Overview of Food Safety Hazards in the European Dairy Supply Chain E. D. van Asselt, H.J. van der Fels-Klerx, H.J.P. Marvin, H. van Bokhorst-van de Veen, and M. Nierop Groot Abstract: Monitoring of dairy products should preferably focus on the most relevant food safety hazards in the dairy supply chain. For this purpose, the possible presence of microbiological, chemical, and physical hazards as well as trends in the dairy supply chain that may affect their presence were assessed. A literature review was combined with available data from EFSA, RASFF, and the Dutch monitoring program on chemical hazards as well as expert information. This study revealed that microbiological hazards are encountered more frequently in dairy products than chemical and physical hazards. Listeria monocytogenes, Staphylococcus aureus, Salmonella, and human pathogenic Escherichia coli were identified as the most important microbiological hazards in dairy products. Soft and semisoft cheeses are most frequently associated with L. monocytogenes and S. aureus enterotoxins, whereas raw milk is most frequently associated with human pathogenic E. coli and Campylobacter spp., Cronobacter spp., and Salmonella spp. are the microbiological hazards of most concern in powdered infant formula. Based on literature, monitoring, and RASFF data, the most relevant chemical hazards in dairy products are aflatoxin M 1 , dioxins, and dioxin-like compounds and residues of veterinary drugs. Chemical hazards primarily occur at the dairy farm and may accumulate during further processing. The most relevant physical hazards are metal, glass, and plastic particles introduced during processing. Analysis of trends in the near future revealed that increased milk production is seen as most relevant in relation to food safety. Other trends affecting food safety are climate change and changes at the farm level, which aim to improve animal welfare and environmental sustainability. Keywords: chemical hazards, microbiological hazards, physical hazards, (raw) dairy products Introduction Milk is widely produced in Europe totaling 165 million metric tons in the 28 European Union (EU) member states for 2014 (Eurostat 2015) and contributing to about 25% of the global milk production (FAO Stat 2015). The EU is a major player in the world dairy market due to its role as lead exporter of many dairy products (Tropea 2015). Germany and France are the main producers within the EU accounting for almost 40% of total EU production. Other major producers are the United Kingdom (10%), the Netherlands (8%), Poland, and Italy (both 7%). Most milk is used to produce cheese and butter, 36% and 29%, respectively. The remaining prod- ucts include cream for consumption (13%), milk for consumption (11%), milk powder (3%), and other products (8%). Most milk (92%) is delivered to dairy companies for further processing; the remainder is processed at the farm into milk for consumption and cheese. Almost all the milk is produced by cows (96.8%) and a small proportion by ewes, goats, and buffalo (3.2%) (Eurostat 2015). Milk composition varies among animal species. For example, CRF3-2016-1098 Submitted 7/11/2016, Accepted 10/31/2016. Authors van Asselt, van der Fels-Klerx, and Marvin are with RIKILT–Wageningen Univ. & Research, P.O. Box 230, 6700 AE Wageningen, the Netherlands Authors van Bokhorst-van de Veen and Nierop Groot are with Wageningen Food & Biobased Research, P.O. Box 17, 6700 AA Wageningen, the Netherlands. Direct inquiries to authors van Asselt (E-mail: [email protected]). sheep milk has a substantially higher fat content, ranging from 60 to 82 g/kg compared to goat and cow milk, which range from 30 to 50 g/kg and from 35 to 40 g/kg, respectively (ter Mors and de Wit 2011). Dairy production follows several stages and generally starts with animal feed production, followed by raw milk production at the farm, and further processing either at a dairy company or at the farm itself. The majority of dairy products within the EU are sold at retail (Figure 1). Along this dairy supply chain, food safety haz- ards may enter at various stages. Intervention measures are taken, such as the implementation of GLOBAL Good Agricultural Prac- tices (GLOBALG.A.P.) or Hazard Analysis Critical Control Points (HACCP) systems, in order to control the presence of these haz- ards. Despite these quality control programs, food safety hazards may still be present and, therefore, monitoring programs have been established to detect the possible presence of food safety hazards (Noordhuizen and Metz 2005). These programs should be risk-based and focus on the most relevant food safety hazards (Regulation (EC) 882/2004), which will enhance the probability of detection (Van Asselt and others 2012). For this purpose, an inventory of potential food safety hazards alongside an identifica- tion of the most important hazards to be included in monitoring programs is required. The current situation for the dairy chain should be re-evaluated, and any relevant trends that may influence hazard presence should also be considered. C 2016 Institute of Food Technologists ® doi: 10.1111/1541-4337.12245 Vol. 16, 2017 Comprehensive Reviews in Food Science and Food Safety 59
Transcript
Overview of Food Safety Hazards in the EuropeanDairy Supply ChainE. D. van Asselt, H.J. van der Fels-Klerx, H.J.P. Marvin, H. van Bokhorst-van de Veen, and M. Nierop Groot
Abstract: Monitoring of dairy products should preferably focus on the most relevant food safety hazards in the dairysupply chain. For this purpose, the possible presence of microbiological, chemical, and physical hazards as well as trendsin the dairy supply chain that may affect their presence were assessed. A literature review was combined with availabledata from EFSA, RASFF, and the Dutch monitoring program on chemical hazards as well as expert information. Thisstudy revealed that microbiological hazards are encountered more frequently in dairy products than chemical and physicalhazards. Listeria monocytogenes, Staphylococcus aureus, Salmonella, and human pathogenic Escherichia coli were identified as themost important microbiological hazards in dairy products. Soft and semisoft cheeses are most frequently associated withL. monocytogenes and S. aureus enterotoxins, whereas raw milk is most frequently associated with human pathogenic E. coliand Campylobacter spp., Cronobacter spp., and Salmonella spp. are the microbiological hazards of most concern in powderedinfant formula. Based on literature, monitoring, and RASFF data, the most relevant chemical hazards in dairy productsare aflatoxin M1, dioxins, and dioxin-like compounds and residues of veterinary drugs. Chemical hazards primarily occurat the dairy farm and may accumulate during further processing. The most relevant physical hazards are metal, glass, andplastic particles introduced during processing. Analysis of trends in the near future revealed that increased milk productionis seen as most relevant in relation to food safety. Other trends affecting food safety are climate change and changes at thefarm level, which aim to improve animal welfare and environmental sustainability.
Keywords: chemical hazards, microbiological hazards, physical hazards, (raw) dairy products
IntroductionMilk is widely produced in Europe totaling 165 million metric
tons in the 28 European Union (EU) member states for 2014(Eurostat 2015) and contributing to about 25% of the global milkproduction (FAO Stat 2015). The EU is a major player in the worlddairy market due to its role as lead exporter of many dairy products(Tropea 2015). Germany and France are the main producers withinthe EU accounting for almost 40% of total EU production. Othermajor producers are the United Kingdom (10%), the Netherlands(8%), Poland, and Italy (both 7%). Most milk is used to producecheese and butter, 36% and 29%, respectively. The remaining prod-ucts include cream for consumption (13%), milk for consumption(11%), milk powder (3%), and other products (8%). Most milk(92%) is delivered to dairy companies for further processing; theremainder is processed at the farm into milk for consumption andcheese. Almost all the milk is produced by cows (96.8%) and a smallproportion by ewes, goats, and buffalo (3.2%) (Eurostat 2015).Milk composition varies among animal species. For example,
CRF3-2016-1098 Submitted 7/11/2016, Accepted 10/31/2016. Authors vanAsselt, van der Fels-Klerx, and Marvin are with RIKILT–Wageningen Univ. &Research, P.O. Box 230, 6700 AE Wageningen, the Netherlands Authors vanBokhorst-van de Veen and Nierop Groot are with Wageningen Food & BiobasedResearch, P.O. Box 17, 6700 AA Wageningen, the Netherlands. Direct inquiries toauthors van Asselt (E-mail:[email protected]).
sheep milk has a substantially higher fat content, ranging from 60to 82 g/kg compared to goat and cow milk, which range from 30to 50 g/kg and from 35 to 40 g/kg, respectively (ter Mors and deWit 2011).
Dairy production follows several stages and generally starts withanimal feed production, followed by raw milk production at thefarm, and further processing either at a dairy company or at thefarm itself. The majority of dairy products within the EU are soldat retail (Figure 1). Along this dairy supply chain, food safety haz-ards may enter at various stages. Intervention measures are taken,such as the implementation of GLOBAL Good Agricultural Prac-tices (GLOBALG.A.P.) or Hazard Analysis Critical Control Points(HACCP) systems, in order to control the presence of these haz-ards. Despite these quality control programs, food safety hazardsmay still be present and, therefore, monitoring programs havebeen established to detect the possible presence of food safetyhazards (Noordhuizen and Metz 2005). These programs shouldbe risk-based and focus on the most relevant food safety hazards(Regulation (EC) 882/2004), which will enhance the probabilityof detection (Van Asselt and others 2012). For this purpose, aninventory of potential food safety hazards alongside an identifica-tion of the most important hazards to be included in monitoringprograms is required. The current situation for the dairy chainshould be re-evaluated, and any relevant trends that may influencehazard presence should also be considered.
doi: 10.1111/1541-4337.12245 Vol. 16, 2017 � Comprehensive Reviews in Food Science and Food Safety 59
Food safety hazards in the dairy chain . . .
Figure 1–Various stages in the dairy production chain from farm-to-fork.
Correspondingly, this paper presents an overview of the mostrelevant microbiological, chemical, and physical hazards that maybe present in the European dairy supply chain. Moreover, futuretrends that may affect these hazards are discussed
Methods UtilizedDemarcation of the study
The Dutch dairy supply chain (Figure 1) was used as a basis forthe European dairy supply chain (Figure 1). Even though produc-tion chains can differ throughout Europe, it was assumed that thefood safety hazards and their points of entry are identical. Severalconsumer products can be produced from raw milk, yet this studyfocused on milk produced for consumption, cheese, butter, andmilk powder produced and imported into the EU derived fromcows, goats, and sheep. Food safety hazards were evaluated from thefarm to the final product and excluded the retail and the consumerstages. The study focused on dairy products, yet did not includepossible hazards introduced from major additional ingredients,such as fruits, nuts, or herbs, nor minor ones, such as salt, vitamins,flavor, or enzymes, which can be encountered during processing.
Data analysisThe Rapid Alert System for Food and Feed (RASFF) portal
was used to extract data for notifications of food safety hazardsin milk and milk products within the EU during 2009 to 2014.All notifications were included, namely, border rejections, publicinformation, and alerts. Furthermore, data on chemical hazardsfrom the Dutch monitoring program on dairy products wereretrieved from 2009 to 2013. These data are from the Dutch
Quality Program for Agricultural Products (KAP) that involvesextensive cooperation between the Dutch government and Dutchagribusinesses. Data originated from the Dutch Dairy Organiza-tion (NZO) and the Netherlands Food and Consumer ProductSafety Authority (NVWA) for the years 2009 and 2010 and fromRIKILT for the years 2009 to 2013. Furthermore, European mon-itoring data from the World Health Organization’s (WHO) GlobalEnvironment Monitoring System–Food Contamination Monitor-ing and Assessment Program (GEMS/Food contaminants) wereobtained for 2009 to 2014 and included about 33000 sample re-sults from 26 countries. Outbreak data on microbiological hazardswere obtained from EFSA reports from 2010 to 2013. Addition-ally, results as reported in the European Food Safety Authority(EFSA) report on residues in animal products were used for 2012to 2014.
Literature studyA literature search was performed to identify the possible micro-
biological, chemical, and physical hazards in the dairy chain usingthe Scopus database and Google Scholar. The relevance of the re-trieved references was 1st determined based on title and keywordsof the reference. Then, these selected references were further eval-uated based on the abstracts. Criteria for selection were: relevanceto the study, geographic origin, and impact factor of the journal.Based on this selection, full text papers were downloaded and ana-lyzed. Additionally, snowball citation was applied and, if necessary,additional papers were examined on specific topics. Furthermore,scientific reports were retrieved from governmental organizations
such as the Food and Agriculture Organization (FAO), EFSA, andWHO.
