Retail oat products survey

Friday 6 February 2004

The preliminary results of an ongoing research study on contamination of oats at harvest by fusarium toxins indicated the potential for retail oat products to be contaminated by trichothecene toxins.

A survey of retail oat products was carried out by the Agency between May and October 2003. The results indicated that exposure to these toxins from this food group in the UK diet is very low. The samples for the survey were purchased between May and September 2003 from a range of UK retail outlets selected at random.

The purpose of the Agency's food surveys is varied. In this case the survey was carried out to monitor consumer dietary exposure against safety limits, but all are planned carefully to reflect market share and range.

In this survey, 335 samples of retail oat products were analysed for the trichothecenes; deoxynivalenol (DON), 3-acetyl deoxynivalenol (3-AcDON), 15-acetyl deoxynivalenol (15-AcDON), diacetoxyscirpenol (DAS), fusarenon X (Fus X), nivalenol (NIV), neosolaniol (NEO), T-2 toxin (T2), HT-2 toxin (HT2) and T-2 triol (T23). All the samples were analysed by RHM Technology Ltd (RHM) using validated methods of analysis.

Trichothecenes were quantified in 173* of 335 samples (limit of quantification 10µg/kg) at levels ranging from 10 to 404 micrograms/kg. A total of 6 trichothecene mycotoxins were detected, with DON, T2 and HT2 being the most common.

There are currently no regulatory limits for trichothecenes. The levels of trichothecenes present in the samples were assessed against the proposed EC limits currently under discussion, which are as follows:

• Deoxynivalenol: 500µg/kg for cereal products and 100µg/kg for cereal-based foods for infants and young children.
• T-2 + HT-2 toxins: 200µg/kg for cereal products and 50µg/kg for cereal-based foods for infants and young children.

Background
Fusarium toxins are naturally occuring chemicals produced by a particular type of mould that can grow on cereal crops such as oats. These mycotoxins are chemically stable and survive food processing, posing a potential risk to human health. Trichothecenes are acutely toxic to humans causing sickness and diarrhoea and in extreme cases death. However, this worst case scenario is not applicable to UK consumers and products. It relates to people in developing countries eating a primarily cereal based diet with contamination levels far higher than those found in the UK.* Chronic effects in animals have been reported to include immunosuppression and teratogenicity.

Those trichothecenes, which are most frequently found as contaminants in cereals and cereal products, were recently considered by the SCF and it set a TDI of 1 µ/kg bodyweight per day for DON, a temporary TDI of 0.7 µ/kg bodyweight per day for NIV and a combined temporary TDI of 0.06µ/kg bodyweight per day for T2 and HT2. The TDI is an estimate of the amount of contaminant expressed on a bodyweight basis that can be ingested daily over a lifetime without appreciable health risks. The results of this survey indicate that human exposure to fusarium toxins is very low in the UK.

Sampling
In this survey 335 samples of retail oat products (for example baby food, rolled/porridge oats, snack bars and flapjacks) were analysed. The samples were purchased between May and September 2003 by Ventress Technical Services Ltd from retail out lets across the UK. The details of each sample were recorded on receipt and are given in Table 1. The absence of a particular brand from this table means only that the product was not included in the survey. The samples were selected to include a wide range of stores; both supermarkets, smaller shops and market stalls. Although the samples were taken to reflect market share, a wide range of brands were covered in order to ensure that the survey was representative of the supply of the products to consumers in the UK.

For mycotoxin surveys carried out by the Agency, a representative retail sample, in this case of approximately 1kg in weight, is taken. Where the product was sold in packs of less than 1kg, a number of retail packs were purchased, ensuring that all came from the same batch, and these were mixed thoroughly before analysis. In a small number of cases the total sample weight was significantly less than 1kg. However, this still represented the combination of individual retail packs from a single production batch.

Methodology
All the samples were delivered to RHM at ambient temperature and, after immediate sample preparation, were stored at -16°C until required for analysis. Samples were allowed to defrost to ambient temperature prior to analysis and returned to -16°C immediately after analysis.

All samples in this survey were ground to a fine powder or aqueous homogenates were prepared by blending at high speed 1kg of product with 1litre of de-ionised water in a Waring blender to produce a slurry. Samples were mixed thoroughly to ensure homogeneity prior to analysis. After homogenisation the sample was split equally among three containers 'A', 'B' and 'C' and stored in a freezer at -16°C. Sample A was used for the analyses and samples B and C were stored frozen; one for analysis by the manufacturer/retailer at their request and the other for referee purposes.

