Sami Ullah (1)*, Ronny Beer (3)*, Uwe Fuhr (1), Max Taubert (1), Markus Zeitlinger (2), Raimund Helbok (3)
(1) Department I of Pharmacology, Hospital of the University of Cologne, Cologne Germany (2) Department of Clinical Pharmacology, Medical University of Vienna, Vienna, Austria. (3) Neurological Intensive Care Unit, Department of Neurology, Medical University of Innsbruck, Innsbruck, Austria. * Joint first author
Objectives:
Piperacillin is an extended-spectrum penicillin antibiotic with excellent activity against Gram-positive and -negative bacteria and is thus frequently used as an empirical therapy in various nosocomial infections. In a microdialysis study conducted in septic shock patients, free concentrations in the interstitial fluid (ISF) of skeletal muscle and subcutaneous adipose tissues were 5-10 times lower than corresponding unbound plasma concentrations with mean penetration (AUCISF/AUCunbound plasma) values of 0.19 and 0.10 respectively (1). However, less data is available regarding effective concentrations of piperacillin reached in the brain of severely ill patients.
The aim of this evaluation was to assess piperacillin penetration into the ISF of the brain in critically ill patients by use of a population pharmacokinetic model.
Methods:
Data were available from 10 comatose patients with acute haemorrhagic stroke (median age and body weight of 48 [range 32-72] years and 75 [60-95] kg) being treated for nosocomial pneumonia in an ICU setting. After getting approval from ethics committee and informed consent from the patients, each patient was administered a standard dose of 4 g of piperacillin in combination with 0.5 g tazobactam every eight hours by a 30-minute infusion. Microdialysis samples from interstitial space fluid of human brain parenchyma, representing unbound ISF concentrations, were collected in one-hour intervals after single dose and at steady state. Concentrations were quantified by reversed-phase high-performance liquid chromatography. The lower limit of quantification was 0.05 mg/L. Population pharmacokinetic modeling and covariate analysis were performed using NONMEM 7.4.1 software. Because plasma samples were not available, a published two compartment plasma pharmacokinetic model of piperacillin was selected to drive brain concentrations (2).
Results:
One additional compartment was added to describe brain microdialysis data. To account for an observed delay in the initial rise of concentrations in the brain, a lag time and a transit compartment were introduced. The model was parameterized in terms of inter-compartmental clearance between plasma and brain (Qb = 1.54 L/h), apparent volume of distribution of brain (Vb = 5.28 L), absorption lag (ALAG = 3.01 h) and a transfer rate in the transit compartment (Ktr = 0.092 h-1). Inter-individual variability [CV%] was found to be 50.5%, 30.3% and 83.4% on Qb, absorption lag and Ktr respectively. At steady state, Monte Carlo simulations suggested that median AUC0-24 ratio (AUCISF/AUCunbound plasma) was 0.059 (95% prediction interval 0.014-0.25). Plasma protein binding of piperacillin was assumed to be 30% in our analysis. No meaningful covariates of pharmacokinetic parameters were identified.
Conclusions:
The empirical model was able to describe the data well. The observed delay for piperacillin to reach the brain advocates for the earliest possible administration of piperacillin in infections involving the CNS.
Further evaluations are needed to ascertain whether the current dosing regimen of piperacillin is sufficient to treat the nosocomial CNS infections in patients with acute brain injuries.
References:
[1] Joukhadar C, Frossard M, Mayer BX, Brunner M, Klein N, Siostrzonek P, Eichler HG, Müller M. 2001. Impaired target site penetration of beta-lactams may account for therapeutic failure in patients with septic shock. Crit Care Med 29:385–91.
[2] Roberts JA, Kirkpatrick CMJ, Roberts MS, Dalley AJ, Lipman J. 2010. First-dose and steady-state population pharmacokinetics and pharmacodynamics of piperacillin by continuous or intermittent dosing in critically ill patients with sepsis. Int J Antimicrob Agents 35:156–163.
Reference: PAGE 27 (2018) Abstr 8495 [www.page-meeting.org/?abstract=8495]
Poster: Drug/Disease Modelling - Infection