Expert studyIn order to obtain more information on future trends, 11 persons
with expertise on different stages of the dairy chain were inter-viewed. They involved experts on dairy farming (cow and goatmilk), dairy processing, and/or the commercial dairy trade. Bothexperts from industry and experts with a scientific backgroundwere approached. Experts from RIKILT, Wageningen Univ., Cen-tral Veterinary Inst., the Dutch Dairy Organization (NZO), theDutch Dairy Goat Organization (NGZO), experts working in pri-mary production, and experts from the dairy industry participatedin the study. The experts were interviewed using a predefinedquestionnaire with questions on foreseen changes within the an-imal feed sector, at the dairy farm and in dairy processing. Theywere asked about different developments for dairy cow, goats, andsheep farmers as well as expected changes in production volumes,the dairy supply chain, and consumer trends that may affect dairyproduction.
Results and DiscussionResults of the monitoring data
RASFF notifications. In total, 243 notifications were retrievedfrom the RASFF database on food safety hazards in dairy productsfor 2009 to 2014. Microbiological contaminations were mainlyreported (203 notifications). Of these microbiological contamina-tion notifications, 24% were related to spoilage microorganisms,while the remaining 76% were related to the following pathogenicmicroorganisms: Listeria monocytogenes (52%), Escherichia coli (11%),Salmonella (10%), Pseudomonas (3%), and Bacillus spp. (2%). Fraudwas reported in 15 cases (6% of the total number of notifications).These cases related to unauthorized operators and illegal importof dairy products. Physical and chemical hazards were reported in10% of the total number of notifications (Figure 2). Physical haz-ards encountered were metal, glass, or plastic fragments. Chemicalhazards included aflatoxins, veterinary drug residues, and disinfec-tants. The latter concerned 3 notifications with exceeding levelsof hydrogen peroxide in dairy products originating from France,Germany, and the Czech Republic.
As indicated in Figure 2, most notifications involved cheeseproducts; these notifications were primarily for pathogenic mi-croorganisms (139 notifications), of which about 45% concernedsoft cheeses. Half of the notifications for milk concerned microbi-ological hazards, whereas 18% concerned aflatoxins, 8% veterinarydrugs (beta-lactams and chloramphenicol), and another 8% phys-ical hazards (glass and plastic fragments).
Outbreak data. The EFSA publishes annual reports on thenumber of reported foodborne outbreaks caused by zoonoses inthe EU (EFSA 2012a, 2013b, 2014b, 2015a). In 2013, 839 strong-evidence foodborne outbreaks were reported within the EU(Table 1). Of these 839 outbreaks, cheese and milk were mostfrequently encountered for dairy products (each 1.3%).
Cheese can be a vehicle for Salmonella, which is the causativeagent most frequently reported for foodborne outbreaks in general.In the period 2010 to 2013, cheese was reported in �1.1% ofthe total number of Salmonella cases, while other dairy productsaccounted for 0 to 0.6% (milk, including raw milk) and 0.6% to2.1% (dairy products other than milk and cheese) of the cases. Oneexception occurred in 2012, in which a relatively high numberof Salmonella outbreaks were reported for cheese coming fromFrance.
The total number of foodborne Listeria outbreaks is relativelylow (between 3 and 7 outbreaks per year). Typically, one ofthese outbreaks is caused by the consumption of cheese. Milkconsumption accounts for a relatively high number of Campy-lobacter outbreaks: in 2013, 32 strong-evidence Campylobacter spp.outbreaks were reported in the EU, of which between 9% (2013)and up to 20% (2012) could be attributed to milk. Although notspecified, it is conceivable that these high numbers result from rawmilk consumption.
Raw milk or processed raw milk also poses a risk for Shigatoxin-producing E. coli (STEC) and a number of outbreaks werereported for this pathogen (Table 1). In 2013, 3% of 860 tested rawmilk samples were found positive for STEC, while other nonrawdairy products were positive for STEC in 5% of the samples.
Staphylococcus enterotoxins in cheese were involved in 6.4%(2013) up to 20% (2012) of the outbreaks for all food categories.
Monitoring data on chemical compounds in dairy productsData from the Dutch monitoring program contained over 2000
samples for chemical compounds in butter, cheese, milk powder,and milk from 2009 to 2013. Most samples were taken from milk(around 70%), the majority of which were tested for the presenceof veterinary drugs. Of the 89 samples tested for aflatoxin M1
(AFM1), 2 samples were positive containing 0.032 and 0.1 µg/kg.Dioxins and dioxin-like polychlorinated biphenyls (dl-PCBs) werepresent in various milk samples, but levels were below the actionand maximum limits as set by Commission Recommendation2011/516/EU and Regulation (EC) 1881/2006. Samples testedfor heavy metals, polycyclic aromatic hydrocarbons (PAH), andpesticides contained levels below the limit of detection (LOD), andall samples tested for veterinary drugs were below the maximumresidue levels (MRLs).
In 2014, EFSA gathered the results of about 30000 milk samplestested on possible residues in milk from EU member states: about7000 samples were tested on banned compounds, primarily onchloramphenicol, about 25000 on regulated veterinary drugs andabout 5000 on other compounds (EFSA 2016). Overall, 0.12%of the 30000 samples were noncompliant. Most of these non-compliant samples regarded regulated veterinary drugs (group B1
and B2 in Directive 96/23/EC), while 2 samples contained chlo-ramphenicol (group A6). In total, 20 samples contained antibioticresidues (group B1), 4 samples contained anthelminthics (groupB2a), 1 sample contained nonsteroidal anti-inflammatory drugs(NSAID; group B2e), 7 samples contained AFM1 (group B3d),and 1 sample contained PCBs (group B3a) above the legal limits(EFSA 2016). Reports from previous years show comparable re-sults (EFSA 2014c, 2015b). The WHO GEMS database does notcontain results on veterinary drug residues. It shows that the mostcommonly reported compounds in dairy products are dioxins andPCBs and AFM1. In some cases heavy metals (primarily lead) arereported above the legal limit.
Results from the literature reviewInitially, 380 publications were retrieved using the specified
search criteria for identification of potential microbiological,chemical, and physical hazards in the dairy chain. Eleven publica-tions focused on goat and only 3 publications focused on sheep.This indicated that limited information is available in the scientificliterature about dairy products made from milk of these producinganimals.
Around 40 articles and reports were retrieved based on expertinput and Google searches resulting in reports from international
Figure 2–Number of RASFF notifications for microbiological, physical, and chemical hazards in dairy products (2009 to 2014).
organizations such as FAO, the International Dairy Federation(IDF), and the EU.
A recent study in which the various hazards in dairy productswere compared showed that microbiological hazards constitutethe highest risk (based on occurrence of the hazard, severity ofthe hazard, and possibilities of detecting the hazard) followed bychemical hazards and physical hazards (Kurt and Ozilgen 2013).The following sections describe the various food safety hazardsencountered in dairy products.
Microbiological hazards at primary production. A recent reportpublished by EFSA (2015b) lists the main microbiological hazardsrelated to consumption of raw milk as identified by a panel of Euro-pean experts: Campylobacter spp., Brucella melitensis, Mycobacteriumbovis, Salmonella spp., STEC, and tick-borne encephalitis virus(TBEV) (Table 2). Several studies performed in both Europeanand non-European countries detected foodborne pathogens inraw cow milk stored in farm bulk tanks and dairy silos, includ-ing L. monocytogenes, Salmonella spp., S. aureus, E. coli O157, andMycobacterium avium subsp. paratuberculosis as reviewed by Koustaand others (2010). At the primary stage of the milk supply chain,there are various routes of contamination for these pathogens.First of all, feed and the drinking water of cows may be a source ofcontamination. Notably, ruminants may become infected by theingestion of feed or water contaminated with Toxoplasma gondii,the causative agent of toxoplasmosis (ACMSF 2013). Furthermore,contaminated feed or improperly fermented silage can eventuallylead to teat surface contamination. After feed digestion, the sur-viving (spore-forming) pathogens can end up, via fecal secretion,in the barn bedding and attach to the udder and teats. When thisdirt is not fully removed from the teats, pathogens can end up inthe milk (Vissers and others 2007a,b). Several studies identifiedsilage as the main source of butyric acid bacteria and clostridia
spores in milk for cheese production (Vissers and others 2007b;Julien and others 2008; Doyle and others 2015).
Second, the dairy farm environment is an important reser-voir of foodborne pathogens such as Salmonella species, Liste-ria monocytogenes, STEC, Campylobacter jejuni, Yersinia enterocolit-ica, and Clostridium spp. (Rohrbach and others 1992; Jayarao andHenning 2001; van Kessel and others 2004; Lindstrom and oth-ers 2010). Milk can become contaminated with pathogens beforeit leaves the teat due to the presence of dirt and fecal materialcontaining pathogens on the teat skin and thereby also (partly) inthe teat canal. Although good dairy farming practices acknowl-edges proper cleaning of teats, this measure only partly eliminatesmicroorganisms present on the teat surface (Vissers and others2007b). Besides presence on the external teat surfaces, the mam-mary gland can be colonized by microorganisms including Staphy-lococcus spp., Streptococcus, Bacillus spp., Micrococcus, and Corynebac-terium and, incidentally, coliforms, without causing any diseasesymptoms in cows (White and others 1989; LeJeune and Rajala-Schultz 2009). Cattle may also be a reservoir of Coxiella burnetii(the causative agent of Q-fever), Mycobacterium spp., Campylobac-ter sp., coliforms (including E. coli and Salmonella enterica), or thefoot-and-mouth disease virus (Hutchinson and others 1985; Spierand others 1991; Gudmundson and Chirino-Trejo 1993; Jensenand others 1996; To and others 1998; Lira and others 2004).
Inflammation of the cow’s udder tissue (mastitis) is a well-known problem at dairy farms. A large variety of bacteria maycause mastitis of which S. aureus, some coliforms, and Brucella (forsome EU countries) are human pathogens frequently encoun-tered in infected animals, which can be transmitted to humans viathe milk (Kousta and others 2010; Motarjemi and others 2014).Several factors increase the risk for mastitis, including poor hy-giene of the milking equipment, barn type, drinking water quality
Corynebacterium spp.Listeria monocytogenesMycobacterium bovisSalmonella spp.Staphylococcus aureusStreptococcus equi subsp. zooepidemicusShiga toxin-producing E. coli (STEC)Yersinia enterocoliticaYersinia pseudotuberculosisCryptosporidium parvumToxoplasma gondiiTick-borne encephalitis virus (TBEV)aCampylobacter was identified as the leading cause of outbreaks.The main microbiological hazards as identified by experts are indicated in bold font type.
(Sampimon and others 2009), housing management (EFSA2009a), and postmilking teat disinfection (Barkema and others1999; Olde Riekerink and others 2007). Animal welfare and inparticular health of dairy cows is negatively affected by geneticselection for high milk yield and has been shown to be positivelycorrelated with the incidence of mastitis (EFSA 2009a).
The use of veterinary drugs to treat clinical mastitis in dairycattle is one control point for food safety management systems(Straley and others 2006). However, selective pressure, from theuse of antimicrobials, can lead to antimicrobial resistance (EFSA2014a). The emergence of, for example, methicillin-resistant S.
aureus (MRSA) and their reported occurrence in dairy cows, inparticular in cases of clinical mastitis has been described (EFSA2009b).
Another source of microbiological contamination of the cowmilk is the milking equipment. Most microorganisms, includ-ing pathogens, have the capacity to adhere to surfaces and residein surface-associated, multicellular communities called biofilms.Once embedded in biofilms, microorganisms often display in-creased resistance to antimicrobial agents because the self-producedmatrix of extracellular polymeric material acts as a protectivebarrier against the effects of detergent and disinfectant solutions(Braunig and Hall 2005; Marchand and others 2012). Biofilmscan form on the equipment, transport line, and storage tank whencleaning and disinfection programs are inadequate.
