Development of UHPLC-MS/MS methods to quantify 25 antihypertensive drugs in serum in a cohort of patients treated for hypertension
AuthorThorstensen, Christian W.; Clasen, Per-Erik; Rognstad, Stine; Haldsrud, Renate; Føreid, Siri; Helstrøm, Trine; Bergland, Ola Undrum; Halvorsen, Lene Vernås; Aune, Arleen; Olsen, Erik; Brobak, Karl Marius; Høieggen, Aud; Gustavsen, Ingebjørg G.; Larstorp, Anne Cecilie Kjeldsen; Søraas, Camilla Lund; Opdal, Mimi Stokke
We developed three ultra-high pressure liquid chromatography coupled to mass spectrometry detection (UHPLCMS/MS) methods to quantify 25 antihypertensive drugs in serum samples. Patient-reported drug lists were collected, and drug concentrations were analysed in samples from 547 patients, half with uncontrolled hypertension, and all treated with ≥ 2 antihypertensive drugs. For sample preparation, serum was mixed with deuterated internal standards and acetonitrile and precipitated. Aliquots of the supernatant were injected on UHPLC-MSMS with a C18 reversed phase column. The mobile phase was 0.1 % HCOOH (formic acid) in water and 0.1 % HCOOH in acetonitrile (except in methanol for spironolactone/canrenone) at a flow rate of 0.4 mL/min. The calibrators and internal controls were prepared in Autonorm™. The calibration ranges were wide, and the models were linear or quadratic with squared correlation coefficients ≥ 0.97. The limits of detection and quantification, specificity, carry-over, and matrix effects were acceptable. The accuracy of the internal controls was in the range 85–121 %, and the intermediate precision for all drugs was 4–28 %. The patient-reported antihypertensive drug use and the detected serum drug concentrations were in accordance with that most frequently prescribed nationally. The percent non-detectable level was 5–10 % for bendroflumethiazide, doxazosin, nifedipine, and ramipril. Often the drug dose chosen was lower than the recommended maximum daily dose. We report the maximum (Cmax) and minimum (Cmin) drug concentrations after drug intake. The inter-individual pharmacokinetic variability at Cmin was 18-fold for hydrochlorothiazide, 22-fold for losartan carboxyl acid, 26-fold for amlodipine, 44-fold for candesartan, and 50-fold for valsartan. Our methods are suitable for measuring antihypertensive drugs in patient serum for therapy control.