Variable Linezolid Exposure in Intensive Care Unit PatientsPossible Role of Drug-Drug Interactions
2016 (English)In: Therapeutic Drug Monitoring, ISSN 0163-4356, E-ISSN 1536-3694, Vol. 38, no 5, 573-578 p.Article in journal (Refereed) Published
Background:Standard doses of linezolid may not be suitable for all patient groups. Intensive care unit (ICU) patients in particular may be at risk of inadequate concentrations. This study investigated variability of drug exposure and its potential sources in this population.Methods:Plasma concentrations of linezolid were determined by high-performance liquid chromatography in a convenience sample of 20 ICU patients treated with intravenous linezolid 600 mg twice daily. Ultrafiltration applying physiological conditions (pH 7.4/37 degrees C) was used to determine the unbound fraction. Individual pharmacokinetic (PK) parameters were estimated by population PK modeling. As measures of exposure to linezolid, area under the concentration-time curve (AUC) and trough concentrations (C-min) were calculated and compared with published therapeutic ranges (AUC 200-400 mg*h/L, C-min 2-10 mg/L). Coadministered inhibitors or inducers of cytochrome P450 and/or P-glycoprotein were noted.Results:Data from 18 patients were included into the PK evaluation. Drug exposure was highly variable (median, range: AUC 185, 48-618 mg*h/L, calculated C-min 2.92, 0.0062-18.9 mg/L), and only a minority of patients had values within the target ranges (6 and 7, respectively). AUC and C-min were linearly correlated (R = 0.98), and classification of patients (underexposed/within therapeutic range/overexposed) according to AUC or C-min was concordant in 15 cases. Coadministration of inhibitors was associated with a trend to higher drug exposure, whereas 3 patients treated with levothyroxine showed exceedingly low drug exposure (AUC approximate to 60 mg*h/L, C-min <0.4 mg/L). The median unbound fraction in all 20 patients was 90.9%.Conclusions:Drug exposure after standard doses of linezolid is highly variable and difficult to predict in ICU patients, and therapeutic drug monitoring seems advisable. PK drug-drug interactions might partly be responsible and should be further investigated; protein binding appears to be stable and irrelevant.
Place, publisher, year, edition, pages
2016. Vol. 38, no 5, 573-578 p.
critical illness, pharmacokinetics, Gram-positive infection, antimicrobial agents
IdentifiersURN: urn:nbn:se:uu:diva-307742DOI: 10.1097/FTD.0000000000000324ISI: 000385404400003PubMedID: 27631464OAI: oai:DiVA.org:uu-307742DiVA: diva2:1048381