Novel LC-MS/MS method for plasma vancomycin : comparison with immunoassays and clinical impactNovel LC-MS/MS method for plasma vancomycin : comparison with immunoassays and clinical impact
Faculty of Pharmaceutical, Biomedical and Veterinary Sciences. Pharmacy
Clinica chimica acta. - Amsterdam
441(2015), p. 63-70
University of Antwerp
Background: Accurate quantification of vancomycin in plasma is important for adequate dose-adjustment. As literature suggests between-method differences, our first objective was to develop a novel liquid chromatography-tandem mass spectrometry (LC-MS/MS) method for total vancomycin in human plasma and to compare frequently used immunoassays with this method. Secondly, we investigated the clinical impact of between-method quantification differences. Methods: For LC-MS/MS, lithium heparin plasma was extracted by adding a precipitation reagent containing the internal standard (vancomycin-des-leucine). Analysis was performed on an Acquity TQD mass spectrometer equipped with an Acquity UPLC 2795 separations module. Our method was analytically validated and compared with four frequently used immunoassays from four different manufacturers. Vancomycin concentrations were clinically classified as toxic, therapeutic and sub-therapeutic. Clinical discordance was calculated using LC-MS/MS as a reference. Results: A novel LC-MS/MS method using protein precipitation as sole pretreatment and an analysis time of 5.0 min was developed. The assay had a total imprecision of 2.6-8.5%, a limit of quantification of 0.3 mg/L and an accuracy ranging from 101.4 to 111.2%. Using LC-MS/MS as reference, three immunoassays showed a mean proportional difference within 10% and one showed a substantial mean proportional difference of > 20%. Clinical discordant interpretation of the obtained concentrations ranged from 6.1 to 22.2%. Conclusions: We developed a novel LC-MS/MS method for rapid analysis of total vancomycin concentrations in human plasma. Correlation of the method with immunoassays showed a mean proportional difference > 20% for one of the assays, causing discordant clinical interpretation in more than I out of 5 samples. (C) 2014 Elsevier B.V. All rights reserved.