Technical University of Denmark Microbiological contaminants in food in the European Union in 2004-2009

Abstract

Zoonotic and non-zoonotic microbiological contaminants in foodstuffs can cause human disease. Zoonotic pertains to zoonoses, which are diseases that may be transmitted between animals and humans. Such infections can be acquired directly from animals or through ingestion of contaminated foodstuffs. Non-zoonotic contaminants are not transmissible between animals and humans. The severity of these diseases in humans can vary from mild symptoms to life-threatening conditions. During the years 2004 to 2009, EU Member States (MSs) submitted information on zoonotic microbiological contaminants (Salmonella, Campylobacter, Listeria, Yersinia, verotoxigenic Escherichia coli (VTEC), Brucella)and non-zoonotic microbiological contaminants (histamine, Enterobacter sakazakii and staphylococcal enterotoxins) in food to the zoonoses database at the European Food Safety Authority (EFSA). Norway and Switzerland also provided information. This report presents a detailed description of the sampling activity over the years, the quality of data reported by MSs, and an in-depth analysis of the occurrence of microbiological contaminants in domestically produced foodstuffs. From 2004-2009, MSs reported tests results from more than 5.5 million samples and non-MSs from more than 200,000 samples from different foodstuffs. Most of these investigations were assumed to reflect the occurrence of microbiological contaminants in domestically produced foodstuffs, however some of the data were reported as HACCP, own-check and outbreak/clinical investigations or described as import, selective or suspect sampling. Due to heterogeneity in the sampling schemes for such data, they are difficult to interpret with regard to the overall occurrence of microbiological contaminants in domestically produced foodstuffs and have therefore not been included in the analyses. However, the data can be of great value in relation to outbreak investigations and as input data for risk assessments. Specific information on sampling stage and sampling unit is essential if MS data are to be comparable, and data without such information have therefore also been excluded from the detailed analyses of the microbiological contaminants in Chapter 3. Initially, 86% of the reported samples were assumed to be representative of the occurrence of microbiological contaminants in domestically produced food. However, when samples without specific sampling unit and/or sampling stage information were excluded, only 44% of all data reported for 2004 to 2009 could be considered for in-depth analyses. Furthermore, some investigations were reported without specific indication of e.g. animal species, type of vegetable or type of cheese. Due to the difficulty concerning the interpretation of these data, they were generally excluded from the analyses and discussions. Fortunately, since 2007 the quality of reported data has improved greatly. The total number of samples reported by MSs was large, however when data were broken down to the fine level of categorization required in this report, data were often too sporadic for calculation of EU totals. The available data did not meet the criteria required for statistical analysis of temporal trends and therefore conclusions regarding temporal trends had to be made based on the interpretation of descriptive tables and figures alone. For some foodstuffs, the occurrence of microbiological contaminants increased along the food production chain, which indicates either cross-contamination during production or microbiological growth during shelf-life. For other foodstuffs, the occurrence of microbiological contaminants decreased along the food production chain. This may be explained by reducing steps such as heat treatment, and may be illustrated by the example that RTE food in most cases was less frequently contaminated with Salmonella than non-RTE food. Despite the large amount of data reported on Salmonella, very few clear temporal trends could be demonstrated. In 2008 and 2009, the proportion of Salmonella positive samples in meat tested at retail seemed to stabilize at a relatively low level of approximately 1% compared to previous years where the proportions were above 1.3%. This might be a consequence of the microbiological criteria that came into force in 2006. However, since meat from poultry, pigs, bovine animals and eggs is widely consumed throughout the EU, the continued presence of Salmonella in these foodstuffs is likely to contribute considerably to Salmonella exposure. The lowest level of compliance to the Salmonella food safety criteria were reported for minced meat and meat preparations made from poultry meat intended to be eaten cooked. During the period 2006 to 2009, number of investigations in compliance increased for minced meat, meat preparations and meat products from poultry meat intended to be eaten cooked (batched based data). No obvious trends were observed for the other food categories during this period. For Campylobacter, poultry meat was the food vehicle with the highest proportions of positive samples compared to pig and bovine meat for all years. Generally, less data were reported on RTE products than on non-RTE products. In RTE products, Campylobacter was only reported in a few cases and at low proportions. A substantial amount of data was reported on Listeria monocytogenes by MSs, especially from retail level. The majority of the samples were collected from different types of RTE meat products. L. monocytogenes was found more frequently in bovine meat products compared to pig meat products and poultry meat products. Fishery products, soft cheeses and meat products were the food categories most frequently found to be in non-compliance with the L. monocytogenes food safety criteria. Relative high levels of non-compliance were also found among samples reported as ‘Other RTE products’ tested at processing. From 2006 to 2009, a decrease in number of investigations in non-compliance was observed in fish and fishery products tested both at processing and retail, as well as for soft cheeses tested at processing Data on Yersinia weremost often reported in pig meat. Very little data were reported specifically on Yersinia enterocolitica and most of these originated from non-RTE pig and bovine meat sampled at retail. Unfortunately, information regarding the presence of human pathogenic biotypes was available from only a few investigations. VTEC was reported from several different types of foods, but generally the occurrence of VTEC, and particularly VTEC O157 was low. Information on serogroup was sparse, but five (O26, O91, O103, O111 and O157) of the six serogroups considered the most important regarding pathogenicity to humans, were reported. Data on VTEC in fruit and vegetables were very limited, but several investigations indicated that vegetables were a source of human VTEC O157 infections. Data on the three non-zoonotic microbiological contaminants histamine, Enterobacter sakazakii and staphylococcus enterotoxins were analysed for the first time for the years 2006-2009. The Regulation (EC) No 2073/2005 on microbiological criteria for foodstuffs came into force in 2006 and this was the first year MSs reported data on these contaminants. However, little data have been reported and the data clearly reflect the foodstuffs included in the microbiological criteria. Due to the small amount of available data, all data, except for HACCP related data were included for analyses. Presence of histamine was reported in fish and fishery products and few MSs reported high levels, especially at retail. Some of these results were not in compliance with the microbiological criteria. For Enterobacter sakazakii, the majority of data were reported on dried infant formulae and dried dietary foods for special medical purpose intended for infants below six months of age. Several MSs reported positive findings in these products which are in non-compliance with the microbiological criteria. Positive findings for staphylococcus enterotoxins were reported for samples of milk, cheese and other dairy product. Positive findings in cheeses, milk powder and whey powder are in non-compliance with the microbiological criteria. A full evaluation of the microbiological criteria for the three non-zoonotic microbiological contaminants was not possible due to missing information. For most microbiological contaminant-food stuff combinations, very little consistency was observed concerning reporting from individual MSs during the period. Therefore, statistical trends could not be evaluated at EU or MS group level. There were no clear changes in the proportion of positive samples at different sampling stages along the food chain. However, it should be emphasized that the vast majority of investigations on RTE products for all six years revealed no or very low levels of microbiological contaminants.