Wolkowiez et al. 2014 piperacillin pharmacokinetic model, using a Bayesian approach, best predicts drug levels in Mexican neonates. This supports model-informed precision dosing for piperacillin/tazobactam in neonatal intensive care units.
Area of Science:
- Pharmacology
- Neonatal Medicine
- Pharmacokinetics
Background:
- Piperacillin/tazobactam is crucial for neonatal sepsis, but limited safety and pharmacokinetic data exist for this vulnerable population.
- Neonatal organic immaturity leads to high piperacillin pharmacokinetic variability, impacting treatment efficacy and resistance.
- Evaluating existing pharmacokinetic models is essential for optimizing piperacillin dosing in neonates.
Purpose of the Study:
- To assess the predictive performance of published piperacillin population pharmacokinetic models.
- To evaluate these models for a model-informed precision dosing strategy in Mexican preterm and term neonates.
- To identify the most suitable pharmacokinetic model for clinical application in neonatal intensive care.
Main Methods:
- Identified and extracted data from published piperacillin population pharmacokinetic models including neonates.
- Replicated models in NONMEM software and collected piperacillin plasma concentrations from 25 neonates.
- Evaluated model performance using a priori and Bayesian approaches, analyzing bias and precision.
Main Results:
- Three population pharmacokinetic models were evaluated; none were suitable using an a priori approach.
- The Cohen-Wolkowiez et al. 2014 model with a Bayesian approach demonstrated the best performance.
- This model, requiring two blood samples, predicted 66.6% of observations with <30% median absolute prediction error.
Conclusions:
- The Cohen-Wolkowiez et al. 2014 model is superior for predicting piperacillin concentrations in Mexican neonates.
- The Bayesian approach using two plasma concentrations is clinically acceptable for bias and precision.
- Model-informed precision dosing with this model can optimize piperacillin dosage in this population.
Background And Objectives:
Piperacillin/tazobactam is extensively used off-label to treat late-onset neonatal sepsis, but safety and pharmacokinetic data in this population are limited. Additionally, the organic immaturity of the newborns contributes to a high piperacillin pharmacokinetic variability. This affects the clinical efficacy of the antibiotic treatment and increases the probability of developing drug resistance. This study aimed to evaluate the predictive performance of reported piperacillin population pharmacokinetic models for their application in a model-informed precision dosing strategy in preterm and term Mexican neonatal intensive care patients.
Methods:
Published population pharmacokinetic models for piperacillin which included neonates in their study population were identified. From the reference models, structured models, population pharmacokinetic parameters, and interindividual and residual variability data were extracted to be replicated in pharmacokinetic software (NONMEM® version 7.4). For the clinical study, a sampling schedule was designed, and 2-3 blood samples of 250 µL were taken from neonates who met the inclusion criteria. Piperacillin plasma concentrations were determined by liquid chromatography/tandem mass spectrometry. The clinical treatment data were collected, and piperacillin plasma concentrations were estimated using reference pharmacokinetic models for an a priori or Bayesian approach. Statistical methods were used in terms of bias and precision to evaluate the differences between observed and estimated neonatal piperacillin plasma concentrations with the different approaches and to identify the pharmacokinetic model that best fits the neonatal data.
Results:
A total of 70 plasma samples were collected from 25 neonatal patients, of which 15 were preterm neonates. The overall median value (range) postnatal age, gestational age, body weight, and serum creatinine at the sampling collecting day were 12 (3-26) days, 34.2 (26-41.1) weeks, 1.78 (0.08-3.90) Kg, 0.47 (0.20-0.90) mg/dL, respectively. Three population pharmacokinetic models for piperacillin in infants up to 2 months were identified, and their predictive performance in neonatal data was evaluated. No pharmacokinetic model was suitable for our population using an a priori approach. The model published by Cohen-Wolkowiez et al. in 2014 with a Bayesian approach showed the best performance of the pharmacokinetic models evaluated in our neonatal data. The procedure requires two blood samples (predose and postdose), and, when applied, it predicted 66.6% of the observations with a relative median absolute predicted error of less than 30%.
Conclusions:
The population pharmacokinetic model developed by Cohen-Wolkowiez et al. in 2014 demonstrated superior performance in predicting the plasma concentration of piperacillin in preterm and term Mexican neonatal intensive care patients. The Bayesian approach, including two different piperacillin plasma concentrations, was clinically acceptable regarding bias and precision. Its application for model-informed precision dosing can be an option to optimize the piperacillin dosage in our population.
Related Concept Videos
01:12Analysis of Population Pharmacokinetic Data
01:20Pharmacokinetic Models: Overview
There are three primary types of models: empirical, compartment, and physiological. Empirical models, with minimal...
01:26Pharmacokinetic Models: Comparison and Selection Criterion
Physiological models take a detailed approach by considering specific molecular processes. They can predict drug distribution, metabolism, and elimination changes, providing a comprehensive understanding of how drugs interact with the body.
00:59Model-Independent Approaches for Pharmacokinetic Data: Noncompartmental Analysis
One important characteristic of noncompartmental analyses is that drug exposure increases proportionally with increasing doses. This...
01:21Factors Affecting Drug Response: Overview
01:07Physiological Pharmacokinetic Models: Incorporating Hepatic Transporter-Mediated Clearance
A recent model describes pravastatin's hepatobiliary excretion,...