Background

Background: A population PK analysis from 3 linezolid (LZD) trials was performed to estimate PK parameters in patients with CAP and SST infections, and evaluate patient covariate influence on LZD PK.

Methods: Sparse samples were obtained from patients administered low (750 mg/day) or high (1125-1250 mg/day) dosage LZD IV, then orally. A total of 3,238 LZD concentrations from 655 patients were available. Patient covariates were explored with univariate analysis and significant findings were combined into a full multivariable model using forward selection and backward elimination in NONMEM.

Results: The most robust model that could be adequately fit was a one compartment model with first-order absorption (Ka) and parallel linear (Ke) and non-linear (Km, Vm) elimination. The mean parameter estimates (% SEM) were: 0.802 (30.7) hr-1 for Ka, 75.5 (11.7) mg/hr and 378 (16.7) mg for Vm and Km. Although the linear elimination process was 111% higher in non-whites compared to other races, there was no significant effect on total clearance. For a 52-year-old white male weighing 84.4 kg, clearance was estimated to be 6.96 and 4.51 L/hr at usual steady-state concentrations for the low and high dose groups. The non-linear clearance component represented 88.5 and 82.3 percent of total clearance, but significant accumulation is not expected based on 60 day simulations. Men have a 12% higher Vd than women. The Vd increased by a factor of 0.351 for total body weight and decreased by a factor of 0.21 for age.

Conclusions: A one-compartment model with linear and non-linear elimination adequately describes lzd PK. Covariate effects were small and not sufficient to require dosage adjustment. At doses expected in clinical use, extensive accumulation is not expected.

Interscience Conference on Antimicrobial Agents and Chemotherapy (ICAAC); Toronto, Ontario, Canada; September 2000

By B Cirincione, Luann Phillips, Thaddeus H. Grasela, S Sardella, Elizabeth Ludwig, J Bruss, E Antal, B Hafkin, D Stalker