Linezolid plasma and intrapulmonary concentrations in critically ill obese patients with ventilator-associated pneumonia: intermittent vs continuous administration.

Abstract

Clinical application of an antibiotic's pharmacokinetic/pharmacodynamic (PK/PD) properties may improve the outcome of severe infections. No data are available on the use of linezolid (LNZ) continuous infusion in critically ill obese patients affected by ventilator-associated pneumonia (VAP). We conducted a prospective randomized controlled trial to compare LNZ concentrations in plasma and epithelial lining fluid (ELF), when administered by intermittent and continuous infusion (II, CI), in obese critically ill patients affected by VAP. Twenty-two critically ill obese patients were enrolled. At the steady state, in the II group, mean ± SD total and unbound maximum-minimum concentrations (C max/C max,u - C min/Cmin,u) were 10 ± 3.7/6.8 ± 2.6 mg/L and 1.7 ± 1.1/1.2 ± 0.8 mg/L, respectively. In the CI group, the mean ± SD total and unbound plasma concentrations (C ss and C ss,u) were 6.2 ± 2.3 and 4.3 ± 1.6 mg/L, respectively. Within a minimum inhibitory concentration (MIC) range of 1-4 mg/L, the median (IQR) time LNZ plasma concentration persisted above MIC (% T > MIC) was significantly higher in the CI than the II group [100 (100-100) vs 100 (89-100), p = 0.05; 100 (100-100) vs 82 (54.8-98.8), p = 0.009; 100 (74.2-100) vs 33 (30.2-78.5), p = 0.005; respectively]. Pulmonary penetration (%) was higher in the CI group, as confirmed by a Monte Carlo simulation [98.8 (IQR 93.8-104.3) vs 87.1 (IQR 78.7-95.4); p < 0.001]. In critically ill obese patients affected by VAP, LNZ CI may overcome the limits of standard administration but these advantages are less evident with difficult to treat pathogens (MIC = 4 mg/L). These data support the usefulness of LNZ continuous infusion, combined with therapeutic drug monitoring (TDM), in selected critically ill populations.