Summary

The Food Monitoring Scheme is a system of repeated representative measurements and evaluations of levels of undesirable substances in and on foodstuffs, including residues of plant protection products, pesticides and veterinary drugs, heavy metals and other contaminants.

Since 2003, the food monitoring scheme has been made up of two complementary analytic programmes. One consists in examination of foodstuffs selected from a market basket developed on the basis of a statistical analysis of dietary habits [Schroeter A, Sommerfeld G, Klein H, Hübner D (1999) Warenkorb für das Lebensmittelmonitoring in der Bundesrepublik Deutschland (Market Basket for Food Monitoring Purposes in the Federal Republic of Germany). Bundesgesundheitsblatt (Federal Health Bulletin) 1-1999, 77-83], with the aim to watch the situation of contamination and residues under representative sampling conditions (market basket monitoring). The other programme consists in examination of particular problems in the framework of special projects (project monitoring). A total of 4,947 samples of domestic and foreign origins were analysed in 2009 in the framework of both programmes.

The following foodstuffs were selected from the market basket:

Food of animal origin

• Goat cheese
• Butter
• Full egg (liquid)
• Butter mackerel (butterfish)

Food of plant origin

• Vegetable margarine
• Wheat grains
• Bulgur wheat
• Spelt wheat flakes
• Lentils
• Chips [French fries] (cooked)
• Rucola (Rocket)
• Cauliflower
• Sweet peppers
• Aubergines
• Celeriac
• Peas (deep frozen)
• Table grapes
• Apricots
• Bananas
• Orange juice

Depending on what undesirable substances were to be expected, the foods were analysed for residues of plant protection products (insecticides, fungicides, herbicides), veterinary drugs, contaminants (for instance, persistent organo-chlorine compounds, musk compounds, elements, nitrate, mycotoxins), and toxic reaction products.

Project monitoring dealt with the seven following subjects:

• Fumonisins in foodstuffs
• Ergot alkaloids in rye products
• Deoxynivalenol in wheat-based cookies
• Ochratoxin A in instant coffee
• Furan in coffee and coffee extracts
• Analyses of single heads of lettuce for residues of plant protection products
• Dioxins and PCB in liver and meat/muscles of sheep

Interpretation of findings included a comparison with findings from previous years, where this was possible. Yet, we explicitly stress that all statements and evaluation about contamination of foodstuffs made in this report solely refer to the foodstuffs and substances or substance groups analysed in 2009. It is not possible to assess the overall exposure to certain substances, because only part of the market basket can be considered in the analyses of one year, while the substances analysed also occur in other foodstuffs.

Generally, the findings of the 2009 food monitoring programme again support the recommendation that nutrition should be manifold and balanced in order to minimise the dietary intake of undesirable substances, which is unavoidable to some degree.

In particular, findings from the 2009 market basket and project monitoring programmes are summarised as follows:

Residues of plant protection products and pesticides

Food of animal origin

Mainly, residue findings in goat cheese, butter, full egg and butter mackerel were again such of ubiquitous, persistent organo-chlorine insecticides which were extensively used in the past and have entered the food chain via environmental contamination. Yet, as a whole, all products showed declining residue concentrations compared to earlier monitoring studies. The residues found did not indicate any acute health risks to consumers. Non-compliance with maximum residue levels was only found in butter mackerel. Here, 13 samples (12%) exceeded the legal maximum residue level (MRL) for mirex in fish (0.01 mg/kg), which is roughly at the analytic quantification level.

Food of plant origin

Residues of plant protection products were found to different degrees in all foodstuffs of vegetable origin analysed therefor. While only 10% of samples of vegetable margarine contained quantifiably residues, this was the case in about 50% of the samples of deep-frozen peas and even 65% of samples of wheat grains, cauliflower, aubergines, celeriac, and orange juice. Even more frequent (about 80% of samples) were quantifiable findings in banana (with peel) and sweet peppers.

Lentils, rucola, table grapes, and apricots contained plant protection product residues in more than 90% of samples. Table grapes, apricots and rucola were also those with the most frequent multiple residues. The highest number was 14 substances found in a sample of table grapes.
Seven per cent of samples of foods of domestic production gave reason to suspect unauthorised use of plant protection products, mostly in deep-frozen peas and rucola.

Wheat grains, cauliflower, bananas and orange juice did not contain residues above MRLs. In the other analysed foods – apart from rucola – non-compliance with MRLs was in the range between 1.6 and 4.7% of samples. In rucola, this percentage was higher, with about 9.4%. As in previous years, the percentage of samples not complying with MRLs in 2009 was clearly lower in food of domestic production with 1.7%, compared to food from other EU countries (2.9%), or Third Countries (5.5%).
Exposure estimates and more detailed risk assessments led to the identification of one finding in sweet peppers from Turkey and one finding each in apricots from Turkey and France which meant a complete exhaustion of the acute reference doses (ARfD) of oxamyl, carbofuran, and methomyl/thiodicarb, respectively.

