Opinion of the Scientific Panel on food additives, flavourings, processing aids and materials in contact with food (AFC) related to Semicarbazide in food

No abstract available

The Scientific Panel on Food Additives, Flavourings, Processing Aids and Materials in Contact with Food (AFC) has been asked to advise the European Commission on the occurrence of semicarbazide (SEM) in food. The European Food Safety Authority (EFSA) issued preliminary advice on SEM in 2003, when the occurrence of SEM in food, derived from food packaging, was first discovered. The Panel was asked on this occasion to gather data on the occurrence of SEM in all types of food, to explain the conditions under which SEM may be formed in food and to evaluate the analytical methods used. In the light of this information, the Panel was asked to assess the risks posed by semicarbazide in all types of food. The approach taken by the Panel on this question was to search the scientific literature and to take account of information from the Commission, national authorities and trade associations.

SEM has been found to occur in different types of foods and the source of SEM varies. SEM is a metabolite of the veterinary medicine nitrofurazone, but since the use of nitrofurazone is illegal in the EU, SEM from this source should not be detectable in foods. SEM can be present in foods as a result of migration from sealing gaskets used in the metal lids of jars and bottles.

In this case, the origin of the SEM is thermal breakdown of azodicarbonamide, a blowing agent used to foam the plastic gaskets. SEM has been found in food products made using flour in which azodicarbonamide has been added as a dough-improver a practice that is not permitted in the EU. Other sources have been suggested but are less well documented. SEM is reportedly formed as a reaction product of the action of hypochlorite on food additives such as carrageenan and on foods such as egg white powder. Finally, SEM may be present at background levels naturally, may be formed at low levels when some foods are dried, and may also derive from as yet unidentified sources.

The method of analysis used to test foods for SEM involves acid hydrolysis and a derivatisation step with 2-nitrobenzaldehyde. The derivative is then determined using liquid chromatography coupled to tandem mass spectrometry with a detection limit in the region of 0.2 μg/kg. The acid hydrolysis step liberates bound residues for analysis and the analysis method therefore measures total SEM (“free” and “bound”) in the sample. The acid hydrolysis conditions used in the analytical method are not dissimilar to normal gastric conditions. Since the current state of knowledge on the bioavailability of any bound residues is incomplete, in this evaluation there is no distinction made between SEM that is present as such in a food sample and any SEM that may have been formed from precursors in the food under acidic conditions used in the analysis.

It is concluded that the method of testing for SEM provides concentration data that are suitable for this risk evaluation.

On the basis of the information available, migration of SEM from the breakdown of azodicarbonamide (ADC) in sealing gaskets is by far the largest source of exposure known. The concentration data available from analyses of food undertaken by different countries were similar. The highest potential intakes of SEM are in infants consuming ready-to-feed infant milk and baby food, attributable to the larger gasket areas involved in the packaging and their small body weight. Reasonable worst case estimates of intake for infants fed on products packaged in glass jars and bottles range from 0.35 to 1.4 μg/kg bw/day.

Adult exposures to SEM from this source are likely to be much lower than infant exposures, due to the lower contribution of foods packaged in bottles and jars to the total diet of adults, the lower contamination levels derived from the smaller gasket areas involved for that packaging, and the higher adult body weight. A reasonable worst case estimate of intake for an adult would be 0.02 μg/kg bw/day.

Commission Directive 2004/1/EC prohibits the use of azodicarbonamide in food contact materials from 2nd August 2005. Once existing stocks of packaged foods are used up, exposure of consumers by this route will have been eliminated.

Other possible sources of SEM in foods contribute far less to exposure than that estimated above for packaging. Bread made using flour treated with ADC can contain SEM. In laboratory tests the SEM concentration in bread was 28 μg/kg. ADC is not permitted as a flour treatment agent in the EU and the importation of bread and bakery ware is likely to be very low. There is the potential for exposure from breaded animal products imported into the EU (e.g. frozen breaded chicken or fish products). Taking an upper figure of 5 μg/kg of product this would give an intake of SEM of 1 μg/person from a consumption of 200g of product.

For a high consumer of egg products that may be contaminated by 50 μg/kg SEM as a result of using hypochlorite as a sanitising solution on production equipment, a reasonable worst-case estimate of exposure is 0.008 μg/kg bw/day. For the food additive carrageenan, that may become contaminated with SEM at a mean concentration of 65 μg/kg from use of hypochlorite in the production process, if consumption was up to the Acceptable Daily Intake (ADI) for carrageenan then the intake of SEM from this source could be up to 0.005 μg/kg bw/day.

SEM has been shown to be carcinogenic in mice, but not rats. Literature data on genotoxicity together with the results of recent studies indicate that SEM is mutagenic but not clastogenic in some test systems in vitro, notably in the absence of an exogenous metabolising system. In vivo, negative results were reported in studies on DNA damage in liver and lung of mice, and in the micronucleus assay in the mouse. Based on the overall weight of evidence provided by the studies performed, which included a study using a highly sensitive methodology, the Panel concluded that the weak genotoxicity exerted by SEM in vitro is not expressed in vivo.

The new data allaying the concern on genotoxicity in vivo, and the likely reductions in exposure following replacement of the most significant, currently known source of SEM in the diet (gaskets on glass jars and bottles), offer further support to the preliminary advice given by EFSA in 2003 that the risk, if any, from consumption of foods containing SEM is judged to be very small, not only for adult consumers but also for infants. In this respect the Panel noted that SEM is a weak non-genotoxic carcinogen for which a threshold mechanism can be assumed. A large margin of at least 5 orders of magnitude exists between the dose causing tumours in experimental animals and human exposure, including that of infants.

The Panel therefore concluded that the issue of carcinogenicity is not of concern for human health at the concentrations of SEM encountered in food.

Semicarbazide, CAS No 57-56-7, SEM, azodicarbonamide, CAS No 123-77-3 nitrofurazone, CAS No 59-87-0, gaskets, hypochlorite.