ABSTRACT
Bone fragments are essential structures for the detection of processed animal proteins (PAPs) in feed by light microscopy for official controls according to Annex VI of European Union Regulation EC/152/2009. The preparation of samples submitted for analysis requires a grinding step to make them suitable for microscopic slide preparation and observation. However, there are no technical guidelines set down for this step despite the fact that it can lead to an increase in bone numbers due to fragmentation. This was demonstrated by an in-house study carried out by the Irish National Reference Laboratory (NRL) for animal protein detection. The present collaborative study investigated the possible effects of three different grinding conditions on the final result for a feed adulterated with 0.05 and 0.01% (w/w) of PAP. The microscopic analysis either combined or not with an Alizarin Red staining was carried out by 10 different laboratories. The results demonstrated that although a large variation in the numbers of bone fragments was noted, five of the six different grinding/staining combinations applied at two levels of PAP adulteration did not significantly (at p = 0.05) differ from one another. The only exception occurred when grinding the feed containing 0.05% of PAP with a rotor mill equipped with a 0.5-mm sieve and combined with a staining which resulted in a greater number of bone fragments by forced fragmentation. Overall, the impact of the grinding/staining combinations on the final results was shown to be negligible when considering the regulatory limit of detection (LOD) requirement for the method and the current rules of implementation of the light microscopic method. From a total of 180 analyses carried out on the feed matrix containing 0.05% of PAP no false-negative result was observed, and at a level of 0.01% PAP only 10 false-negative results occurred.