In order to assess the influence of the distribution of residues on actual short-term exposure of consumers, special analyses in single heads of lettuce were carried out, to compare results with those obtained in the usual procedure, which is to analyse homogenised mixed samples consisting of at least 10 heads of lettuce. Although the analysis of single heads of lettuce produced higher exposure to 19 out of 109 residues (17%) than the exposure stated after the usual procedure based on homogenised mixed samples and a variance factor, this did not noticeably influence the assessment of the acute risk, because ARfD coverage was still extremely low.

The low ARfD coverage in all residues allowed safely excluding an acute risk to health from these residues. Regarding those residues with ARfD coverage higher than 10%, the usual exposure estimate model (based on the results from homogenised mixed samples) is the even safer model, as the results obtained in analyses of single heads of lettuce would have led to even lower exposure estimates.

Regarding such single piece analyses, one has to balance the high expense (ten individual processing and measurement procedures) against the profit in terms of information suitable for acute risk assessment. None of the single piece studies produced an exposure finding which would have meant a higher consumer acute risk estimate. As a result we can state that the risk estimate model usually applied in official food control guarantees a high level of consumer protection.

Residues of pharmacologically active substances

Chicken eggs carry only low levels of residues of coccidiostats. This is the result of analyses of 68 samples of full eggs for coccidiostats, nicotine, and cotinine, which produced residues findings of nicarbazin below the MRL in two of the samples.

Persistent organo-chlorine compounds

Dioxins and polychlorinated biphenyls (PCB) are present everywhere in the environment at low levels. This leads to what is called unavoidable background contamination, also in foods. Analytic results about non-dioxin-like PCB (ndl-PCB) indicate that contamination of most of the food groups tested in the framework of market-basket monitoring is very low. There was only one sample of vegetable margarine which showed an increased level of ndl-PCB.

The 2011-2015 Monitoring Plan provides for further such tests because such contamination cannot be generally precluded. Over a medium term, we believe it is necessary to reach EU-wide harmonisation of the different maximum levels now permitted in individual states, including very strict regulations with regard to ndl-PCB in foods. Food monitoring contributes to this end by consolidating the necessary data basis about background contamination of foods.

The findings of the monitoring project “Dioxins and PCB in liver and muscle of sheep” confirm earlier findings by some Bundesländer that a large proportion of livers of sheep exceeds the maximum levels set by Regulation (EC) No. 1881/2006 for dioxins and for the total of dioxins and dioxin-like PCB (dl-PCB). This monitoring project found that some 70% of sheep livers tested throughout Germany exceeded EU maximum levels. As regards muscle meat of sheep, in contrast, there were only few cases of non-compliance with EU maximum levels of these contaminants.

The findings of the monitoring project also show that there is a correlation between contamination of liver and muscle, and in particular, that lower concentrations of dl-PCB in liver correlate with increased levels of dl-PCB in muscle, because accumulation factors in liver are clearly lower. With a concentration of about 50 pg WHO-PCDD/F-PCB-TEQ/g fat in sheep liver and more, one can no longer preclude that the sum of residues of dioxins and dioxin-like PCB exceeds the EU maximum permissible level in sheep meat.

On the basis of present findings, the Federal Institute of Risk Assessment (Bundesinstitut für Risikobewertung, BfR) keeps up its precautionary recommendation to avoid eating sheep liver. In contrast to that, sheep meat is much less contaminated with dioxins and dioxin-like PCB and can be normally consumed without concern.

Highly volatile halogenated hydrocarbons

Vegetable margarine did not contain any quantifiably findings of the highly volatile chlorinated hydrocarbons (CHC) trichloromethane, tetrachloroethene and trichloroethene. Trichloroethene was also not detected in butter and goat cheese. Tetrachloroethene was detected at a very low level only in one sample of butter. Trichloromethane, in contrast, was detected in nearly every second sample of butter or goat cheese, with concentrations sometimes amounting to more than 50% of the legal maximum level, and once even exceeding the legal maximum level in goat cheese.

