RESUMO
An intercomparison study was conducted to determine the presence of processed animal proteins (PAPs), including meat and bone meal (MBM) from various species, in animal feed. The performances of different methods, such as microscopy, polymerase chain reaction (PCR), immunoassays, and a protocol based on iquid chromatography (LC), were compared. Laboratories were asked to analyze for PAPs from all terrestrial animals and fish (total PAPs); mammalian PAPs; ruminant PAPs; and porcine PAPs. They were free to use their method of choice. In addition, laboratories using microscopy were asked to determine the presence of PAPs from terrestrial animals, which is applicable only to microscopy. For total PAPs microscopy, LC and some immunoassays showed sufficient results at a concentration as low as 0.1% MBM in the feed. In contrast, PCR was not fit for purpose. In differentiating between MBM from terrestrial animals and fishmeal, microscopy detected 0.5% of terrestrial MBM in feed in the presence of 5% fishmeal, but was less successful when the concentration of MBM from terrestrial animals was 0.1%. The animal-specific determination of MBM from mammals or, more specifically from either ruminants or pigs, by PCR showed poor results, as indicated by a high number of false-positive and false-negative results. The only PCR method that scored quite well was applied by a member of the organizer team of the study. Immunoassays scored much better than PCR, showing sufficient sensitivity but some deficiency in terms of specificity. The results also demonstrated that the reliable determination of MBM from ruminants has not been resolved, especially for low concentrations of MBM (0.1%) in feed. Comparison of the results for mammalian MBM from all methods indicated that, for control purposes, the immunoassay method, especially when applied as dipsticks, could be used as a rapid screening method combined with microscopy to confirm the positive samples. However, implementation of such a system would require that the immunoassays were previously validated to demonstrate that this approach is fit for purpose. The determination of ruminant or porcine PAPs by immunoassays was more difficult, partly because the MBM in this study contained about 50% bovine and porcine material, thereby reducing the target concentration level to 0.05%.