Only limited data are available in literature concerning microbi-ological hazards associated with milk produced by goats and sheep.A review of available data in the scientific literature on microbio-logical hazards of raw milk produced from goats and sheep showedthat this milk can contain these human pathogens: Salmonella spp.Campylobacter spp., STEC, L. monocytogenes, S. aureus, B. cereus,Streptococcus spp., C. burnetii, Helicobacter pylori, Mycobacterium spp.,or TBEV (Farrokh and others 2013; Verraes and others 2014).No data on frequency or occurrence of Toxoplasma gondii in goatand sheep milk were found in the scientific literature. This datagap may, in part, be the result of the relative complexity of theanalysis. However, this parasite can infect both goats and sheep andcan be transferred from the animal to the milk. As a consequence,it is considered a microbiological hazard for consumption of rawmilk, but not the most severe hazard (EFSA 2015c). According to
Verraes and others (2014), the major microbiological hazards asso-ciated with raw goat milk consumption are infection with STEC,Campylobacter, Brucella spp., and TBEV. This list largely correspondswith the most relevant EU-wide hazards reported in goat or sheepmilk in a recent EFSA scientific opinion report; Campylobacterwas considered the leading cause of outbreaks for these milk types(EFSA 2015c). The EFSA study additionally considered Salmonellaspp. as a main hazard for goat and sheep milk.
Microbiological hazards at dairy processing. Milk for consump-tion. Pasteurization processes were initially designed to inacti-vate Mycobacterium tuberculosis, which is a relatively heat-resistant,nonspore-forming human pathogen present in milk. Pasteuriza-tion standards today aim for a 6-log reduction of C. burnetii,which is the most heat-resistant milk-borne zoonotic pathogenknown. Pasteurization of milk according to standard procedure(at least 72°C for minimally 15 s) reduces the probability of veg-etative pathogens survival by a factor of 106. However, spores ofpathogens, including those of Clostridium botulinum and B. cereus,are not eliminated by pasteurization (Papademas and Bintsis 2010),but disease incidences with pasteurized milk are rare. Milk that hasbeen pasteurized correctly is, therefore, unlikely to cause disease(Claeys and others 2013). However, inadequate pasteurization orrecontamination events after pasteurization can lead to the pres-ence of Salmonella spp, L. monocytogenes, C. jejuni, Yersinia enterocol-itica, STEC, B. cereus, Mycobacterium spp., S. aureus, or C. botulinumin dairy products (Braunig and Hall 2005).
To prevent outgrowth of surviving microorganisms or incidentalrecontaminations, pasteurization and temperature control (rapidcooling, chilled storage) are critical control points for foodbornepathogens associated with milk (WHO/FAO 2004). Historicaldata support that pasteurization of milk has resulted in improvedpublic health (Farrokh and others 2013).
More intense heating regimes are required to ensure microbi-ological safety and shelf-life of ambient stable milk for consump-tion. Hazards potentially present in shelf-stable milk (ultra-hightemperature-treated) are C. botulinum and toxigenic bacilli, butcases are rare. The applied heating regime is sufficient to ensure a12-log reduction of C. botulinum (Hutchinson and others 1985).
Cheese. Cheese has been reported as a vehicle for foodborneoutbreaks with pathogens including L. monocytogenes, S. aureus,Salmonella spp., or STEC (Kousta and others 2010). Industriallyproduced cheese is typically manufactured from pasteurized milk.However, at dairy farms, cheese may be produced from raw milkand pathogens present in the raw milk (Table 1) form a poten-tial source of cheese contamination (Kousta and others 2010 andreferences therein). Soft and semisoft cheeses contain a high mois-ture content and allow growth of different pathogens. In particular,L. monocytogenes forms a risk in these types of cheeses as it can growduring refrigerated storage, and most reports in the literature focuson this pathogenic species (EFSA 2015a). The 2013 L. monocyto-genes data (EFSA 2013a) revealed that this pathogen was morefrequently detected in cheeses made from raw or mildly heatedmilk compared to pasteurized milk. Moreover, the proportionof positive samples was higher in cheeses made from cow milkcompared to that made of milk from other producing species.However, the proportion of samples with more than 100 cfu/gcheese, the food safety criterion set for ready-to-eat food products(Regulation (EC) 2073/2005), was highest for cheese made fromsheep milk (Lomonaco and others 2009). The prevalence of L.monocytogenes in hard cheeses is low (0.4% for cheese made fromraw or mildly heated milk and 0.6% for cheese made form pas-teurized milk) and the levels did not exceed 100 cfu/g for positive
samples (Dalmasso and Jordan 2014). This suggests that the hardcheese environment does not support growth of L. monocytogenes.A challenge study with L. monocytogenes in Gouda cheese indeedshowed that this pathogen cannot grow during cheese ripening(Wemmenhove and others 2013).
Another relevant pathogen in soft and semisoft cheeses is S. au-reus. It can grow, in particular, in the 1st phase of the cheese processfrom inoculation to salting (Cretenet and others 2011). IncreasedS. aureus levels can, in part, be explained by water loss duringcurd-draining depending on the ripening temperature and pH ofthe products (Cretenet and others 2011). In general, fermenta-tion processes reaching high numbers of lactic acid bacteria willreduce S. aureus numbers and, thereby, inhibit enterotoxin forma-tion. However, enterotoxins may be produced when fermentationis retarded due to starter culture failure or when S. aureus is al-ready present in high numbers in the milk (>104 to 105 cells/mL)(Anonymous 2003). S. aureus can also grow during the initial stageof manufacturing semihard and hard cheeses. If the initial popu-lation in the milk is high (above 103 cfu/mL), enterotoxins maybe produced before the pH drops to inhibitory levels (Cretenetand others 2011). Toxins produced by coagulase-positive S. aureusare heat-stable and remain active after pasteurization, whereas theproducing population has declined and may no longer be de-tectable (Longhi and others 2003). Absence of viable S. aureus cellsis, therefore, no guarantee for absence of its toxins. Thus, Euro-pean standards for coagulase-positive staphylococci in cheese relyon controlled analyses carried out during the production processat time points where numbers are expected to be high (Cretenetand others 2011).
Several studies are available reporting STEC in cheese and otherdairy products, which are primarily produced from raw milk. Sub-stantial differences in STEC occurrence in dairy products is re-ported, varying from 0% to 27% in a selection of studies involvingat least 100 samples (Farrokh and others 2013). A study performedin Italy showed that STEC O157 could not be detected in dairyproducts from pasteurized milk (Pintado and others 2005).
Butter. Microbiological hazards reported for butter includeSalmonella spp., STEC, L. monocytogenes, and S. aureus. However,there are relatively few reported cases of butter-associated food-borne illness (Roberts and others 2005).
Butter is a water-in-oil emulsion with at least 80% fat, around18% water and a maximum of 2% salt. Microorganisms will bemainly concentrated within the droplets of the aqueous phase(Roberts and others 2005; Wilbey 2005). There are 3 mainprocesses applied at industrial scale for butter production: batchchurning of cream, continuous churning of cream (also called theFritz-process), and high-fat processes (Wilbey 2005). The latter isvulnerable for contamination, because pasteurization occurs justbefore packaging, whereas for the other 2 processes, the creamis heated at the start of the butter making process. Wash waterto remove buttermilk can be a source of pathogens if it is not ofpotable quality. Preferably, the wash water should be pasteurizedbefore use. The presence of coliforms is an indicator of poorhygiene and may be a potential risk for food poisoning (Wilbey2005).
Milk powder. Most information in the literature can be found forpowders intended for powdered infant formula (PIF) as this con-cerns an age group likely to experience higher disease rates. Over-all, Cronobacter sakazakii (formerly known as Enterobacter sakazakii)and Salmonella spp. (FAO/WHO 2004, 2006; WHO and FAO2007; Strydom and others 2012) can be considered microbio-logical hazards of most concern in PIF. Although both bacteria
do not grow in dry PIF, they can survive in it for long peri-ods of time (Forsythe 2005) and may grow out at a later stage,for example, after rehydration of the powder. The milk powderproduction process involves high heat, which will result in a re-duction of Cronobacter numbers and Salmonella in excess of 10 logunits (FAO/WHO 2006). End-product contamination is, there-fore, considered to be caused by recontamination of the productafter drying rather than survival of the microorganisms duringprocessing (FAO/WHO 2006; Strydom and others 2012). Re-contamination can result from addition of contaminated dry-mixingredients after spraydrying of the PIF (FAO/WHO 2006; Stry-dom and others 2012) or from the factory environment betweenspraydrying of the PIF and packaging (Strydom and others 2012;Farrokh and others 2013). Besides vegetative pathogens, sporesformed by pathogenic microorganisms (for example, B. cereus, C.botulinum, and Clostridium perfringens) can contaminate dry powderproducts. Although spores of C. botulinum have incidentally beenfound in powdered dairy products (Carlin and others 2004), theygenerally represent a low level of risk for illness (FAO/WHO 2004,2006; CAC (Codex Alimentaris Commission) 2008). Several stud-ies have detected the presence of C. perfringens in infant formula(Barash and others 2010); however, C. perfringens illnesses havenever been associated with the consumption of PIF (FAO/WHO2004; Doyle and others 2015). An evaluation by the InternationalCommission on the Microbiological Specifications of Foods con-cerning the usefulness for testing C. botulinum presence in infantformula led to the recommendation that routine testing for thispathogen is not recommended and end-product testing shouldonly serve a function in source identification during outbreaks.A B. cereus criterion has been set in European legislation for PIFand dietary food for special medical purpose intended for infantsunder 6 mo of age (Regulation (EC) 2073/2005).
Chemical hazards at primary production. Most chemical haz-ards enter the dairy chain at primary production, either throughthe ingestion of contaminated feed, via the uptake of chemicalcompounds as a result of grazing on contaminated soil or via theadministration of veterinary medicines. Another possible cause ofcontamination is through fraud, which may occur at various stagesalong the dairy supply chain. Examples are the presence of dioxinin German animal feed due to the illegal use of contaminatedtechnical fats in 2010 (Kupferschmidt 2011), the recent incidentwith furazolidone in Dutch animal feed (Dijksma 2014), and themelamine crisis in China in 2008 (Chen 2009; Pei and others2011).
Mycotoxins. Several mycotoxins may be present in feed that canbe transferred into the milk. These are aflatoxins, ochratoxin-A,fusarium toxins (for example, fumonisins, trichothecenes, andzearalenone), and ergoline alkaloids like cyclopiazonic acid(FSANZ 2006). Aflatoxins are considered the most importantmycotoxins for dietary exposure from dairy products and, sub-sequently, the only mycotoxins for which maximum limits havebeen established in milk and milk products (Regulation (EC)1881/2006). Kleter and others (2009) stated that 93% of all myco-toxin notifications from RASFF were for aflatoxins. Aflatoxins (B1,B2, G1, and G2) are mainly produced by Aspergillus spp. AflatoxinB1 (AFB1) is the most toxic of these aflatoxins and is primarilyproduced by A. flavus and A. parasiticus (Diener and others 1987).Aspergillus spp. can colonize plants in the field in hot, humid cli-mates, typically in (sub)tropical areas (Prandini and others 2009;Marin and others 2013) and their levels may increase postharvestif crops have not been adequately dehydrated (Prandini and oth-ers 2009). Contamination of feed in Europe is mainly caused by
imported products. However, in rare occasions with exceptionallyhot and dry growing seasons, aflatoxin contamination may occurin Southern Europe, as was the case in 2013 with Balkan maize(Schatzmayr and Streit 2013).
When aflatoxin B1 (AFB1) is present in feed, it can be con-verted by ruminants resulting in the presence of AFM1 in themilk, which is harmful to human health (EFSA 2004; Prandiniand others 2009). Transfer factors range from 0.015 to 0.024 forcattle, goats, and sheep (MacLachlan 2011). In general, AFM1
levels in dairy milk in Europe are low: only around 0.06% ofabout 12000 samples were above the EU limit of 0.05 µg/kg milk(EFSA 2004). However, when incidents occur, this may lead towidespread AFM1 contamination in milk as was the case in Italyin 2003 (Perrone and others 2014). It is expected that the tightrestrictions on controlling AFB1 in feed intended for dairy cattlemay not be applied in the same way for feedstuffs intended forother animals, such as goats and sheep. Therefore, milk from goatsand sheep could possibly exceed the legal limits for AFM1 (EFSA2004).
