GC/MS detection of central nervous tissues as specified BSE risk material in meat products: validation by an externally controlled blind trial

The addition of central nervous tissues (CNT), such as brain and spinal cord, in the manufacturing of meat products is either forbidden--if the material falls under the legal definition of specified risk material (SRM)--or must be labelled on the packed product. To foster official food control, several CNT detection methods were developed, but only fatty acid patterns as detected by gas chromatography/mass spectrometry (GC/MS) allow the further characterization of the detected CNT as to both the animal species and--surprisingly--the age of the animal from which the CNT was derived in accordance with the legal definition. Complementing a previous report in this journal by Lucker et al. 2010 (doi: 10.1007/s00216-010-3956-5) on CNT quantification by GC/MS, we now report results of the validation of this new analytical approach by an externally controlled blind trial elucidating its potential to identify the species and age of the CNT detected. The 72 samples (24 standards of emulsion-type sausage, each heated in three different batches: 75 degrees C, 30 min; 115 degrees C, 25 min; 133 degrees C, 40 min) containing porcine, ovine or bovine muscle tissue and differing amounts of CNT (bovine or ovine brain: 0, 0.1, 0.5, 1.0, 3.0% m/m) were produced externally and provided blind for analyses to our laboratory. In accordance with the previous study, heating had no detectable effect on the GC/MS analysis. Judged by the present sensitivity of this method (cut-off 0.2% CNT), all of the samples containing 0.5% or more CNT (n = 57, 100%) were identified correctly as CNT-positive. The CNT species was identified correctly in 54 samples (94.7%), with three samples of one specific standard (0.5% ovine CNT) falsely classified as bovine CNT. However, the CNT age of these samples was correctly classified (more than 12 months). Overall, 57 samples (100%) were correctly classified as SRM-positive and 15 samples (100%) as SRM-negative. To the best of our knowledge, this is the first time that a legal demand for the (1) detection of traces of a specific tissue in a food matrix, (2) the identification of its taxonomic origin and (3) the classification of its age has been shown to be analytically possible in totally blind samples. The very positive validation results of this externally controlled blind trial recommend the present GC/MS approach for the detection of CNT in meat products as a reference method. However, our results also demonstrate the need for further studies, in particular to increase sensitivity and to conduct ring trials including more than one laboratory