AFNOR validation of Premi®Test, a microbiological-based screening tube-test for the detection of antimicrobial residues in animal muscle tissue

Abstract

Premi®Test contains viable spores of a strain of Bacillus stearothermophilus which is sensitive to antimicrobial residues, such as beta-lactams, tetracyclines, macrolides and sulphonamides. The growth of the strain is inhibited by the presence of antimicrobial residues in muscle tissue samples. Premi®Test was validated according to AFNOR rules (French Association for Normalisation). The AFNOR validation was based on the comparison of reference methods (French Official method, i.e. four plate test (FPT) and the STAR protocol (five plate test)) with the alternative method (Premi®Test). A preliminary study was conducted in an expert laboratory (Community Reference Laboratory, CRL) on both spiked and incurred samples (field samples). Several method performance criteria (sensitivity, specificity, relative accuracy) were estimated and are discussed, in addition to detection capabilities. Adequate agreement was found between the alternative method and the reference methods. However, Premi®Test was more sensitive to beta-lactams and sulphonamides than the FPT. Subsequently, a collaborative study with 11 laboratories was organised by the CRL. Blank and spiked meat juice samples were sent to participants. The expert laboratory (CRL) statistically analysed the results. It was concluded that Premi® Test could be used for the routine determination of antimicrobial residues in muscle of different animal origin with acceptable analytical performance. The detection capabilities of Premi®Test for beta-lactams (amoxicillin, ceftiofur), one macrolide (tylosin) and tetracycline were at the level of the respective maximum residue limits (MRL) in muscle samples or even lower.

Keywords:

• validation
• Premi®test
• screening
• antimicrobial residues
• muscle
• collaborative study
• routine analysis

Acknowledgements

The authors thank Annie Rault, Anne de Courville, Sophie Gautier and Marie-Pierre Fourmond for their technical participation in the preliminary study and Catherine Creff-Froger, responsible of the Four Plate Test analyses. We also thank Jean-Michel Fabre (Phylum, Labège, France) for help with data analysis. Thanks to Françoise Goeijen (DSM, The Netherlands) and to the six French field veterinary laboratories (LVD) that participated to the preliminary study. Thanks finally to all the participants of the collaborative study.