Plant toxins. Plant toxins that may be transferred into the milkare pyrrolizidine alkaloids (PAs). These compounds may be foundin forage plants and weeds, such as comfrey, Patterson’s curse,heliotrope, and ragwort (FSANZ 2006). Grazing cows will omiteating these plants, but when the meadow is mown and the grassis used to produce silage or hay, these plants and their toxins maybe consumed. Ingestion of these toxins may result in illness andeven death of the animals (EFSA 2011).
According to a recent EFSA opinion, GAP prevents the pres-ence of PAs in the cow’s diet. Only occasionally, cows are exposedto PAs and, consequently, the probability of PA poisoning in dairycows is limited. Animals have different susceptibilities toward Pas;for example, goats and sheep are more resistant toward PAs thancows (EFSA 2011). The amount of PAs excreted into the milkof animals that are exposed to PAs is low and, therefore, milkdoes not contribute highly to human PA exposure (EFSA 2011;Hoogenboom and others 2011). Nevertheless, an investigationin an outbreak of hepatic veno-occlusive disease caused by 1,2-unsaturated PAs showed that goat milk was one of the sourcescontributing to human poisoning (Kakar and others 2010). Fur-thermore, even though carry-over is low, it may still pose a humanhealth risk due to the genotoxic and carcinogenic properties ofthe compounds (Hoogenboom and others 2011).
Pesticides. Pesticides are sometimes found in milk due to theuse of contaminated feed. The major chemical groups of pes-ticides are: organochlorine pesticide (OCPs), organophosphorus(OPPs), carbamates, and pyrethroids. OCPs, such as cyclodienes,dichlorodiphenyltrichloroethane (DDT), lindane, and hex-achlorocyclohexanes, accumulate in animals and humans due totheir environmental persistence and fat solubility. The use of thesepesticides has been restricted and is now banned in most countries(Pico 2008). Most of the OPPs, such as dichlorvos, malathion, anddiazinon, are insecticides, which are widely used because of theirbroad spectrum of action as well as their lower persistence and ac-cumulation in the environment compared to OCPs (Pagliuca andothers 2006; Pico 2008). Carbamates affect the nervous system ofinsects and have a comparable mechanism to OPPs. Pyrethroidsmimic the structure and action of pyrethrins, which are insecticidalcompounds naturally found in the chrysanthemums (Chrysanthe-mum cinerariaefolium) (Pico 2008; Gupta 2012). Due to emerg-ing livestock diseases, such as blue tongue, which is transmittedby midges, the use of OPPs and pyrethroids may increase in thefuture (Maclachlan and Mayo 2013).
In the past OCPs were extensively used in tropical areas and, asthey are very persistent, they may still be found in the environ-ment. Crops grown there may, thus, become contaminated and,consequently, pesticide residues are transferred to milk when thesecrops are fed to cows (Nag 2010b). Several studies in tropical areashave shown positive milk samples. For example, a recent study inPakistan showed that more than 70% of the 150 raw milk samplescontained pesticides residues (organochlorine and pyrethroid pes-ticides) with 35% of the milk samples polluted with aldrin (Hassanand others 2014). Another study in India showed that 9.6% ofthe cow milk samples and 8.9% of buffalo milk samples were con-taminated with endosulphan residues. In total, 6.5% of the sampleshad levels above the CODEX Maximum Residue Limit (MRL) of0.1 mg/kg on fat basis. One of the factors influencing the pres-ence of residues in milk is the farmers’ lack of knowledge in theseregions regarding withdrawal periods for pesticides used on cropsthat are fed to cows (Karabasanavar and Singh 2013).
Organic pollutants and heavy metals. The use of phosphate fertil-izers, the application of contaminated material on the soil, such assewage sludge or industrial waste, and the atmospheric depositionfrom nearby industrial activities in the past or via recent incidentshave resulted in a broad range of contaminants that can be found inthe environment (Logonathan and others 2008; Nag 2010b). Thisgroup consists of PAHs, organochlorines (for example, dioxinsand dioxin-like PCBs), heavy metals, perfluorinated substances,and brominated flame retardants (Lutz and others 2006; Hoogen-boom and Fink-Gremmels 2012). Background levels of perfluori-nated substances and brominated flame retardants that are presentin the environment do not cause human health effects. They mayonly end up in cow milk during pollution events (Van Asselt andothers 2013). PAHs are largely metabolized in the cow (Rychenand others 2008) and heavy metals primarily accumulate in theliver and kidneys. Carry-over to milk is low and only significant athigh intake levels through feed (MacLachlan 2011). Therefore, themost relevant compounds in this group of organic pollutants arethe organochlorines as these compounds are very persistent andcan bioaccumulate in livestock (Rychen and others 2008). Dueto their long half-life, organochlorines can remain in the environ-ment for long periods of time (Nag 2010a). Overall, carry-overrates for organochlorines vary from 0.2% to 77% depending on thedioxin type (Hoogenboom 2005; Nag 2010a). Background levelsof dioxins and PCBs in raw milk and dairy products have beendeclining over the years in the EU (EFSA 2012c; Hoogenboomand Fink-Gremmels 2012).
However, when cows or sheep are grazing on contaminatedland, levels of dioxins and PCBs in milk may become much higher(Nag 2010b). Sheep ingest relatively more soil per kg body weightthan cows (de Vries and others 2007). Furthermore, as sheep milkcontains more fat than cow milk, it is expected that sheep milkwill contain higher levels of dioxins than cow milk when theseanimals are grazing on the same contaminated land (Jones andothers 1989; Logonathan and others 2008).
Radionuclides. The nuclear accident at Chernobyl, Ukraine, in1986 caused a deposition of radionuclides to the grass, which weretaken up by cows during grazing and subsequently transferred intotheir milk (Steinhauser and others 2014). Milk was shown to bethe dominant source contributing to iodine-131 (131I) exposurefor the local population after the incident (Steinhauser and others2014). 131I has a short half-life, around 8 d, and its uptake bycows is, thus, primarily important directly after an incident. Otherradionuclides, such as cesium-137 (137Cs) have a much longerhalf-life of around 30 y. This radionuclide can, thus, remain in the
environment long after an incident (US EPA 2015). Even yearsafter the Chernobyl incident, milk remained the major route forintake of 137Cs and contributed more than 50% to the averageintake. Recent data from the affected region still show elevated137Cs-levels in milk, but they are below the EU limits (Steinhauserand others 2014).
Veterinary drugs. Veterinary drugs are prescribed to cure animaldiseases. Antibiotics in dairy cattle are mainly used to control mas-titis (Oliver and others 2011). They have prescribed withdrawalperiods, meaning that farmers need to wait a certain period oftime after treatment before they can sell products of animal originto consumers (Directive 2001/82/EC). These withdrawal peri-ods have been established to prevent the occurrence of antibioticresidues in animal products above the MRLs as laid down in Reg-ulation (EU) 37/2010. Antibiotics use may result in the presenceof antibiotic residues in the milk when milk is delivered within thewithdrawal period (Vishnuraj and others 2016). For food safety,but also for technical reasons, the dairy industry established pri-vate monitoring programs focusing on antibiotic residues. Un-fortunately, the results from these monitoring programs are notpublicly available. The EU food safety authorities also monitorantibiotics as indicated by Directive 96/23/EC. In 2014, 0.13%of the milk samples contained antibiotic residues above the MRL.The most frequently encountered antibiotic residues in milk werebeta-lactams, such as amoxicillin, ampicillin, and benzylpenicillin(EFSA 2016). Due to an increased concern for antibiotic resis-tance, the overall antibiotics sales for livestock within the EUhas declined from 2010 to 2013, although there are differencesbetween countries (EMA 2015).
Apart from antibiotics, other veterinary medicines may be ad-ministered to cows such as painkillers (for example, NSAIDs)and antiparasitic drugs. Painkillers that are used for lactating cowshave short withdrawal periods, ranging from 0 h to 1 to 2 d(www.cbg-meb.nl). Thus, it is expected that these compoundswill not be found above the indicated MRLs in Regulation (EU)37/2010. Antiparasitic drugs are applied to treat flukes, tapeworms,and nematodes (Khaniki 2007). As with antibiotics, withdrawalperiods should be observed in order to prevent the presence ofresidues in dairy products. Antiparasitic drugs are the 2nd mostimportant group of residues found in milk, and were found tohave a noncompliance percentage of 0.07% in the EU (EFSA2016). Compounds found include endectocides and flukicides.Endectocides are broad spectrum anthelminthics that work againstnematods and arthropods. They are globally used parasiticides andinclude avermectins and milbemycins (moxidectin) (Floate 2006).Flukicides are used to treat liver flukes cause by Fasciola spp. andinclude compounds such as albendazole, triclabendazole, and lev-amisole (Shokier and others 2013). Triclabendazole is most effec-tive against liver fluke infections (Shokier and others 2013) and asa result has been used most frequently in dairy cattle (Imperialeand others 2011). Some of the endectocides and flukicides havean MRL, such as eprinomectine and albendazole, but others, suchas ivermectin, doramectin, and levamisole, are not allowed to beused in milk-producing animals according to Regulation (EU)37/010. When these compounds are found above the LOD, theyare reported as noncompliant (EFSA 2016).
Apart from the regulatory drugs, forbidden compounds aresometimes found in milk. Chloramphenicol is the most frequentlyreported compound within this group (EFSA 2016).
Chemical hazards at dairy processing. Contaminants that arealready present in the milk are usually unaffected by further pro-cessing at the dairy factory or at the farm. However, in some cases
contaminants may be concentrated, for example, in the produc-tion of milk powder, which causes a higher level of contaminantsin the final product (Prandini and others 2009). Organic pollu-tants such as dioxins and PCBs are lipophilic and will accumulatein butter, which typically contains 80% fat (Kalantzi and others2001). When contaminants are water-soluble, their levels will re-duce in butter, as is also the case for the radionuclides strontium-90(90Sr), cesium-134 (134Cs), and 137Cs (Nag 2010b). This also ap-plies to AFM1, which will primarily end up in skimmed milk andbuttermilk, yet in a much lower level in butter. Due to a concen-tration factor, AFM1 levels may be 5 times higher in hard cheesethan in milk (Prandini and others 2009). The following sectionsgive an overview of the chemical hazards that may occur duringprocessing.
Detergents and disinfectants. Both at primary production and sub-sequent processing, equipment needs cleaning and disinfection.A wide range of chemicals are currently used in dairy process-ing: acidic compounds, aldehyde-based biocides, caustic products,chlorine (for example, as sodium hypochlorite), hydrogen per-oxide, iodine, isothiazolinones, ozone, peracetic acid, phenolics,biguanidines, and surfactants (Simoes and others 2010). Sodiumhypochlorite is a compound that is used as a disinfectant. It mayresult in disinfection byproducts in food when chlorine formstrichloromethane (TCM), also called chloroform (Danaher andJordan 2013). As is to be expected, increased chlorine concentra-tions and reduced amounts of rinsing water will increase the TCMlevel in milk (Ryan and others 2012). Disinfectants have emergedas a residue in milk in recent years. Iodine residues are found as wellas quaternary ammonium compound (QAC) residues and TCMresidues (Danaher and Jordan 2013). QACs that are frequently ap-plied include didecyldimethylammonium chloride (DDAC) andbenzalkonium chloride (BAC). Residues of these compounds havebeen found in milk and ice cream as a result of disinfection;sometimes levels above the MRL of 0.1 mg/kg (Regulation (EC)396/2005) have been found (BfR 2012a,b).
