Apinya Boonpeng (1), Monchana Nawakitrangsan (2), Maseetoh Samaeng (2), Sutthiporn Pattharachayakul (1), Sutep Jaruratanasirikul (2)
(1) Department of Clinical Pharmacy, Faculty of Pharmaceutical Sciences, Prince of Songkla University, Songkhla, Thailand, (2) Division of Internal Medicine, Faculty of Medicine, Prince of Songkla University, Songkhla, Thailand.
Introduction/Objectives: Imipenem is a broad-spectrum carbapenem antibiotic and is frequently prescribed in critically ill patients with severe nosocomial infections. These patients show several pathophysiological changes that may alter the imipenem pharmacokinetics (PK) normally found in other populations. Although the PK of imipenem has been widely studied, most studies have been conducted on small patient populations and have not directly compared the PK between critically ill and non-critically ill patients. Therefore, we pooled several studies into a larger PK dataset to estimate the population PK parameters of imipenem and to investigate the influence of critically ill status on these parameters. We also performed Monte Carlo simulations to determine an optimal dosing regimen of imipenem in critically ill patients.
Methods: The unbound imipenem plasma concentrations in adult patients with nosocomial infections including both non-critically ill (n=25, 8-10 samples/subjects) and critically ill patients (n=15, 5-9 samples/subjects) from 4 studies were integrated into a population pharmacokinetic analysis. The model building was performed using NONMEM (v 7.4.3) along with PsN (v 4.9.0) and Pirana (v 2.9.9). The first-order conditional estimation method with eta-epsilon interaction (FOCE-I) was used for parameter estimation throughout the model-building process. Different structural and error models were compared in terms of changes in objective function value (OFV) and goodness-of-fit plots. After an appropriate structural model was established, the influence of potential covariates including critically ill status, age, sex, body weight, creatinine clearance, glomerular filtration rate (GFR), acute kidney injury, serum albumin, inotropic/vasopressor use, APACHE score, and SOFA score on PK parameters was evaluated. Model adequacy was assessed by goodness-of-fit plots and prediction-corrected visual predictive checks (pcVPC). A Monte Carlo simulation was performed to determine the probability of target attainment (PTA) of achieving 40% of the time that free plasma concentration remained above the MIC (fT>MIC) and 75% fT>MIC.
Results: A two-compartment model with first-order elimination best described the pooled plasma imipenem concentration-time profiles from 4 studies. Residual variability was modelled by an additive-plus-proportional error model with separate proportional error terms for each study. In our final model, the GFR (calculated by CKD-EPI equation) and body weight were the only significant covariates on clearance and central volume of distribution, respectively. Critically ill status appeared to have only a minor impact on PK parameters and was dropped from the model. The final population PK parameter estimates (between-subject variability) were: clearance 13.9 L/h (46.6%), central volume of distribution 15.6 L (55.4%), peripheral volume of distribution 16.1 L (70.9%), and inter-compartment clearance 21.8 L/h. These parameters were then used to perform Monte Carlo simulation of various dosing regimens. Considering 40%fT>MIC as the target for non-critically ill patients, a standard dosage regimen of 500 mg every 6 hours or 1000 mg every 8 hours was sufficient to achieve 90% PTA for pathogens with MICs of 2 mg/L. For critically ill patients, it is suggested that a more aggressive target of 75-100%fT>MIC should be ensured for optimal exposure. In order to achieve the target 75%fT>MIC, a dose of 1000 mg every 6 hours or extended 3-h infusion of 500 mg every 6 hours with loading dose was required for treating pathogen with MIC less than 2 mg/L.
Conclusions: Our population PK model successfully characterized the imipenem pharmacokinetics in adult patients with nosocomial infections. Critically ill status did not show a significant influence on PK parameters. It might be due to the small proportion of critically ill subjects in our analysis. Further studies with larger populations may help elucidate more fully whether a critically ill status has an effect on PK parameters.
References:
[1] Blot SI, Pea F, Lipman J. The effect of pathophysiology on pharmacokinetics in the critically ill patient–concepts appraised by the example of antimicrobial agents. Adv Drug Deliv Rev. 2014;77:3-11.
[2] Jaruratanasirikul S, Sudsai T. Comparison of the pharmacodynamics of imipenem in patients with ventilator-associated pneumonia following administration by 2 or 0.5 h infusion. J Antimicrob Chemother. 2009;63(3):560-3.
[3] Jaruratanasirikul S, Wongpoowarak W, Jullangkoon M, Samaeng M. Population pharmacokinetics and dosing simulations of imipenem in serious bacteraemia in immunocompromised patients with febrile neutropenia. J Pharmacol Sci. 2015;127(2):164-9.
[4] Jaruratanasirikul S, Wongpoowarak W, Nawakitrangson M, Thengyai S, Samaeng M. Population pharmacokinetics and Monte Carlo dosing simulations of imipenem in patients with ventilator-associated pneumonia. Lung Breath J. 2017;1(1):1-4.
Reference: PAGE () Abstr 9463 [www.page-meeting.org/?abstract=9463]
Poster: Drug/Disease Modelling - Infection