Laura Hermann (1), Sarah Dräger (2,3), Verena Schöning (1), Katharina Rentsch (4), Nicolas Gürtler (2), Stephan Moser (2,3), Parham Sendi (5,6), Michael Osthoff (2,3,7), Felix Hammann (1)
(1) Division of Clinical Pharmacology & Toxicology, Department of Internal Medicine, University Hospital Bern, Switzerland, (2) Division of Internal Medicine, University Hospital Basel, Basel, Switzerland, (3) Department of Clinical Research, University of Basel, Basel, Switzerland, (4) Division of Laboratory Medicine, University Hospital Basel, Basel, Switzerland, (5) Department of Infectious Diseases and Hospital Epidemiology, University Hospital Basel, Basel, Switzerland, (6) Institute for Infectious Diseases, University of Bern, Bern, Switzerland, (7) Department of Internal Medicine, Cantonal Hospital Winterthur, Winterthur, Switzerland,
Objectives: Target attainment of flucloxacillin in patients with Staphylococcus aureus bloodstream infections (BSI) is challenging due to the high inter-individual variability in pharmacokinetics (PK), particularly in critically ill patients. Although flucloxacillin is highly protein-bound, in a clinical setting it is total concentration which is usually measured , with an assumed unbound fraction of 5%. Especially in critically ill patients with altered renal function and hypoalbumenia, this might lead to supratherapeutic levels. A PK model can be used to optimize and individualize antibiotic dosing. The objective of this study was to develop a flucloxacillin population PK model for precision dosing, and to retroactively assess target attainment in patients with BSI.
Methods: We included hospitalized, adult, critically and non-critically ill patients with confirmed BSI in this single-centre, prospective, observational pharmacokinetic cohort study conducted at the University Hospital Basel, Switzerland. We measured mid-dose and trough level total and unbound flucloxacillin concentrations after intermittent bolus administration using high-performance liquid chromatography mass spectrometry. Based on these data, we developed and validated a population PK model for total and unbound serum flucloxacillin, considering different models for protein binding. We used Monte Carlo dosing simulations to assess and compare target attainment (100% fT>MIC) for intermittent bolus and continuous infusion.
Results:
We included 178 unbound and total flucloxacillin concentrations of 49 patients, where 47% were critically ill. The majority of patients (69%) were treated with flucloxacillin 12g qd. We found that the unbound fraction was higher compared to literature (5%) at 11.5% and 16.7% for non-critical and critically ill patients, respectively. A one-compartment model with linear elimination best described the concentration profiles. We improved model performance by incorporating the estimated glomerular filtration rate according to the Cockcroft-Gault equation (eGFR-GC), and albumin as covariates. Dosing simulations revealed better target attainment for continuous compared to intermittent bolus infusion.
Conclusions: The base structural model underlying all analyses only uses a single compartment with linear elimination, whereas other authors have proposed more elaborate configurations [1-3]. We believe this to be due to the nature of drug dosing schedules for flucloxacillin (short intervals of 2-8h) that do not capture any slow elimination from deep compartments. The identification of eGFR-GC as one of the informative covariates is plausible as flucloxacillin is mainly eliminated by the kidneys [4]. Serum albumin was the other informative covariate and was mainly responsible for the determination of the ratio of unbound and total flucloxacillin concentration. Preference should be given to measuring unbound flucloxacillin concentrations, as the actual unbound fractions may significantly exceed those calculated using an assumed fraction of 5%. Other studies [5, 6] also showed potential improvements in target attainment by using continuous infusions over intermittent bolus infusions. Additionally, continuous infusion reduces peak concentrations and thus might decrease the risk of toxicity.
References:
[1] Jager, N.G.L., R.M. van Hest, J. Xie, G. Wong, M. Ulldemolins, R.J.M. Brüggemann, J. Lipman, and J.A. Roberts, Optimization of flucloxacillin dosing regimens in critically ill patients using population pharmacokinetic modelling of total and unbound concentrations. J Antimicrob Chemother, 2020. 75(9): p. 2641-2649.
[2] Ulldemolins, M., J.A. Roberts, S.C. Wallis, J. Rello, and J. Lipman, Flucloxacillin dosing in critically ill patients with hypoalbuminaemia: Special emphasis on unbound pharmacokinetics. Journal of Antimicrobial Chemotherapy, 2010.
[3] Wilkes, S., I. van Berlo, J. Ten Oever, F. Jansman, and R. Ter Heine, Population pharmacokinetic modelling of total and unbound flucloxacillin in non-critically ill patients to devise a rational continuous dosing regimen. Int J Antimicrob Agents, 2019. 53(3): p. 310-317.
[4] Swissmedic. Arzneimittelinformation. Available from: www.swissmedicinfo.ch.
[5] Lee, Y.R., P.D. Miller, S.K. Alzghari, D.D. Blanco, J.D. Hager, and K.S. Kuntz, Continuous Infusion Versus Intermittent Bolus of Beta-Lactams in Critically Ill Patients with Respiratory Infections: A Systematic Review and Meta-analysis. European Journal of Drug Metabolism and Pharmacokinetics, 2018. 43(2): p. 155-170.
[6] Kasiakou, S.K., K.R. Lawrence, N. Choulis, and M.E. Falagas, Continuous versus Intermittent Intravenous Administration of Antibacterials with Time-Dependent Action. Drugs, 2005. 65(17): p. 2499-2511.
Reference: PAGE 32 (2024) Abstr 11034 [www.page-meeting.org/?abstract=11034]
Poster: Drug/Disease Modelling - Infection