Recently, chlorine dioxide has become more widely used inthe dairy industry. It can generate chlorinated byproducts (chloriteand chlorate) that could be toxic (Gomez-Lopez 2012). A recentopinion of BfR indicated that perchlorate residues are found ina range of food products in particular in plant-based foods as aresult of fertilizer use, but these residues can also result from theuse of disinfectants. Perchlorate intake can have health effects asit inhibits the iodine uptake (BfR 2013; EFSA CONTAM Panel2015).
If no proper rinsing is applied after cleaning and disinfections,residues of detergents and disinfectants may end up in the dairyproduct. Over the past 5 y, 3 notifications were reported inRASFF with too high levels of hydrogen peroxide in butter anddesserts.
Neoformed contaminants. When milk is heated, compoundspresent in the milk (lactose, protein) may follow the Maillard re-action resulting in the formation of neoformed contaminants suchas lactulosyl- or fructosyl-lysine, pyrraline, and carboxymethylly-sine (CML) (Nguyen and others 2013). The level of CML in-creases with increased temperatures and is also influenced by thewhey-to-casein ratio as well as the lactose levels in the milk. Arecent review within Europe showed a large variability in CMLlevels in infant formula. This may have been caused by the heat-sterilization techniques applied as well as the composition of themilk used (Birlouez-Aragon and others 2010). A good controlof the production process will thus diminish the CML levels inthe final product (Birlouez-Aragon and others 2010; Nguyen and
others 2013). Currently, there is no legislation for these processingcontaminants (Birlouez-Aragon and others 2010).
Migrants from food contact materials. Compounds that can migratefrom packaging material or equipment into the dairy products arephthalates. Phthalates are the most used plasticizers and are addedto plastic polymers (such as PVC) to enhance flexibility. Variousphthalate esters have been found in dairy products, presumablyfrom PVC tubing used during the milking process or transferfrom bulk milk to storage tanks but also from packaging materialssuch as cartons and bottles (Danaher and Jordan 2013). A Belgianstudy found phthalates in almost all milk and dairy products testedin concentrations ranging from below the LOD to a maximum of743 µg/kg for di(2-ethylhexyl) phthalate (DEHP) (Fierens andothers 2012). Currently, there are no EU MRLs for these com-pounds. Migration of phthalates, such as DEHP, may be ofconcern in dairy products due to the high lipid content ofdairy products. As a result, DEHP is being replaced with di-(2-ethylhexylexyl)adipate, which is less toxic than DEHP (FSANZ2006). A recent U.S. study indicated that high meat and dairyconsumption may lead to elevated DEHP-intake (Samantha ESerrano1 2014).
Another group of compounds that may come into contact withdairy products is printing ink ingredients, such as 2-isopropylthioxanthone (ITX) and 2-ethylhexyl-4-dimethylaminobenzoate.In 2005, ITX contamination of milk for children was reportedto RASFF. In that year, ITX notifications contributed to 32% ofall food contact substances reported to RASFF (Kleter and oth-ers 2009). According to EFSA, young children may have a higherexposure to these printing inks as about half of their foods and bev-erages are packed in cartons printed with these inks (EFSA 2005).As a result, the EU has adopted legislation requiring that transferof printing ink compounds to food surfaces (through “set-off” ormigration) must not occur (Regulation (EC) 2023/2006).
A more recent concern is the presence of aluminum in infantformula. Most packaging material contains an aluminum layer,which may result in migration of aluminum into the product(Chuchu and others 2013). Use of brand-specific infant formulascan lead to higher exposure levels because of the variety in alu-minum levels between brands. EFSA (2008) determined that themain foods contributing to dietary exposure were cereals (prod-ucts), vegetables, beverages, and certain infant formulas (EFSA2008). The relative contributions of the various products to alu-minum exposure were not mentioned in this report, but dairyproducts are not the major contributor.
Melamine. A well-known example of food fraud in the dairysector is the melamine case in China. In 2008, melamine wasillegally added to milk products to produce an incorrectly highreading in the measurement of protein content based on totalnitrogen (Ai and others 2009). By the end of November 2008,some 294000 infants and young children had been diagnosed withurinary tract stones (Chen 2009). More than 50000 infants werehospitalized, with 6 confirmed deaths (WHO 2008). It is hard toimagine how the adulterated milk could have passed all the qualityinspections along their supply chains to reach the marketplace onsuch a massive scale (Chen and others 2014). Specific controls formilk quality like fat content should have detected such fraud, butthese tests were either not carried out properly or were ineffective(Pei and others 2011). Recently, rapid surveillance methods havebeen developed to detect nitrogen-containing compounds in milk(Abernethy and Higgs 2013).
Physical hazards. Physical hazards that might be present indairy products include metal parts, sand/soil, stones, wood, plastic,
Figure 3–Overview of microbiological (blue), chemical (purple), and physical (green) hazards for milk production at the dairy farm. The most importanthazards are bolded.
rubber, or glass fragments and hair. They may be introduced dur-ing processing via parts of machinery (such as metal parts fromstirrers or rubber shreds from seals) or equipment, due to pack-aging materials, via jewelry worn by personnel, or via presencein raw materials. At the farm, physical hazards may be introducedduring milking (such as machine parts). However, in most cases,physical hazards are introduced during the processing of dairyproducts. When glass bottles are used, glass fragments may endup in the final products. Hair may be present in dairy productsdue to poor hygiene control. Hair can be a source of microbio-logical contamination. Furthermore, improper cleaning and dis-infection may result in the presence of what is considered “soil”(FAVV 2012).
Physical hazards are controlled by Good Manufacturing Prac-tices and are prevented in the final products through visual obser-vations or methods such as metal detection as part of a HACCP-system (van Schothorst and Kleiss 1994; Peariso 2007).
Future trendsWorldwide, the challenges in the dairy sector are: changing
consumer demands, a growing concern about sustainability, anda need for greater efficiency. The global demand for dairy prod-ucts is rapidly increasing, driven by the growing population andpurchasing capacity in the developing countries (FAO and IFCN2010; Gerosa and Skoet 2012). Climate change is considered an-other important driver that can have an impact on future foodsafety (FAO and IFCN 2010). Increased temperatures and humid-ity may affect mycotoxin production (Paterson and Lima 2010),animal diseases transmitted by insects (Maclachlan and Mayo2013), and the microbiological ecology, including pathogens and
molds. It may become difficult to maintain the cold supply chainwhen ambient temperatures rise (Havelaar and others 2010).
On the other hand, developments in microbiological and chem-ical detection methodologies can have a positive effect on foodsafety and can be used to improve tracking and tracing systems(Havelaar and others 2010).
Animal feed. It is expected that dairy production will increaseglobally leading to increased demands for animal feed. Further-more, as a result of climate change, higher mycotoxin occurrenceis expected in animal feed. This results in a stricter demand formycotoxin-free feed. According to experts, there is an increaseddemand for feed that does not contain genetically modified or-ganisms (GMOs). These factors may cause scarcity in certain areasand thus shifts to either other trade partners, other regions, orother feed ingredients. In order to secure the feed supply, morefeed materials will have to be grown within Europe, possibly re-sulting in new or different grain varieties (NZO 2015). Examplesare lupine, which may contain the mycotoxin phomopsin (EFSA2012b) and rapeseed, as well as the use of alternative proteins (suchas from eby-products from biofuel production, algae or insects).These developments may lead to a more complex and dynamicfeed chain with increased numbers of trade partners. This willhave consequences for food safety as the chain will become lesstransparent.
Another trend affecting feed materials is the increased produc-tion of biofuel. Some of the raw materials used may contain chem-ical hazards. For example, maize may be contaminated with myco-toxins, sorghum may contain plant toxins, and castor beans (usedin oil production in Africa) contain the toxic ricin. Due to theproduction process, chemical hazards present in the raw materialsfor biofuel will be concentrated in the byproducts such as dried
Figure 4–Overview of microbiological (blue), chemical (purple), and physical (green) hazards reported in scientific literature for dairy productsprocessed at the dairy farm. The most important hazards are bolded.
distillers grains with solubles that are used in the feed industry(van Asselt and others 2011). Changes within the animal feed sec-tor are more likely affecting chemical and physical hazards thanmicrobiological hazards.
Dairy farm. The predominant trend foreseen for the coming10 y is the intensification of the dairy sector both for cowand goat milk, which is caused by the abolishment of themilk quota (Rabobank 2015; Tropea 2015) and the growingglobal population (NZO 2015; Promar International 2015).Milk quotas were 1st introduced in 1984 with the aim to limitmilk production in Europe. In April 2015, the milk quotawas dropped allowing farmers to expand their productionprovided their phosphate output is within limits (Jongeneel andvan Berkum 2015). Worldwide production to date is around700 million tons/y (FAOSTAT). The world milk productionis expected to have increased by 180 million tons by 2023.Most of this increase will be produced in developing countries(OECD/Food and Agriculture Organization of the UnitedNations 2014), with an expected increase in antibiotics use asa result (van Boeckel and others 2015). Within Europe, factorssuch as environmental restrictions, limited land availability, or alack of capital can negatively affect dairy production (Koeleman2015). Nevertheless, the number of farms is expected to decrease,and the average herd size is expected to increase (EU AgriculturalEconomics 2013). Currently, EU farm types differ substantially,ranging from extensive pasture-based farms to intensive (family-run) housed systems (Promar International 2015). The dairysector is expected to shift from the family business model towardmore specialized and intensive industrial farming (Tropea 2015)
leading to increased automation (such as milking robots). As aconsequence, the farmer will have less direct contact with the cowsand will be more dependent on technology to monitor health ofthe cow and hygiene during milking (EU 2014). The advantage offurther intensification is that farmers will be able to invest in theirfarm and will become more aware of food safety (NZO 2015).
Other factors that may affect cattle rearing, and consequently,the farming business stem from consumer wishes such as increasedattention to animal welfare and environmental sustainability (LTONederland 2011; NZO 2015; Rabobank 2015) For example, in-vesting in sustainable livestock housing can contribute to a higheraverage age of the cow and increased animal health (Rabobank2013; NZO 2015). Sustainability measures may also negatively af-fect food safety. For example, the implementation of recycling atthe farm (Petersen and others 2007).
Milk processing. Increased global demands for dairy productswill result in increased trade in dairy products, predominantlyproducts with a long shelf-life such as milk powder, butter, and,to a lesser extent, cheese. By 2024, production of skim milk pow-der is expected to reach 1.6 million kg (EU 2014). The highercheese production in the EU is mainly driven by a higher domes-tic consumption rather than by increased export. By 2024, EUcheese production could reach 11 million kg, which is 1.15 mil-lion kg more compared to 2014 levels (EU 2014). Whey powderis an important ingredient for infant formulas and trade in wheypowder is expanding, especially toward China. Its production isexpected to increase by 20% to 2.5 million kg in 2024 (EU 2014).Of this increase, 35% is expected to be exported as whey pow-der in its original form. The remainder will be used for animal
Figure 5–Overview of microbiological (blue), chemical (purple), and physical (green) hazards reported in scientific literature for dairy productsprocessed at the dairy factory. The most important hazards are bolded.
feed (although this market is declining), or will be used as foodsupplements, sports drinks, and, predominantly, infant formulas(EU 2014). Dairy processors require flexibility and need to adapttheir product mix based on global supply and demand (NZO2015).
Furthermore, consumer trends are affecting dairy processing.There is an increasing demand for convenience products suchas precut slices of cheese and grated cheese. These products aretypically vulnerable to contamination, especially contaminationby L. monocytogenes (Schoder and others 2015). There is also anincreasing demand for milk produced from cows that have outdooraccess, for organic dairy products, and for fresh and shelf-stableproducts. In the next years, organic farming is expected to increase.
Additionally, there is a trend toward raw and minimally processedproducts (Falguera and others 2012), which may pose humanhealth issues as raw milk cheeses (especially soft cheeses) havefrequently been contaminated with L. monocytogenes, Salmonellaspp., and E. coli (see RASFF). In order to meet consumer demandsregarding minimally processed foods, alternative techniques can beused such as extreme fast heating, pulsed electric field processing,ultraviolet treatment, ultrasound processing, the addition of carbondioxide, and high-pressure processing (Ross and others 2003).Such novel technologies require an adequate validation of safetybefore implementation, as shelf-life extension may pose a risktoward the microbiological safety of dairy products (Ross andothers 2003; Sobrino-Lopez and Martın-Belloso 2009).
