In this work analytical methods for the determination and identification of residues of 11 glucocorticosteroids have been developed: dexamethasone (DEX), betamethasone (BET), beclomethasone (BCL), prednisolone (PLN), prednisone, methylprednisolone (MPLN), flumethasone (FLU) and triamcinolone (TRI), triamcinolone acetonide, fludrocortisone, cortisone. Each step of these methods has been optimised in different biological matrixes, by means of comparison studies modifying the different critical parameters. Detection, confirmation and identification methods are based on liquid-chromatography with tandem-mass spectrometry (LC-MS/MS).
Extraction and sample cleanup methods were developed for urine, liver, kidney and muscle samples. In order to achieve the highest concentrations of unconjugated compounds as possible, a 4 h (liver, kidney) or overnight incubation of the samples was necessary with Helix pomatia juice. Further extraction contained a liquid extraction followed by a solid phase extraction by either C18 SPE cartridges or OASIS columns. The extraction of hair samples is still under investigation.
As for the chromatographic separation, it was important that epimers beta- and dexamethasone were separate adequately by using an appropriate HPLC column. More than 25 different combinations of columns and mobile phases have been studied in an HPLC-Diode array system. Best results for the separation of beta- and dexamethasone were obtained using a Hypercarb column, a column designed specifically for the separation of structurally similar compounds. The Synergi column also gave good results.
As for the detection based on tandem mass spectrometry, electrospray (ESI) was used as ionisation mode. Postive ESI gave good gave good results for some glucocorticoids, but not for all. The negative ion mode however gave much more specific fragmentation, more specifically, the loss of formaldehyde, which good be found for all analysed glucocorticoids.
The complete LC-MS/MS methods for all matrices(liver, kideny, muscle and urine) were validated according to European criteria. Detection capabilities (CCb)were below 0.86 µg/kg for all matrices or for compounds with an established MRL, the CCbs were not higher than 0.62 µg/kg above these MRLs. Repeatabilities and trueness were very satisfactory. A ringtest was organised amongst several European Laboratories and results were evaluated. From this ringtest it could be concluded that the developed LC-MS/MS methods were very reliable and very reproducible.
The developed methods have been applied in the Spanish National Monitoring Plan for Veterinary Drug Residues and will also be applied in the routine analyses of the LAboratory of Food analysis of the Ghent University.