Population Pharmacokinetics of High-Dose Tigecycline in Patients with Sepsis or Septic Shock

Abstract

Tigecycline is a glycylcycline often used in critically ill patients as the antibiotic of last resort. The pharmacokinetics (PK) of tigecycline in intensive care unit (ICU) patients can be affected by severe pathophysiological changes so that standard dosing might not be adequate. The aim of this study was to describe population PK of high-dose tigecycline in patients with sepsis or septic shock and evaluate the relationship between individual PK parameters and patient covariates. The study population consisted of 37 adult ICU patients receiving a 200-mg loading dose of tigecycline followed by multiple doses of 100 mg every 12 h. Blood samples were collected at 0.5, 2, 4, 8, and 12 h after dose administration. A two-compartment model with interindividual (IIV) and interoccasion (IOV) variability in PK parameters was used to describe the concentration-time course of tigecycline. The estimated values of mean population PK parameters were 22.1 liters/h and 69.4 liters/h for elimination and intercompartmental clearance, respectively, and 162 liters and 87.9 liters for volume of the central and peripheral compartment, respectively. The IIV and IOV in clearance were less than 20%. The estimated values of distribution volumes were different from previously published values, which might be due to pathophysiological changes in ICU patients. No systematic relationship between individual PK parameters and patient covariates was found. The developed model does not show evidence that individual tigecycline dosing adjustment based on patient covariates is necessary to obtain the same target concentration in patients with sepsis or septic shock. Dosing adjustments should be based on the pathogens, their susceptibility, and PK targets.

Keywords: pharmacokinetics; population pharmacokinetics.

FIG 1

Individual tigecycline concentration-time profiles of the studied population.

FIG 2

Visual predictive check showing the simulation-based 90% confidence intervals around the 5th, 50th, and 95th percentiles of the PK data in the form of turquoise (50th) and violet (5th and 95th) areas; the corresponding percentiles from the observed data are plotted in black color.

FIG 3

Estimates of individual PK parameters from the final model in relation to time-independent continuous covariates and categorical covariates. The gray lines indicate the trend in the data (using locally weighted scatterplot smoothing [LOESS]). CL, clearance; V_ss, volume of distribution at steady state; ECMO, extracorporeal membrane oxygenation; CRRT, continuous renal replacement therapy; SDT, started during therapy.

FIG 4

Individual estimates for kappa of the final PK parameters in relation to time-dependent covariates. The gray lines indicate the trend in the data (using LOESS). κ, deviation of individual parameters between occasions; CL, clearance; V₂, volume of distribution of the peripheral compartment; ELWI, extravascular lung water index; CO, cardiac output; SOFA, sequential organ failure assessment score; PCT, procalcitonin concentration.

FIG 5

Comparison of the population PK parameter values estimated in our study and the studies of Van Wart et al. (18), Rubino et al. (19), and Ramirez et al. (16). The boxplots show the mean values and standard deviations of the parameter. For the study by Rubino et al. (19), the standard deviations of the mean clearance estimates were obtained from bootstrap results. CL, clearance; V_ss, volume of distribution at steady state; Q, intercompartmental clearance; V₁, volume of distribution of the central compartment; V₂, volume of distribution of the peripheral compartment.

FIG 6

Simulation of tigecycline concentrations after 200-mg loading dose and seven subsequent 100-mg doses in 12-h intervals based on typical population PK parameters from the model developed in this study (black solid lines) and the models published by Van Wart et al. (18), Rubino et al. (19), and Ramirez et al. (16) (red dashed lines).