Population Pharmacokinetics of Tigecycline in Critically Ill Patients with Severe Infections

Abstract

We sought to describe the population pharmacokinetics of tigecycline in critically ill patients and to determine optimized dosing regimens of tigecycline for different bacterial infections. This prospective study included 10 critically ill patients given a standard dose of tigecycline. Blood samples were collected during one dosing interval and were analyzed using validated chromatography. Population pharmacokinetics and Monte Carlo dosing simulations were undertaken using Pmetrics. Three target exposures, expressed as ratios of the 24-h area under the curve to MICs (AUC_0-24/MIC), were evaluated (≥17.9 for skin infections, ≥6.96 for intra-abdominal infections, ≥4.5 for hospital-acquired pneumonia). The median age, total body weight, and body mass index (BMI) were 67 years, 69.1 kg, and 24.7 kg/m², respectively. A two-compartment linear model best described the time course of tigecycline concentrations. The parameter estimates (expressed as means ± standard deviations [SD]) from the final model were as follows: clearance (CL), 7.50 ± 1.11 liters/h; volume in the central compartment, 72.50 ± 21.18 liters; rate constant for tigecycline distribution from the central to the peripheral compartment, 0.31 ± 0.16 h^-1; and rate constant for tigecycline distribution from the peripheral to the central compartment, 0.29 ± 0.30 h^-1 A larger BMI was associated with increased CL of tigecycline. Licensed doses were found to be sufficient for Enterobacter cloacae, Escherichia coli, Klebsiella pneumoniae, and methicillin-resistant Staphylococcus aureus for an AUC_0-24/MIC target of 4.5 or 6.96. For a therapeutic target of 17.9, an increased tigecycline dose is required, especially for patients with higher BMI. The dosing requirements of tigecycline differ with the indication, with pathogen susceptibility, and potentially with patient BMI.

Keywords: critically ill patients; population pharmacokinetics; severe infections; tigecycline.

FIG 1

Diagnostic plot for the final population pharmacokinetic covariate model. (Bottom) Observed tigecycline concentrations versus population-predicted tigecycline concentration. (The r², bias, and imprecision values were 0.733, 0.391 mg/liter, and 4.54 mg²/liter².) (Top) Observed tigecycline concentrations versus individual-predicted tigecycline concentration (the r², bias, and imprecision values were 0.944, −0.0768 mg/liter, and 0.975 mg²/liter²).

FIG 2

Visual predictive check of tigecycline serum data.

FIG 3

Probability of target attainment for various loading doses. Dashed lines represent different AUC_0–24/MIC targets of 4.5, 6.96, and 17.9.

FIG 4

Probability of target attainment from 120 to 144 h for a patient with normal weight or obesity administered different tigecycline maintenance doses of 50 mg/12 h, 100 mg/12 h, 150 mg/12 h, and 200 mg/12 h. The PK/PD target is achieved when the PTA value is 90% coverage. The targets for which results are shown are AUC_0–24/MIC ratios of 4.5 (A), 6.96 (B), and 17.9 (C).