Due to increasing consumption and production, pharmaceuticals are omnipresent in the environment. Regardless of whether they are present in low concentrations, their detection in the environment is of great concern for human health. Due to the appearance of new pharmaceuticals on the market, there is always a need to develop new analytical methods to determine their presence in various environmental samples. So, in this paper, new LC-MS/MS method was developed for the determination of 13 pharmaceuticals (azithromycin, tiamulin, imatinib, febantel, torasemide, omeprazole, linezolid, praziquantel, etodolac, sulfamethazine, sulfafurazole, albendazole, levamisole) in water and sediments samples.Pharmaceuticals were isolated from water by solid-phase (SPE) and stir-bar sorptive extraction (SBSE), while extracts from sediment samples were obtained by matrix-solid phase dispersion (MSPD) and ultrasound-assisted extraction (UAE). After optimization of extraction conditions, the methods were validated in terms of linearity, LOQ, LOD, precision, recovery and matrix effect. The LOQ of the optimized method were 2.5 ng/L-0.5 µg/L for SPE extraction and 0.1–10 µg/L for SBSE extraction from water, and 2.5 ng/g-25 μg/g for MSPD extraction and 0.025 do 5 µg/g for UAE extraction from sediment depending on the pharmaceuticals. Recoveries were matrix dependent (20–150 %), particularly for SBSE extraction and at the lowest investigated concentration tested, so standard addition calibrations were required for quantification in real samples. The standard addition calibration lines showed high linearity (R2 > 0.97) and replicate extractions showed good reproducibility (RSD≤%).Results obtained show that SPE is much more sensitive than SBSE, while for extraction from sediments both methods are equally sensitive, whereas for some tested pharmaceuticals MSPD is better and for others UAE is better. By comparing the sample preparation methods used, attention is drawn to the fact that, despite the large number of possible methods, not all of them are always equally good with regard to the target components and the type of sample.
Read full abstract