Certified reference materials for testing of the presence/absence of Staphylococcus aureus enterotoxin A (SEA) in cheese : certification report

This report describes the preparation of three cheese powder matrix reference materials (IRMM-359a-c) and their certification for testing of the presence/absence of Staphylococcus aureus enterotoxin A (SEA). Raw milk cheese was decrusted, cut into cubes, chopped in a kitchen-type food processor for a short time, freeze-dried, cryogenically milled, and mixed (blank material IRMM-359a). Moreover, a second portion of raw milk cheese was decrusted and cut into cubes. After addition of water and spiking with a solution of SEA, the sample was homogenised using a high-speed grinder (Ultra-Turrax). The cheese slurry was freeze-dried, cryogenically milled and mixed with blank cheese powder to obtain the two SEA-containing materials at SEA target levels of 0.1 and 0.25 ng/g cheese, respectively (IRMM-359b, IRMM-359c). Between unit-homogeneity was quantified and stability during dispatch and storage were assessed in accordance with ISO Guide 35:2006 [1]. The minimum sample intake is 15.1 g cheese powder (representing 25 g of cheese after reconstitution) per replicate analysis (n=5), as stipulated in Commission Regulation 1441/2007 [2], and therefore no dedicated study on the minimum sample intake was performed. The reference material was characterised in an interlaboratory comparison of laboratories of demonstrated competence and adhering to ISO/IEC 17025 [3] and using the European Screening Method with the VIDAS SET2 and the Ridascreen SET Total for detection (further on named ESM/VIDAS and ESM/Ridascreen, respectively) [4]. Technically invalid results were removed, but no outlier was eliminated on statistical grounds only. Certified values are reported as probability of detection and expressed as either diagnostic specificity (ratio of true negatives divided by the sum of true negatives and false positives) for the blank material, or diagnostic sensitivity (ration of true positives divided by the sum of true positives and false negatives) for the SEA-containing materials. Uncertainties for homogeneity and stability were estimated, but not used for an uncertainty budget due to the nature of the certified values (presence/absence certification). Instead, the certified values are expressed as intervals with a 95% level of confidence. The preparation and processing of the material, homogeneity and stability studies, and the characterisation are described hereafter and the results are discussed.