There is also an increased consumer demand for goat milk,which causes increased goat milk processing. However, goat milkis expected to remain a marginal product compared to cow milk(IDF 2013).
Furthermore, Internet commerce is expanding and sales alsoinclude dairy products. Sampling, control, and trader identificationposes challenges to authorities (Schoder and others 2015). A recentmicrobiological investigation of cheeses purchased online revealedlabeling (such as raw or pasteurized milk cheese), hygiene, andsafety of those products as major points of concern (Schoder andothers 2015).
ConclusionsIn general, most information found in the scientific literature
on food safety hazards was related to dairy cows and dairy productsfrom cow milk. Information on goat and sheep milk was limited,which is expected as goat milk and especially sheep milk representonly a fraction of the total volumes of dairy products produced inEurope. Food safety hazards can enter the dairy cow supply chainat various points, via animal feed, through the dairy farm envi-ronment (Figure 3), or during further processing (Figure 4 and5). When comparing various food safety hazards, microbiologicalhazards were encountered more often than chemical or physicalhazards in dairy products. The most important microbiologicalhazard introduced at the dairy farm is S. aureus, which is presentin cows suffering from mastitis and is subsequently transferred tothe milk. Cheese is most frequently associated with L. monocy-togenes followed by Salmonella spp. and Shiga toxin-producing E.coli. Soft and semisoft cheese form a higher risk for presence ofL. monocytogenes but S. aureus enterotoxins may also be present insoft and semisoft cheese, especially when the fermentation pro-cess is retarded or initial cell numbers in the milk are high. Formilk powder products, Cronobacter spp. and Salmonella spp. can beconsidered microbiological hazards of most concern, especially inPIF intended for infants younger than 6 mo. Chemical hazardsare primarily introduced at the dairy farm, via contaminated feed,through the environment, or due to veterinary treatments of theanimals. Further processing may cause an accumulation of thesehazards depending on the product produced. Based on monitoringdata and literature review, the most important chemical hazards fordairy products are aflatoxin M1, environmental contaminants (pri-marily dioxin and dioxin-like compounds), and veterinary drugs.The most important physical hazards in dairy products are metal,glass, and plastic particles. However, these hazards can easily bedetected and dairy producers have an incentive to prevent them asrecalls are expensive. Therefore, physical hazards are less relevantfor dairy products than microbiological and chemical hazards.
Various trends are foreseen in the near future that may affectfood safety, such as climate change, increased global trade, andchanging consumer demands. The most important developmentis the further intensification of the dairy chain and higher milkproduction due to the abolishment of the milk quota and growingdemands. This may have positive effects for food safety as farmerscan invest in improving their farms and will be more aware of pos-sible food safety issues. However, an increased livestock populationat the farm may also result in a raise in the occurrence of animaldiseases and, consequently, more veterinary drug use. Anotherconsequence of increased production is the increased pressure onthe animal feed market, which may affect the quality and safety offeed products. At the farm level, changes in production systems asa result of increased awareness for sustainability and animal welfare
should be followed closely in order to evaluate their consequencesfor food safety.
AcknowledgmentsThe authors thank the dairy experts who cooperated in this
study. Mariel Pikkemaat (RIKILT) is kindly thanked for her con-tribution to the section on veterinary drugs. Jennifer Banach(RIKILT) is thanked for critically reading this paper and improvingthe English. Furthermore, the financial contribution from theNVWA is highly appreciated.
Author contributionsEsther van Asselt collected the information on chemical and
physical hazards and wrote this paper. Hermien van Bokhorst-vande Veen collected and described the information on microbio-logical hazards. Ine van der Fels-Klerx, Hans Marvin, and MasjaNierop Groot were involved in defining the study setup, con-tributing expert opinions on the food safety hazards, and criticallyreading and commenting on the paper.
AbbreviationsAFB1 Aflatoxin B1
AFM1 Aflatoxin M1
BAC benzalkonium chlorideCAC Codex Alimentaris CommissionCfu Colony forming unitsCML CarboxymethyllysineDDAC Didecyldimethylammonium chloridDDT DichlorodiphenyltrichloroethaneDEHP Di(2-ethylhexyl) phthalateEFSA European Food Safety AuthorityEU European UnionFAO Food and Agriculture Organization of the United Na-
tionsG.A.P. Good Agricultural PracticesGEMS Global Environment Monitoring System of the WHOGMO Genetically Modified OrganismsHACCP Hazard Analysis Critical Control PointsIDF International Dairy FederationITX 2-isopropyl thioxanthoneLOD Limit of detectionMRL Maximum Residue Level ((EC) 396/2005) or Maxi-
Abernethy G, Higgs K. 2013. Rapid detection of economic adulterants infresh milk by liquid chromatography–tandem mass spectrometry.J Chromatogr A 1288:10–20.
ACMSF. 2013. ACMSF Report on risk profile in relation to toxoplasma inthe food chain. Advisory Committe on the Microbiological Safety of FoodMicrobiology Reports.
Ai K, Liu Y, Lu L. 2009. Hydrogen-bonding recognition-induced colorchange of gold nanoparticles for visual detection of melamine in raw milkand infant formula. J Am Chem Soc 131:9496–7.
Anonymous. 2003. Opion of the scientific committtee on veterinarymeasures relating to public health on staphylococcal enterotoxins in milkproducts, particularly cheeses.
Available from: http://www2.esb.ucp.pt/twt/seg_alim/codigosguias/SaureusAlimentosParecerCE.pdf. Accessed 2015 May 16.
Barkema HW, Deluyker HA, Schukken YH, Lam TJGM. 1999.Quarter-milk somatic cell count at calving and at the first six milkings aftercalving. Prev Vet Med 38:1–9.
BfR. 2012a. Gesundheitliche Bewertung der Ruckstande vonBenzalkoniumchlorid in Lebensmitteln. Berlin, Germany: The Federal Inst.for Risk Assessment.
BfR. 2012b. Health assessment of didecyldimethylammonium chloride(DDAC) residues in food. Berlin, Germany: The Federal Inst. for RiskAssessment (BfR).
BfR. 2013. Health assessment of perchlorate residues in foods. Berlin,Germany: The Federal Inst. for Risk Assessment (BfR).
Birlouez-Aragon I, Morales F, Fogliano V, Pain JP. 2010. The health andtechnological implications of a better control of neoformed contaminants bythe food industry. Pathol Biol 58:232–8.
Braunig J, Hall P. 2005. Milk and dairy products. In: Roberts TA, CordierJL, Gram L, Tompkin RB, Pitt JI, Gorris LGM, Swanson KMJ, editors.Micro-organisms in foods 6. New York, USA: Kluwer Academic/PlenumPublishers. p 643–715.
CAC (Codex Alimentaris Commission). 2008. Code for hygienic practicefor powdered infant formula for infants and young children. CAC/RPC66-2008 Viewed 12 March 2015.
Carlin F, Broussolle V, Perelle S, Litman S, Fach P. 2004. Prevalence ofClostridium botulinum in food raw materials used in REPFEDs manufacturedin France. Int J Food Microbiol 91:141–5.
Chen C, Zhang J, Delaurentis T. 2014. Quality control in food supply chainmanagement: an analytical model and case study of the adulterated milkincident in China. Int J Prod Econ 152:188–99.
Chen J-S. 2009. What can we learn from the 2008 melamine crisis in China?Biomed Environ Sci 22:109–11.
Chuchu N, Patel B, Sebastian B, Exley C. 2013. The aluminium content ofinfant formulas remains too high. BMC Pediat 13:162.
Claeys WL, Cardoen S, Daube G, de Block J, Dewettinck K, Dierick K, deZutter L, Huyghebaert A, Imberechts H, Thiange P, Vandenplas Y,Herman L. 2013. Raw or heated cow milk consumption: review of risksand benefits. Food Control 31:251–62.
Cretenet M, Even S, Le Loir Y. 2011. Unveiling Staphylococcus aureusenterotoxin production in dairy products: a review of recent advances toface new challenges. Dairy Sci Technol 91:127–50.
Dalmasso M, Jordan K. 2014. Absence of growth of Listeria monocytogenes innaturally contaminated Cheddar cheese. J Dairy Res 81:46–53.
Danaher M, Jordan K. 2013. Identification of existing and emerging chemicalresidue contamination concerns in milk. Ir J Agric Food Res 52:173–83.
de Vries W, Romkens PAM, Schutze G. 2007. Critical soil concentrations ofcadmium, lead, and mercury in view of health effects on humans andanimals. In: Ware GW, Whitacre DM, editors. Reviews of environmentalcontamination and toxicology. New York: Springer. p 91–130.
Diener UL, Cole RJ, Sanders TH, Payne GA, Lee LS, Klich MA. 1987.Epidemiology of aflatoxin formation by Aspergillus flavus. Annu RevPhytopathol 25:249–70.
Dijksma SAM. 2014. Kamerbrief over de vervolgmaatregelen na vondstverboden antibioticum in de veehouderij. The Hague, The Netherlands:Ministry of Economic Affiars.
Doyle CJ, Gleeson D, Jordan K, Beresford TP, Ross RP, Fitzgerald GF,Cotter PD. 2015. Anaerobic sporeformers and their significance with
respect to milk and dairy products. Int J Food Microbiol 197:77–87.
EFSA. 2004. Opinion of the scientific panel on contaminants in the foodchain on a request from the commission related to aflatoxin B1 asundesirable substance in animal feed. EFSA J 39:1–27.
EFSA. 2005. Opinion of the scientific panel on food additives, flavourings,processing aids and materials in contact with food on a request from thecommission related to 2-Isopropyl thioxanthone (ITX) and2-ethylhexyl-4-dimethylaminobenzoate (EHDAB) in food contactmaterials. EFSA J 293:1–15.
EFSA. 2008. Scientific opinion of the panel on food additives, flavourings,processing aids and food contact materials (AFC) - Safety of aluminiumfrom dietary intake. EFSA J 754: 1–34
EFSA. 2009a. Scientific opinion of the panel on animal health and welfare ona request from the commission on the risk assessment of the impact ofhousing, nutrition and feeding, management and genetic selection on udderproblems in dairy cows. EFSA J 1141:1–60.
EFSA. 2009b. Scientific opinion of the panel on biological hazards on arequest from the European commission on food safety aspects of dairy cowhousing and husbandry systems. EFSA J 1189:1–27.
EFSA. 2011. Scientific opinion on pyrrolizidine alkaloids in food and feed.EFSA J 9:2406.
EFSA. 2012a. The European Union summary report on trends and sources ofzoonoses, zoonotic agents and food-borne outbreaks in 2010. EFSA J10:1–442.
EFSA. 2012b. Scientific opinion on the risks for animal and public healthrelated to the presence of phomopsins in feed and food. EFSA J 10:2567.
EFSA. 2012c. Update of the monitoring of levels of dioxins and PCBs infood and feed. EFSA J 10:2832.
EFSA. 2013a. Analysis of the baseline survey on the prevalence of Listeriamonocytogenes in certain ready-to-eat (RTE) foods in the EU, 2010-2011Part A: Listeria monocytogenes prevalence estimates. EFSA J 11:1–75.
EFSA. 2013b. The European Union summary report on trends and sourcesof zoonoses, zoonotic agents and food-borne outbreaks in 2011. EFSA J11:1–250.
EFSA. 2014a. The European Union summary report on antimicrobialresistance in zoonotic and indicator bacteria from humans, animals and foodin 2012. EFSA J 12:3590.
EFSA. 2014b. The European Union summary report on trends and sourcesof zoonoses, zoonotic agents and food-borne outbreaks in 2012. EFSA J12:1–312.
