Official and Alternative Methods for Sampling Food Commodities for Mycotoxin Analysis Within the EU
EC Documentation
EC Regulations setting maximum permissible limits for mycotoxins are normally accompanied by associated Directives on sampling and analysis. These regulations are described in the EC Legislation fact sheet. Commission Directive 98/53/EC as amended by Commission Directive 2002/27/EC lays down sampling methods and methods of analysis for aflatoxins in foodstuffs, while Commission Directive 2002/26/EC lays down sampling methods and methods of analysis for ochratoxin A in foodstuffs. More recently, a Draft Commission Directive laying down the sampling methods and methods of analysis for the official control of patulin in foodstuffs has been prepared.
Settings maximum limits for mycotoxins together with the Directives on sampling and analysis is a continuous and active process between representatives of Member States, so that these are regularly being updated or extended in scope. Maximum limits for aflatoxins in maize and proposals for limits for ochratoxin A in a further list of food commodities have been agreed (for aflatoxins in maize), or are under discussion.
Purpose of Directives for Sampling and Analysis
The plans laid down in the Directives are for official control/enforcement purposes. These are statitistically based as far as possible to ensure that the results obtained by sampling according to these Directives give results that will fall within an accepted and agreed range of the ‘true’ mean. It is essential that the combination of sampling and analysis provides results that are robust and reliable, as consignments moving in international trade may be rejected, and products placed on the market may be recalled on the basis of these results. By inference, they must stand up to legal challenge should the situation arise. Methods of analysis must meet agreed specific criteria but any method meeting these criteria may be used.
Sampling plans have been developed for circumstances ranging from the bulk commodity situation where consignments may be as large as several thousand tones, to the retail stage where packs may be in the 0.1 kg to several kg range.
Alternative Official sampling plans
The Directives allow for alternative sampling to be used in certain circmstances: 1) if it is not possible to carry out the sampling because of the commercial consequences resulting from damage to the lot (this might be due to packaging type such as the use of vacuum or to means of transport and so on); 2) for aflatoxins in nuts other than groundnuts, pistachios and Brazil nuts, dried fruit other than figs, and cereal lots under 50 tons, because the incidence of contamination in these products is low. In these circumstances the sampling method to be applied must be as representative as possible and be fully described and documented.
Other sampling schemes
It is further accepted that alternative, ‘less onerous’ sampling plans could be used by industry for routine quality control and to demonstrate ‘due diligence’. However, in such cases if a dispute should arise, it would be for a court of law to decide whether this approach is acceptable. Therefore, any alternative sampling scheme has to be documented, as detailed as possible, and beyond question. The situation for analysis is no different to that for official sampling plans in that performance criteria are laid down so that any method meeting these criteria can be used.
Practical difficulties associated with following Official sampling plans
The theory and basic criteria required for sampling is described in the Practical Sampling fact sheet and is not discussed here. Problems in using these prescribed procedures may arise for the authorities charged with sampling, or for the analytical laboratories handling the samples. Most of these problems can be overcome by provision of the funding, equipment and staff required. The following examples illustrate some of the concerns that may need to be addressed.
Samples may be taken from large bulk stored commodities such as cereals, from boxes, bags, containers, or from retail packages in food stores. The number of incremental samples required to make the aggregate sample depends on the commodity and consignment size and is typically 100 resulting in 10-30 kg for bulk commodities, or at the other extreme, 10 or more for retail packs to give 1 to 10 kg in total. This regime can be carried out relatively easy for bulk commodities during transfer by flow stream systems equipped with automated sampling, but becomes labour intensive and costly in other situations where access is difficult or manual sampling is necessary.
Samples may need to be retained and stored for subsequent reference purposes under refrigeration at –20 Deg C to avoid breakdown of mycotoxins. Because aggregate samples may be as large as 30 kg (e.g. when sampling figs or maize for aflatoxins), considerable quantities of materials may quickly accumulate. This can equate to almost 0.1 tonnes of sample for a single cargo consisting of 1500 tonnes of maize. Health and Safety regulations in many countries limit the weight of material that can be handled by one person and laboratories have to invest in supporting technical equipment for large quantity sample handling.
Analytical laboratories have been traditionally equipped to handle samples, typically between 0.2 and 1 kg, and the size of laboratory mixers, grinders and macerators has been matched to accommodate these quantities. The need to process up to 30 kg quantities of cereals or dried fruit requires much larger and expensive grinders and equipment for maceration before the analytical sample can be withdrawn. The increased scale requires disposal of large quantities of solvents, and makes it important to avoid chlorinated solvents as far as possible on environmental grounds. Samples are usually required to be stored for some time after the analysis for reference should any subsequent question arise about the result. To ensure that the contaminant is not degraded, samples are usually stored frozen at -20 Deg C, and therefore a laboratory dealing with large numbers of samples will require extensive storage facilities.
Financial costs will include the labour required for sampling and analysis, loss of commodity sampled, equipment and disposables for analysis, and storage costs.
Future
Consideration is being given to legislation for an extended range of mycotoxins in more food commodities. It is thus important that the number of sampling methods is kept to a minimum, so that ideally, one plan is satisfactory for all mycotoxins in a specific commodity. Increasing regulations on mycotoxins in cereals demand validated and internationally agreed automated sampling plans based on sampling during transfer by flow stream systems or conveyor belts. The additional costs could be relatively insignificant compared to those involved with rejection of consignments or law suits over contested analytical results.
Suggested Bibliography
Gilbert, J. and Vargas, E. A. (2003) Advances in sampling and analysis for aflatoxins in food and animal feed. Journal of toxicology – Toxin reviews, 22 (2 &3), 381-422
Johansson, A. S.; Whitaker, T. B.; Giesbrecht, F. G.; Hagler Jr., W. M.; Young, J. H. (2000) Testing shelled corn for aflatoxin, Part III: Evaluating the performance of aflatoxin sampling plans. J. AOAC. Int. 85 (5), 12791284
Coker, R. D., Nagler, M. J., Defize, P. R., Derksen, G. B., Buchholz, H., Putzka, H. A., Hoogland, H. P., Roos, A. H., Boenke, A. (2000) Sampling plans for the determination of aflatoxins B1 in large shipments of animal feedstuffs. J. AOAC. Int., 83 (5), 1252-1258
Schatzki, T.F. (1995) Distribution of aflatoxins in pistachios. 1. Lot distributions. J. Agr. Food. Chem. 43, 1561-1564
Food and Agriculture Organization (1993) Sampling plans for aflatoxin analysis in peanuts and corn. FAO Food and Nutrition Paper 55. FAO, Viale della Terme di Caracalla, Rome, Italy
Park, D. L. and Pohland, A.E. (1989) Sampling and sample preparation dor detection of natural toxicants in food and feed. J. Assoc. Off. Anal. Chem. 72, 399-404