Development of a multi-analyte method for the determination of anabolic hormones in bovine urine by isotope-labelled GC-MS/MS

The use of natural and synthetic hormones for growth promotion purposes in meat producing animals is prohibited in the European Community since 1986 to protect consumers from possible developmental, neurobiological, genotoxic and carcinogenic effects due to the intake of hormone residues and their metabolites. Consequently, every EU Member State has to monitor a set proportion of the total annual production of different animal food commodities for their anabolic residues, determining them either in biological fluids (blood or urine) or muscle tissue and edible organs. Screening analysis of anabolic hormones using immunoassays suffer from cross-reactivity with structurally related hormones. In order to reduce analytical interferences and improve accuracy, we developed a method using gas chromatography coupled to ion trap tandem mass spectrometry (GC-MS/MS) for the simultaneous determination of 11 anabolics in bovine urine (hexestrol, diethylstilbestrol, dienestrol, 17α-estradiol, 17β-estradiol, 17α-ethynylestradiol, 19-nortestosterone, 17α-methyltestosterone, β-testosterone, α-zeranol and β-zeranol). The assay consists of a simple sample preparation procedure, which utilizes only one solid-phase extraction step and one liquid-liquid extraction step for sample clean-up prior to derivatization for positive electron impact GC-MS/MS analysis. Two types of derivatization agents were assayed, N,O-bis(trimethylsilyl)trifluoroacetamide (BSTFA) with 1% trimethylchlorosilane and heptafluorobutyric anhydride (HFBA). The three possible acquisition modes of the GC-MS system, Full Scan, selective ion monitoring (SIM) and tandem mass spectrometry (MS/MS), were compared and best results were obtained in the MS/MS mode in the narrow concentration range of 0.25 to 8.0 ng/mL. Better results and good linearity were obtained using BSTFA as a derivatization agent by which no matrix effects were observed. The limits of detection were in the range of 0.005 ng/mL (diethylstilbestrol) to 0.38 ng/mL (17α-methyltestosterone and 17α-ethynylestradiol). Recoveries varied from 81 ± 4% (α-zeranol) to 149 ± 24% (17α-methyltestosterone) using the MS/MS mode. Repeatability values were obtained from 4 replicate analysis of spiked urine samples at 1 ng/mL and ranged from 2.1 to 22.7%.