Details of the analytical procedures used and the limit of quantification (LOQ) are summarised in Annexe 1. All results were corrected for recovery and the recovery values are also detailed in Annexe 1.

Quality assurance
The methods described are UKAS accredited and have been validated. Evidence has been, and continues to be, generated to demonstrate that the methods do what is intended and are reliable and reproducible. Control of mycotoxin analyses is monitored using naturally contaminated in-house reference materials and by participation in UK and EU proficiency schemes, collaborative and inter-comparison trials. All analyses were conducted with spiked samples, i.e. a known amount of toxin was added to each sample matrix prior to extraction on each day. These samples were used to assess recovery, with recoveries between 70% and 110% classed as valid.

RHM has participated in inter-comparison exercises organised by FAPAS for deoxynivalenol with satisfactory results.

Measurement uncertainty
The analytical results have a variability known as measurement uncertainty. For any analytical method each result is reported as the best estimate for the sample and it is always qualified by a measurement uncertainty, e.g. x microgram/kg ± y microgram/kg. For this survey the calculation of the measurement uncertainty was carried out using in-house data, performance in international collaborative trials and FAPAS thus incorporating repeatability and reproducibility data. The expanded measurement of uncertainty was calculated using a standard coverage factor of 2, equivalent to a confidence of approximately 95% that the actual level of the mycotoxin being measured lies within the quoted range. The assessment of measurement uncertainty was targeted at samples containing mycotoxins at a range of levels. The standard measurement of uncertainty for these analyses was assessed to be ±42µg/kg at a level of 320µg/kg for DON. The expanded measurement of uncertainty for these analyses was assessed to be ±84µg/kg at a level of 320µg/kg for DON.

Results
The full results of the main survey are detailed in Table 2. All results are given as the best estimate value, corrected for recovery, but do not take into account the measurement uncertainty. Food Standards Agency surveys of contaminants are usually reported in this manner since the best estimate values are used to calculate dietary intakes of the various contaminants. Where surveys are conducted for other purposes, the data are interpreted differently. In their own surveys, for statutory enforcement purposes, public analysts and enforcement authorities always take the uncertainty associated with an analytical measurement into account when considering whether a product breaches regulatory limits. As a result of this they only take action when they are sure that levels of contaminants are above the regulatory limits. The Agency has followed their practice and taken action where the best estimate, less measurement uncertainty, exceeded the proposed guideline limit.

Dietary exposure estimates
Dietary exposure of mycotoxins can be estimated for average (mean) consumers and for those who eat significantly more than average (high level consumers). These estimates can be compared with the TDIs assigned by the Scientific Committee on Food (SCF) and used as a guideline to assess whether mycotoxin exposure through cereal consumption is a risk to consumer health. Exposure assessments were only conducted for those mycotoxins that have a temporary TDI or TDI assigned to them by the SCF, and where sufficient data were available to allow a meaningful analysis.

The Food Standards Agency's INTAKE computer programme was used to combine mycotoxin concentration of retail oats products with food consumption data from the National Diet and Nutrition Surveys of adults, young people aged 4 to 18 years, toddlers aged 1.5 to 4 years and infants aged from 6 to 12 months. Exposure estimates were calculated for all mycotoxins.

Estimates of exposure to trichothecenes, based on the results of this survey, for all population groups (adults, young people and toddlers) indicate that for all groups, exposures are below the tolerable daily intakes established by the European Commission's SCF.

Conclusion
Trichothecene toxins were quantified in 184 of 335 samples although, in most cases, the levels found were low. A total of 6 trichothecene mycotoxins were detected, with DON, T2 and HT2 being the most common. Estimates of exposure based on the results of this survey indicate that exposure to these toxins from this food group in the UK diet is very low.

Fungal contamination of crops with consequent exposure of the population to mycotoxins is a hazard that has always existed for humans. The only way to avoid exposure to mycotoxins is to completely stop eating the foods in which they are likely to occur. However, to do so would eliminate nutritionally valuable foods from our diet. The Food Standards Agency is of the opinion that provided the level of mycotoxins are kept to the lowest levels achievable, the benefits of eating cereal based foods outweigh any possible risks to health. This survey does not raise any safety concerns and consequently consumers do not need to change their diets as a result of these findings.

*Updated 11 February 2004