The use of mycotoxin methodology in practice: a need for harmonization

Background In the EU, sampling and analysis for the official control of the levels of mycotoxins in foodstuffs should be performed in accordance with the methods and criteria set out in Commission Regulation 401/2006. For each mycotoxin, the values of recovery, repeatability and reproducibility of the analytical method selected by each laboratory must fall within the range of acceptability as prescribed in the Regulation. Aims Carry out a survey on current practices concerning the use and application of mycotoxin test methods for what are considered to be the most current commercially significant mycotoxins. Materials and Methods Nineteen control, commercial and research laboratories from 12 countries (United Kingdom, Italy, Belgium, Spain, Germany, The Netherlands, Bulgaria, Hungary, Greece, Turkey, New Zealand and China) participated in a survey of current practices concerning the use and application of methods for the determination of the principal mycotoxins found in foods and subject to regulatory control: [aflatoxins (AFs: AFB(1), AFB(2), AFG(1), AFG(2)), aflatoxin M(1) (AFM(1)) fumonisins (FBs: FB(1), FB(2)), ochratoxin A (OTA), deoxynivalenol (DON), patulin (PAT), zearalenone (ZEA), and T-2 and HT-2 toxins]. Results and Discussion Fourteen of the laboratories surveyed were accredited to ISO 17025:2005 and the accreditation paralleled participation in proficiency testing schemes such as FAPAS (R). Most of the laboratories declared that they received laboratory samples weighing between 0.004-1 kg. The number and types of food matrices analysed for each mycotoxin or group of mycotoxins varied consistently between mycotoxins, laboratories and countries. In general the highest number of food matrices capable of being assessed for a particular mycotoxin was for OTA followed - in decreasing order - by AFs, DON, ZEA, T-2/HT-2 toxins, FBs, AFM(1) and PAT. Analysis for OTA, AFs, PAT, ZEA, DON, FBs, T-2/HT-2 toxins and AFM(1), were performed in 95%, 84%, 74%, 74%, 63%, 58%, 58% and 53% of the laboratories, respectively. Most laboratories stated that they used HPLC coupled with either a fluorometer, ultraviolet or mass spectrometric (MS) detectors for detection and quantification of mycotoxins. Only one laboratory used GC/MS for analysis of T-2 and HT-2 toxins whereas two laboratories used TLC based methods for the determination of all mycotoxins except fumonisins. The use of LC-MS methodology by eight laboratories is remarkable because LC/MS is not an official method for mycotoxins within a CEN or AOAC context. Some mycotoxins are not amenable to all detection techniques reported above. ELISA kits were used in three laboratories for the analysis of AFs, OTA, ZEA, DON, FBs and/or T-2/HT-2 toxins. Several other test kits were used in one laboratory for the determination of OTA (six different test kits) and DON (eight different test kits). Six different definitions of limit of detection (LOD) and nine different definitions of limit of quantification (LOQ) have been reported by participating laboratories with the signal/noise ratio being the most popular (used by 40% of laboratories). In some cases the values of LOD, LOQ and measurement uncertainty for the same mycotoxin varied from laboratory to laboratory. In particular, a large variability of measurement uncertainty was noted that was probably due to non-harmonized interpretation of the term. Conclusion This survey suggests that the primary issues needing to be harmonized are: accreditation needed, ppropriate size of laboratory sample, guidelines on the most convenient analytical method for each combination of mycotoxins/matrix, use of method validated through a collaborative study, participation in proficiency testing, use of reference/certified materials/standard solutions, use of the same definition/calculation for LOD, LOQ, recovery and measurement uncertainty.