BTEX (aromatic hydrocarbons)

A number of foods were analysed for the aromatic hydrocarbons benzene, toluene, ethylbenzene, xylene, and styrene, collectively named BTEX. Vegetable margarine did not have any BTEX findings. Goat cheese and butter did not have any findings of benzene and ethyl benzene, and only few quantifiable findings of xylene and styrene. The highest levels found were 121 µg/kg xylene and 1300 µg/kg styrene in goat cheese. Toluene was the aromatic hydrocarbon most frequently found in goat cheese and butter, with levels of up to 140 and 396 µg/kg, respectively. The BTEX findings in butter confirmed the findings of the 2006 national monitoring.

Polycyclic aromatic hydrocarbons

Butter mackerel and vegetable margarine were tested for polycyclic aromatic hydrocarbons (PAH) in the monitoring programme for the first time in 2009. The findings here were not significantly different from earlier findings in comparable foodstuffs. Chrysene was found and quantified most frequently in both foods, benzo(a)pyrene least frequently. In this context, it showed again that it makes sense to consider the four leading substances in this group – benzo(a)anthracene, benzo(a)pyrene, benzo(b)flouranthene, and chrysene – altogether, and that this allows to draw conclusions about the overall contamination with PAH. The contamination levels actually found in butter mackerel and vegetable margarine do not give rise to further studies beyond routine surveillance.

Nitro musk compounds

The nitro musk compounds musk ketone and musk xylene were looked for and not found in goat cheese, butter, full egg, butter mackerel and vegetable margarine. This confirmed the finding of earlier monitoring tests in other foodstuffs that contamination of foods with nitro musk compounds has decreased.

Mycotoxins

Ochratoxin A

Contamination of wheat grains as well as bulgur and spelt wheat flakes with the mycotoxin ochratoxin A (OTA) is at a low level. The small number of analyses in wheat grains in 2009 makes it necessary to make more tests to confirm this assessment of the contamination situation. We think it necessary to periodically test bulgur/hard wheat and spelt wheat flakes/spelt wheat, as well as other kinds of cereals such as buckwheat, and cereal products for OTA. The 2011-2015 Monitoring Plan therefore provides for a large variety of foodstuffs to be tested for OTA.

Soluble coffee was found to contain OTA in the majority of samples, yet the concentrations were mostly far below the currently valid maximum level. There was no concentration above this level. Obviously, manufacturers’ self control and prevention programmes are effective. Soluble coffee does not contribute noticeably to consumer’s exposure to ochratoxin A, as it is.

Deoxynivalenol

Wheat grains, bulgur, and spelt wheat flakes carried only slight contaminations with the mycotoxin deoxynivalenol (DON). The small number of analyses in wheat grains makes it necessary to make more tests to confirm this assessment of the contamination situation. Also, we think it necessary to periodically test bulgur/hard wheat and spelt wheat flakes/spelt wheat for DON contents.
Wheat cookies and other small wheat baking ware showed relatively low contamination with DON. So, future testing seems to make sense only when adverse weather let increased attack with mould fungi be suspected.

Zearalenone

Contamination of wheat grains and the cereal products bulgur and spelt wheat flakes with the mycotoxin zearalenone (ZEA) was at a low level. The small number of analyses in wheat grains in 2009 makes it necessary to make more tests to confirm this assessment of the contamination situation. Analyses of bulgur/hard wheat and spelt wheat flakes/spelt wheat for ZEA should be occasionally repeated.

Fumonisins

Nearly half of the foodstuffs analysed (maize flour, polenta, maize husk meal, grits, cornflakes, crunchy products on maize basis, and foods for gluten-free nutrition) were contaminated with fumonisins. Concentrations were yet very low on average, and the number of samples with concentrations actually higher than half the EU maximum level, or even higher than the maximum level, was extremely small. The findings of the 2009 monitoring tests thus confirmed the findings of earlier years.
Surveillance of the situation of contamination should continue still, because it cannot be precluded that fumonisin levels will rise again.

Ergot alkaloids

Testing of Type 997 and Type 1150 rye flour, rye wholemeal flour and rye bruised grain for 12 of the 30 known ergot alkaloids showed that the sums of single alkaloids (total ergot alkaloid content) were always below the orientation value of 1000 µg/kg.

Nitrate

Nitrate was found to measurable degree in nearly all samples of rucola and celeriac. As it was to be expected, nitrate levels in rucola harvested in the winter season were markedly higher than levels in summer. Rucola, together with spinach and fresh head lettuce, is among the foods with the highest nitrate levels analysed in the framework of this monitoring. The situation looks better with celeriac, which showed clearly decreasing levels compared to previous years. The high nitrate findings in rucola underline the need for an EU-wide maximum level.

