Determination of amphenicol antibiotics and their glucuronide metabolites in urine samples using liquid chromatography with quadrupole time-of-flight mass spectrometry

•Amphenicol antibiotics are determined in human urine samples.•Free antibiotics are mainly excreted from the body in the urine as glucuronide conjugates.•An enzymatic hydrolysis allowed discrimination between free and conjugated forms.•The use of LC avoids derivatization of the amphenicols.•The high resolution of QTOF-MS detection provides unequivocal identification.•Detection limits are considerably improved using DLLME. A rapid procedure for the determination of amphenicol antibiotics in human urine by liquid chromatography with quadrupole time-of-flight mass spectrometry (LC-QTOF-MS) is proposed. The presence of thiamphenicol (TAP), florfenicol (FF) and chloramphenicol (CAP) in the human body can be attributed to their administration to treat certain diseases or by eating food of animal origin. The TAP, FF and CAP excreted in urine is mainly in the form of glucuronide conjugates, although their free forms may also be excreted to a lesser extent. In the procedure described, the enzymatic hydrolysis of amphenicol glucuronide forms in urine was carried out using β-glucuronidase and sulfatase at pH 5 (37 °C, overnight) in order to discriminate the free and conjugated forms. Then, amphenicol antibiotics were submitted to dispersive liquid–liquid microextraction (DLLME) for preconcentration. All the parameters affecting DLLME efficiency were optimized, and the following conditions were selected: 0.9 g NaCl in 10 mL of urine, to which 1.2 mL methanol (as dispersant solvent) and 1 mL of 4-methyl-2-pentanone (as extractant solvent) were added. The absence of a matrix effect allowed quantification of the samples against aqueous standards. Detection limits were 29, 6 and 3 pg mL−1 for TAP, FF and CAP, respectively. Relative standard deviations were calculated to evaluate the intra- and inter-day precision and values lower than 10% were obtained in all cases. The trueness of the method was tested through recovery studies, obtaining satisfactory values (83–104%). Ten urine samples obtained from volunteers were analysed and all of them were free of the studied antibiotics.