EFSA. 2014c. Report for 2012 on the results from the monitoring ofveterinary medicinal product residues and other substances in liveanimals and animal products. EFSA supporting publication 2014:EN-540,65 p.
EFSA. 2015a. The European Union summary report on trends and sources ofzoonoses, zoonotic agents and food-borne outbreaks in 2013. EFSA J13:1–162.
EFSA. 2015b. Report for 2013 on the results from the monitoring ofveterinary medicinal product residues and other substances in live animalsand animal products. EFSA supporting publication 2015:EN-723, 69 p.
EFSA. 2015c. Scientific opinion on the public health risks related to theconsumption of raw drinking milk. EFSA J 13:1–95.
EFSA. 2016. Report for 2014 on the results from the monitoring ofveterinary medicinal product residues and other substances in live animalsand animal products. EFSA supporting publication 2016:EN-923, 70 p.
EFSA CONTAM Panel. 2015. Scientific opinion on risks for public healthrelated to the presence of chlorate in food. EFSA J 13:4135.
EMA. 2015. Sales of veterinary antimicrobial agents in 26 EU/EEAcountries in 2013. (EMA/387934/2015).
EU. 2014. Medium-term prospects for EU agricultural markets and income2014-2024. EU report 12/2014.
EU Agricultural Economics. 2013. Structure and dynamics of EU farms:changes, trends and policy relevance. Available from:http://ec.europa.eu/agriculture/rural-area-economics/briefs/pdf/09_en.pdf.Accessed 2016 January 26.
Eurostat. 2015. Milk and milk product statistics. Available from:http://ec.europa.eu/eurostat/statistics-explained/index.php/Milk_and_milk_product_statistics#Publications. Accessed 2016 January 22.
Falguera V, Aliguer N, Falguera M. 2012. An integrated approach to currenttrends in food consumption: moving toward functional and organicproducts? Food Control 26:274–81.
FAO and IFCN. 2010. Global dairy sector: status and trends. In: Hemme T,Otte J, editors. Status and prospects for smallholder milk production—aglobal perspective. Rome, Italy: FAO. p 16–33.
FAO Stat. 2015. Production—livestock primary. Available from:http://faostat3.fao.org/browse/Q/QL/E. Accessed 2015 January 21.
FAO/WHO. 2004. Enterobacter sakazakii and other microorganisms inpowdered infant formula. Microbiological risk assessment series no. 6.
FAO/WHO. 2006. Enterobacter sakazakii and Salmonella in powdered infantformula. Microbiological risk assessment series no. 10.
Farrokh C, Jordan K, Auvray F, Glass K, Oppegaard H, Raynaud S,Thevenot D, Condron R, de Reu K, Govaris A, Heggum K, HeyndrickxM, Hummerjohann J, Lindsay D, Miszczycha S, Moussiegt S, Verstraete K,Cerf O. 2013. Review of Shiga-toxin-producing Escherichia coli (STEC) andtheir significance in dairy production. Int J Food Microbiol 162:190–212.
FAVV. 2012. Autocontrolegids voor de productie en verkoop vanzuivelproducten op het landbouwbedrijf. Brussels, Belgium: Federaalagentschap voor de veiligheid van de voedselketen.(G-034).
Fierens T, Servaes K, van Holderbeke M, Geerts L, de Henauw S, Sioen I,Vanermen G. 2012. Analysis of phthalates in food products and packagingmaterials sold on the Belgian market. Food Chem Toxicol 50:2575–83.
Floate KD. 2006. Endectocide use in cattle and fecal residues: environmentaleffects in Canada. Can J Vet Res 70:1–10.
Forsythe SJ. 2005. Enterobacter sakazakii and other bacteria in powdered infantmilk formula. Matern Child Nutr 1:44–50.
FSANZ. 2006. A risk profile of dairy products in Australia. Kingston,Australia: Food Standards Australia New Zealand.
Gerosa S, Skoet J. 2012. Milk availability—trends in production and demandand mid-term outlook. Rome, Italy: FAO–Agricultural DevelopmentEconomics Div.
Gomez-Lopez VM. 2012. Chlorine dioxide. In: Gomez-Lopez VM, editor.Decontamination of fresh and minimally processed produce. Oxford, UK:Wiley-Blackwell. p 165–75.
Gudmundson J, Chirino-Trejo JM. 1993. A case of bovine mastitis caused byCampylobacter jejuni. Zentralblatt fur Veterinarmedizin Reihe B J Vet MedSeries B 40:326–8.
Hassan AU, Tabinda AB, Abbas M, Khan AM. 2014. Organochlorine andPyrethroid pesticide analysis in dairy milk samples collected from cottongrowing belt of Punjab, Pakistan. Pak J Agric Sci 51:321–5.
Havelaar AH, Brul S, de Jong A, de Jonge R, Zwietering MH, ter Kuile BH.2010. Future challenges to microbial food safety. Int J Food Microbiol 139Supplement:S79–94.
Hoogenboom LAP. 2005. Behavior of polyhalogenated and polycyclicaromatic hydrocarbons in food-producing animals. In: Preedy VR, WatsonRR, editors. Reviews in food and nutrition toxicity. Boca Raton, USA:CRC Press. p 269–99.
Hoogenboom LAP, Mulder PPJ, Zeilmaker MJ, van den Top HJ,Remmelink GJ, Brandon EFA, Klijnstra M, Meijer GAL, Schothorst R,van Egmond HP. 2011. Carry-over of pyrrolizidine alkaloids from feed tomilk in dairy cows. Food Addit Contam A 28:359–72.
Hoogenboom R, Fink-Gremmels J. 2012. Animal feed contamination bydioxins, polychlorinated biphenyls (PCBs) and brominated flame retardants.In: Fink-Gremmes L, editor. Animal feed contamination. Camrbidge, UK:Woodhead Publishing. p 131–82.
Hutchinson DN, Bolton FJ, Hinchliffe PM, Dawkins HC, Horsley SD,Jessop EG, Robertshaw PA, Counter DE. 1985. Evidence of udderexcretion of Campylobacter jejuni as the cause of milk-borne campylobacteroutbreak. J Hyg (Lond) 94:205–15.
IDF. 2013. The world dairy situation 2013. Brussels, Belgium: IDF.
Imperiale F, Ortiz P, Cabrera M, Farias C, Sallovitz JM, Iezzi S, Perez J,Alvarez L, Lanusse C. 2011. Residual concentrations of the flukicidalcompound triclabendazole in dairy cows’ milk and cheese. Food AdditContam A 28:438–45.
Jayarao BM, Henning DR. 2001. Prevalence of foodborne pathogens in bulktank milk. J Dairy Sci 84:2157–62.
Jensen NE, Aarestrup FM, Jensen J, Wegener HC. 1996. Listeriamonocytogenes in bovine mastitis. Possible implication for human health. Int JFood Microbiol 32:209–16.
Jones D, Safe S, Morcom E, Holcomb M, Coppock C, Ivie W. 1989.Bioavailability of grain and soil-borne tritiated
2,3,7,8-tetrachlorodibenzo-p-dioxin (TCDD) administered to lactatingHolstein cows. Chemosphere 18:1257–63.
Jongeneel R, van Berkum S. 2015. What will happen after the EU milk quotasystem expires in 2015? The Hague, the Netherlands: LEI Wageningen UR.
Julien MC, Dion P, Lafreniere C, Antoun H, Drouin P. 2008. Sources ofclostridia in raw milk on farms. Appl Environ Microbiol 74:6348–57.
Kakar F, Akbarian Z, Leslie T, Mustafa ML, Watson J, van Egmond HP,Omar MF, Mofleh J. 2010. An outbreak of hepatic veno-occlusive disease inwestern Afghanistan associated with exposure to wheat flour contaminatedwith pyrrolizidine alkaloids. J Toxicol 2010:Article ID 313280,7 pages.
Kalantzi OI, Alcock RE, Johnston PA, Santillo D, Stringer RL, ThomasGO, Jones KC. 2001. The global distribution of PCBs and organochlorinepesticides in butter. Environ Sci Technol 35:1013–8.
Karabasanavar NS, Singh SP. 2013. Occurrence of endosulphan residues indairy milk in plains of Uttarakhand, India: short communication. Int J DairyTechnol 66:189–93.
Khaniki GRJ. 2007. Chemical contaminants in milk and public healthconcerns: a review. Int J Dairy Sci 2:104–15.
Kleter GA, Prandini A, Filippi L, Marvin HJP. 2009. Identification ofpotentially emerging food safety issues by analysis of reports published bythe European Community’s Rapid Alert System for Food and Feed(RASFF) during a four-year period. Food Chem Toxicol 47:932–50.
Koeleman E. 2015. Milk quota end, new limiting factors arise. Dairy Global:8 Jan 2015.
Kousta M, Mataragas M, Skandamis P, Drosinos EH. 2010. Prevalence andsources of cheese contamination with pathogens at farm and processinglevels. Food Control 21:805–15.
Kupferschmidt K. 2011. Dioxin scandal triggers food debate in Germany.Can Med Assoc J 183:E221–2.
Kurt L, Ozilgen S. 2013. Failure mode and effect analysis for dairy productmanufacturing: practical safety improvement action plan with cases fromTurkey. Safety Sci 55:195–206.
LeJeune JT, Rajala-Schultz PJ. 2009. Unpasteurized milk: a continued publichealth threat. Clin Infec Dis 48:93–100.
Lindstrom M, Myllykoski J, Sivela S, Korkeala H. 2010. Clostridium botulinumin cattle and dairy products. Crit Rev Food Sci Nutr 50:281–304.
Lira WM, Macedo C, Marin JM. 2004. The incidence of Shigatoxin-producing Escherichia coli in cattle with mastitis in Brazil. J ApplMicrobiol 97:861–6.
Logonathan P, Hedley MJ, Grace ND. 2008. Pasture soils contaminated withfertilizer-derived cadmium and fluorine: livestock effects. Rev EnvironContam T 192:29–66.
Lomonaco S, Decastelli L, Nucera D, Gallina S, Manila Bianchi D, Civera T.2009. Listeria monocytogenes in gorgonzola: subtypes, diversity and persistenceover time. Int J Food Microbiol 128:516–20.
Longhi C, Maffeo A, Penta M, Petrone G, Seganti L, Conte MP. 2003.Detection of Listeria monocytogenes in Italian-style soft cheeses. J ApplMicrobiol 94:879–85.
LTO Nederland. 2011. melkveehouderij - Midden in de maatschappij. TheHague, The Netherlands: LTO Nederland vakgroep melkveehouderij.
Lutz S, Feidt C, Monteau F, Rychen G, Le Bizec B, Jurjanz S. 2006. Effectof exposure to soil-bound polycyclic aromatic hydrocarbons on milkcontaminations of parent compounds and their monohydroxylatedmetabolites. J Agric Food Chem 54:263–8.
MacLachlan DJ. 2011. Estimating the transfer of contaminants in animalfeedstuffs to livestock tissues, milk and eggs: a review. Anim Prod Sci51:1067–78.
Maclachlan NJ, Mayo CE. 2013. Potential strategies for control ofbluetongue, a globally emerging, culicoides-transmitted viral disease ofruminant livestock and wildlife. Antivir Res 99:79–90.
Marchand S, de Block J, de Jonghe V, Coorevits A, Heyndrickx M, HermanL. 2012. Biofilm formation in milk production and processingenvironments; influence on milk quality and safety. Comp Rev Food SciFood Saf 11:133–47.
Marin S, Ramos AJ, Cano-Sancho G, Sanchis V. 2013. Mycotoxins:occurrence, toxicology, and exposure assessment. Food Chem Toxicol60:218–37.
Motarjemi Y, Moy GG, Jooste PJ, Anelich LE. 2014. Milk and dairyproducts. In: Motarjemi Y, Lelieveld H, editors. Food safety management: apractical guide for the food industry. London, UK: Academic Press. p83–117.
Nag SK. 2010a. Contaminants in milk: routes of contamination, analyticaltechniques and methods of control. In: Griffiths MW, editor. Improving thesafety and quality of milk volume 1: milk production and processing.Oxford, UK: Woodhead Publishing Ltd. p 146–78.