Elements

Lead

Analyses in the chosen foodstuffs of animal origin essentially confirmed findings of previous years. Among the foodstuffs of vegetable origin, rucola, celeriac and bananas showed a clear decline in lead contents. Wheat grains, sweet peppers, and orange juice confirmed the trends of previous years. Lead contents in vegetable margarine, bulgur and spelt wheat flakes, which were analysed for lead for the first time in the framework of the monitoring programme, were not conspicuous against other comparable foods.

Cadmium

Tests for cadmium carried out under the 2009 monitoring programme essentially confirmed the results of previous years.
Cadmium contents in vegetable margarine and bulgur, which were analysed for cadmium for the first time in the framework of the monitoring programme, were not conspicuous against other comparable foods. In spelt wheat flakes, which were also tested for the first time, cadmium was found at comparable frequency and levels as in wheat grains. Orange juice and rucola showed slightly declining levels. Celeriac as a root vegetable contained higher cadmium levels than the other foodstuffs tested. This food also had a relatively high number of non-compliant samples exceeding the maximum level (10 samples from Germany, one sample from the Netherlands). This must be attributed to the relatively high contamination of soils with cadmium. Information from other EU Member States is that contamination levels of celeriac are roughly the same as in Germany.
Against the background of a new calculation of toxicologically tolerable exposure on EU and WHO level, experts from Member States and the EU Commission are currently discussing a revision of cadmium maximum levels. The purpose is to protect susceptible consumer groups (vegetarians, pregnant or nursing women, and infants) from potential health risks. The discussions have not come to an end yet.

Mercury

As it was already found in the 2001 monitoring, butter mackerel carried quantifiable contents of total mercury in all samples. Yet the concentrations have declined from the last study, as have mercury concentrations in goat cheese. The same holds for the noncompliance rate. In goat cheese, there was no non-compliant sample in 2009.
Bulgur and spelt wheat flakes, which were analysed for mercury for the first time in the 2009 monitoring, did not show any conspicuous findings against comparable other foodstuffs. Full egg and wheat grains had a higher percentage of samples with quantifiable residues compared to previous years, but levels measured were actually very low and therefore did not give rise to consider studies beyond routine surveillance.

Copper

The proportion of samples with quantifiable copper contents in foodstuffs tested in 2009 was roughly the same as in tests in previous years. Rucola showed a clear decline in copper contents compared to 2004, while orange juice showed a slight increase. Among the foodstuffs analysed for copper for the first time, vegetable margarine did not contain any quantifiable contents. Spelt wheat flakes contained slightly more copper than wheat grains.
The legal maximum level was exceeded in one sample of goat cheese from France, one of sweet peppers from Spain, and one of celeriac from Germany. The levels actually measured did not give rise to consider studies beyond routine surveillance.

Aluminium

Aluminium was found in quantifiable amounts in 32 – 98% of samples in the foods tested therefor. Apart from rucola, which had the highest aluminium finding with 820 mg/kg, all other foodstuffs had median contents below 5 mg/kg. In some kinds of vegetables, aluminium can be present in higher concentrations, as aluminium findings in spinach in the 2008 food monitoring and data assessments by the European Food Safety Authority (EFSA) have shown.

Arsenic

The proportion of samples with quantifiable total arsenic contents in foodstuffs tested therefor ranged from 1.6% in bananas to 100% in butter mackerel. Sweet peppers, goat cheese and orange juice did not contain quantifiable amounts, different from previous years. Median values of arsenic levels ranged from 0.01 mg/kg (full egg, wheat grains, rucola, celeriac), and 0.77 mg/kg (butter mackerel). The levels measured did not give rise to consider studies beyond routine surveillance.

Nickel

All the foodstuffs apart from wheat grains were analysed for nickel the first time in the 2009 monitoring. There were no conspicuous findings compared to other foods, apart from findings in vegetable margarine. Vegetable margarine had a maximum finding of 16.7 mg/kg, which is relatively high. In that particular case, the finding might be attributable to the use of nickel-containing catalysts for fat hardening, residues of which were not thoroughly removed to the technically unavoidable degree.

Acrylamide

A comparison with monitoring findings of the year 2004 shows that acrylamide levels in deep-fried potato chips (French fries) could not be reduced. This trend of stagnation has continued for several years now. While medium levels are comparable to those reported from other EU Member States in 2008, 90th percentile values were noticeably higher in Germany than EU wide. The levels found should be reason to further monitor acrylamide levels in deep-fried chips.

Furan

Monitoring analyses in coffee and coffee extracts showed that coffee is an average adult’s most important source of exposure to furan. Although there is no acute health risk at the present state of knowledge, minimizing furan levels seems to be a requirement of preventive health protection of consumers until we have a reference value. Furan levels in roasted coffee have not changed compared to the previous year.