Avaliação de desempenho e validação de ensaios analíticos qualitativos comerciais para triagem de resíduos de antimicrobianos em leite cru

Abstract

The present study sought to validate and to assess the performance of commercial screening tests for antibiotic residues in raw, commingled bovine milk according to The Decision 2002/657/EC of the European Community, which is part of the requirements for performance evaluation and validation of established analytical tests by the Brazilian Ministry of Agriculture, Animal Production and Food Supply (MAPA). In particular, its National Residue Control Plan (PNCR-Milk) was fully observed to define the scope of this study and the references for Maximum Residue Limits (MRL) for each analyte. Thus, five analyte groups were covered (beta-lactams, tetracyclines, sulfonamides, amphenicols and quinolones), yielding 30 PNCR-Milk analytes being examined. Five products belonging to the Charm®ROSA® product line (Rapid One Step Assay, based on immunoreceptors) were assessed, each covering one of the five analyte groups. Also evaluated was the Charm® CowSideII®, a product based on a sensitive indicator microorganism. Results indicate that the products under the Charm® ROSA® line could satisfactorily screen for 26 out of 30 PNCR-Milk analytes, whereas Charm® CowSideII® could satisfactorily screen for 16 PNCR-Milk analytes. In fact, detection capability (CCβ, with β=5%) values for Charm® ROSA® products based on the 26 analytes were for the majority of the cases (20 out 26) equal to, or less than, 0.50 of the respective MRL. Similar performance was obtained for the 17 analytes that were satisfactorily detected by Charm® CowSideII®. Charm® ROSA® products also showed consistent, adequate specificity/selectivity characteristics; furthermore, all products - but the Charm® Chloroamphenicol Test® - can reliably screen milk samples containing the preservatives Azidiol® or Bronopol®, which are generally adopted by the RBQL laboratories’ overseen by MAPA. However, qualified milk samples containing such preservatives yielded false non-compliant results when tested with Charm® CowSideII®. This evidence should preclude the use of Charm® CowSideII® to screen for antibiotic residues in samples containing one of these preservatives. Results for test robustness under small changes of seven test variables were mostly adequate for all products tested. Nonetheless, experimental data support our recommendation to avoid the testing of milk samples showing increased titratable acidity (18°D), or increased fat content (ca. 6%), and, notably, this finding further reinforces the fundamental need to ensure adequate milk agitation prior to sampling, in addition to using milk of fair quality when screening for antibiotic residues. Collectively, the body of evidence hereby generated support an indication for the intelligent, educated insertion of the assessed commercial screening test products to supplement the existing national residue and chemical contaminants control framework for Brazilian milk, as a contribution to national policies in food safety and public health.