Nag SK. 2010b. Pesticides, veterinary residues and other contaminants inmilk. In: Griffiths MW, editor. Improving the safety and quality of milkvolume 1: milk production and processing. Oxford, UK: WoodheadPublishing Ltd. p 113–45.
Nguyen HT, van der Fels-Klerx HJ, van Boekel MAJS. 2013. Nϵ-(carboxymethyl)lysine: a review on analytical methods, formation, andoccurrence in processed food, and health impact. Food Rev Int 30:36–52.
Noordhuizen JPTM, Metz JHM. 2005. Quality control on dairy farms withemphasis on public health, food safety, animal health and welfare. LivestProd Sci 94:51–9.
NZO. 2015. De witte motor - De kracht en de uitdagingen van deNederlandse Zuivelsector.
OECD/Food and Agriculture Organization of the United Nations. 2014.ECD-FAO Agricultural Outlook 2014.
Olde Riekerink RGM, Barkema HW, Stryhn H. 2007. The effect of seasonon somatic cell count and the incidence of clinical mastitis. J Dairy Sci90:1704–15.
Oliver SP, Murinda SE, Jayarao BM. 2011. Impact of antibiotic use in adultdairy cows on antimicrobial resistance of veterinary and human pathogens: acomprehensive review. Foodborne Pathog Dis 8:337–55.
Pagliuca G, Serraino A, Gazzotti T, Zironi E, Borsari A, Rosmini R. 2006.Organophosphorus pesticides residues in Italian raw milk. J Dairy Res73:340.
Papademas P, Bintsis T. 2010. Food safety management systems (FSMS) inthe dairy industry: a review. Int J Dairy Technol 63:489–503.
Paterson RRM, Lima N. 2010. How will climate change affect mycotoxinsin food? Food Res Int 43:1902–14.
Peariso D. 2007. Preventing foreign material contamination of foods.Hoboken, USA: Wiley-Blackwell, 306 pages.
Pei X, Tandon A, Alldrick A, Giorgi L, Huang W, Yang R. 2011. TheChina melamine milk scandal and its implications for food safety regulation.Food Pol 36:412–20.
Perrone G, Logrieco AF, Gallo A. 2014. Biodiversity of Aspergillus sectionFlavi in Europe in relation to the management of aflatoxin risk. FrontMicrobiol 5:1–5.
Petersen SO, Sommer SG, Beline F, Burton C, Dach J, Dourmad JY, Leip A,Misselbrook T, Nicholson F, Poulsen HD, Provolo G, Sørensen P, VinnerasB, Weiske A, Bernal MP, Bohm R, Juhasz C, Mihelic R. 2007. Recyclingof livestock manure in a whole-farm perspective. Livest Sci 112:180–91.
Pico Y. 2008. Food contaminants and residue analysis. Amsterdam, theNetherlands: Elsevier.
Pintado CMBS, Oliveira A, Pampulha ME, Ferreira MASS. 2005. Prevalenceand characterization of Listeria monocytogenes isolated from soft cheese. FoodMicrobiol 22:79–85.
Prandini A, Tansini G, Sigolo S, Filippi L, Laporta M, Piva G. 2009. On theoccurrence of aflatoxin M1 in milk and dairy products. Food Chem Toxicol47:984–91.
Promar International. 2015. The European dairy industry towards2020—report extracts challenges, strategies and change. London, UK:Promar Intl. Ltd.
Rabobank. 2013. Sneller naar een integraal duurzamemelkveehouderijsector. Available from: https://www.rabobank.nl/images/m624_rabobank_su_melkveehouderij_def0711_2_29591343.pdf?ra_resize=yes&ra_width=1024&ra_height=768&ra_toolbar=yes&ra_locationbar=yes. Accessed 2015 March 10.
Rabobank. 2015. Melkveehouderij - Ontwikkelingen in de sector. Availablefrom: https://magazine.rabobank.nl/rn/melkveehouderij/#1. Accessed2015 March 10.
Roberts TA, Cordier JL, Gram L, Tompkin RB, Pitt JI, Gorris LGM,Swanson KMJ. 2005. Oil- and fat-based foods. In: Roberts TA, editor.Micro-organisms in foods 6. New York, USA: Kluwer Academic/PlenumPublishers. p 480–521.
Rohrbach BW, Draughon FA, Davidson PM, Oliver SP. 1992. Prevalence ofListeria-monocytogenes, Campylobacter-jejuni, Yersinia-enterocolitica, andSalmonella in bulk tank milk - risk-factors and risk of human exposure. JFood Prot 55:93–7.
Ross AIV, Griffiths MW, Mittal GS, Deeth HC. 2003. Combiningnonthermal technologies to control foodborne microorganisms. Int J FoodMicrobiol 89:125–38.
Ryan S, Gleeson D, Jordan K, Furey A, O’ Brien B. 2012. Evaluation oftrichloromethane formation from chlorine-based cleaning and disinfectionagents in cow’s milk. Int J Dairy Technol 65:498–502.
Rychen G, Jurjanz S, Toussaint H, Feidt C. 2008. Dairy ruminant exposureto persistent organic pollutants and excretion to milk. Animal 2:312–23.
Serrano SE, Braun J, Trasande L, Dills R, Sathyanarayana S. 2014. Phthalatesand diet: a review of the food monitoring and epidemiology data. EnvironHealth 13:43–57.
Sampimon OC, Barkema HW, Berends IMGA, Sol J, Lam TJGM. 2009.Prevalence and herd-level risk factors for intramammary infection withcoagulase-negative staphylococci in Dutch dairy herds. Vet Microbiol134:37–44.
Schatzmayr G, Streit E. 2013. Global occurrence of mycotoxins in the foodand feed chain: facts and figures. World Mycotoxin J 6:213–22.
Schoder D, Strauß A, Szakmary-Brandle K, Wagner M. 2015. How safe isEuropean Internet cheese? A purchase and microbiological investigation.Food Control 54:225–30.
Shokier KM, Aboelhadid SM, Waleed MA. 2013. Efficacy of fiveanthelmintics against a natural Fasciola species infection in cattle. BJBAS2:41–5.
Simoes M, Simoes LC, Vieira MJ. 2010. A review of current and emergentbiofilm control strategies. LWT - Food Sci Technol 43:573–83.
Sobrino-Lopez A, Martın-Belloso O. 2009. Review: potential ofhigh-intensity pulsed electric field technology for milk processing. FoodEng Rev 2:17–27.
Spier SJ, Smith BP, Cullor JS, Olander HJ, Roden LD, Dilling GW. 1991.Persistent experimental Salmonella dublin intramammary infection in dairycows. J Vet Int Med 5:341–50.
Steinhauser G, Brandl A, Johnson TE. 2014. Comparison of the Chernobyland Fukushima nuclear accidents: a review of the environmental impacts.Sci Total Environ 470–471:800–17.
Straley BA, Donaldson SC, Hedge NV, Sawant AA, Srinivasan V, Oliver SP,Jayarao BM. 2006. Public health significance of antimicrobial-resistantGram-negative bacteria in raw bulk tank milk. Foodborne Pathog Dis3:222–33.
Strydom A, Cawthorn DM, Cameron M, Witthuhn RC. 2012. Species ofCronobacter—a review of recent advances in the genus and their significancein infant formula milk. Int Dairy J 27:3–12.
ter Mors R, de Wit J. 2011. Gezondheidsvoordelen van schapenmelk.Driebergen, The Netherlands: Louis Bolk Instituut. (2011-011 LbD).
To H, Htwe KK, Kako N, Kim HJ, Yamaguchi T, Fukushi H, Hirai K.1998. Prevalence of Coxiella burnetii infection in dairy cattle withreproductive disorders. J Vet Med Sci 60:859–61.
Tropea F. 2015. The future of the EU dairy sector after the end of milkquotas. European Parliamentary Research Service. (PE 564.359).
US EPA. 2015. Commonly encountered radionuclides. Available from:http://www.epa.gov/radiation/radionuclides/. Accessed 2015 March 10.
van Asselt ED, Meuwissen MPM, van Asseldonk MAPM, Sterrenburg P,Mengelers MJB, van der Fels-Klerx HJ. 2011. Approach for a pro-activeemerging risk system on biofuel by-products in feed. Food Pol 36:421–9.
van Asselt ED, Sterrenburg P, Noordam MY, van der Fels-Klerx HJ. 2012.Overview of available methods for risk based control within the EuropeanUnion. Trends Food Sci Tech 23:51–8.
van Asselt ED, Kowalczyk J, van Eijkeren JCH, Zeilmaker MJ, Ehlers S,Furst P, Lahrssen-Wiederhol M, van der Fels-Klerx HJ. 2013. Transfer ofperfluorooctane sulfonic acid (PFOS) from contaminated feed to dairy milk.Food Chem 141:1489–95.
van Boeckel TP, Brower C, Gilbert M, Grenfell BT, Levin SA, RobinsonTP, Teillant A, Laxminarayan R. 2015. Global trends in antimicrobial use infood animals. Proc Natl Acad Sci 112:5649–54.
van Kessel JS, Karns JS, Gorski L, McCluskey BJ, Perdue ML. 2004.Prevalence of salmonellae, Listeria monocytogenes, and fecal coliforms in bulktank milk on US dairies. J Dairy Sci 87:2822–30.
van Schothorst M, Kleiss T. 1994. HACCP in the dairy industry. FoodControl 5:162–6.
Verraes C, Claeys W, Cardoen S, Daube G, de Zutter L, Imberechts H,Dierick K, Herman L. 2014. A review of the microbiological hazards of rawmilk from animal species other than cows. Int Dairy J 39:121–30.
Vishnuraj MR, Kandeepan G, Rao KH, Chand S, Kumbhar V,Gonzalez-Redondo P. 2016. Occurrence, public health hazards anddetection methods of antibiotic residues in foods of animal origin: acomprehensive review. Cogent Food Agric 2:1235458.
Vissers MM, te Giffel MC, Driehuis F, de Jong P, Lankveld JM. 2007a.Predictive modeling of Bacillus cereus spores in farm tank milk duringgrazing and housing periods. J Dairy Sci 90:281–92.
Vissers MMM, Driehuis F, te Giffel MC, de Jong P, Lankveld JMG. 2007b.Short communication: quantification of the transmission of microorganismsto milk via dirt attached to the exterior of teats. J Dairy Sci 90:3579–82.
Wemmenhove E, Stampelou I, van Hooijdonk ACM, Zwietering MH,Wells-Bennik MHJ. 2013. Fate of Listeria monocytogenes in Goudamicrocheese: no growth, and substantial inactivation after extended ripeningtimes. Int Dairy J 32:192–8.
White DG, Harmon RJ, Matos JE, Langlois BE. 1989. Isolation andidentification of coagulase-negative Staphylococcus species from bovine bodysites and streak canals of nulliparous heifers. J Dairy Sci 72:1886–92.
WHO. 2008. Experts set tolerable level for melamine intake. Geneva,Switzerland: WHO.
WHO and FAO. 2007. Safe preparation, storage and handling of powderedinfant formula: guidelines. Geneva, Switzerland: WHO.
WHO/FAO. 2004. Risk assessment of Listeria monocytogenes in ready-to-eat foods : Technical report. Microbiological risk assessment series no. 5:1–307.
Wilbey RA. 2005. Microbiology of cream and butter. In: Robinson RK,Wilbey RA, editors. Dairy microbiology handbook. Hoboken, USA: JohnWiley & Sons, Inc. p 123–74.
![NOT REPORTABLE IN THE HIGH COURT OF SOUTH AFRICA …[1] The plaintiff herein is a partnership existing between Kevin Glynn Roux and Elizabeth Johanna Van Nierop who trade as estate](https://static.documents.pub/doc/80x56/613092fa1ecc515869442f29/not-reportable-in-the-high-court-of-south-africa-1-the-plaintiff-herein-is-a-partnership.jpg)
