**7.** *Salmonella* **in raw milk and milk products**

Nowadays, according to the EC regulations in force, we can obtain milk for human consumption from all mammalian species, without exception, provided that the animals are reared and milked when they are in good health and nutrition conditions (EC Regulation 853/04). In developed countries the milk that humans use as food is almost always cow's milk; much more seldom we also consume sheep's, goat's and buffalo milk. In poor developing countries, in addition to cow's milk, buffalo, sheep's and goat's milk is also regularly drunk. In different geographical areas and with varying eating habits, along with "traditional" dairy species, man also employs various other animal species (camels, dromedaries, horses, reindeer, etc.) as a source of milk. Thanks to its chemical composition, milk is an almost ideal food for humans and for this reason it is part of the daily ration of most of the world's population. Over time the different peoples on Earth have developed a remarkable range of food products that use milk as raw material: in the world today there are approximately 1,600 different types of cheeses and over 100 different types of fermented dairy products. From raw milk many kind of milk products are obtained, such as pasteurized or UHT milk, cheese, fermented milks and probiotics, ice cream, butter, ricotta and milk drinks or whey. This wide range of food is obtained by subjecting raw milk to one or more technical processes that change the components of milk and its rheological properties to a greater or lesser extent. These "treatments" may be the addition of salt and the removal of water (*seasoning*) or the addition of natural enzymes and/or milk ferments that trigger these complex biochemical processes that we call *ageing* of cheese or fermented milks. *Salmonella*, as well as other pathogenic agents of foodborne disease, can contaminate raw milk: (1) directly inside the mammary gland (very rare event); (2) during milking, because the bacteria are often present in the faeces of milk animals and on their coat; (3) after milking, because salmonellae can contaminate work surfaces with which the raw milk comes into contact; (4) in subsequent phases, still due to the presence of *Salmonella* on work surfaces and/or cross contamination. The fate of salmonellae in milk and milk products widely depends on the antimicrobial effects the different transformation processes may have on the bacteria, as bactericidal effect or more simply bacteriostatic effect. This explains why in developed countries cases of human salmonellosis caused by the consumption of dairy products and milk are much rarer than those caused by the consumption of fresh meat or fish products (Jayarao & Henning, 2001). Raw milk, of course, represents an exception: in recent years it has caused a number of fairly many outbreaks of human salmonellosis (Newkirk *et al*., 2011). We must not forget that in recent years in many European states the consumption of raw milk purchased directly from the dairy by means of automatic vending machines has greatly increased. The following study will, therefore, give information about the possible presence of *Salmonella* in raw milk and different products that are derived from it, focusing above all on the possibility for *Salmonella* to multiply in different dairy products. According to the epidemiology data of foodborne illnesses provided each year by the EFSA, milk and milk products are not by far the greatest sources of danger for consumers. Like in previous years, in 2009 too there were few cases of *Salmonella* detected in cow's milk. Only three EU Member States conducted specific tests on raw milk sold in vending machines: Austria (71 samples tested), Germany (173 samples) and Hungary (50 samples). *Salmonella* was never detected in any of these samples. On the other hand, as regards pasteurized or UHT milk, seven states reported data: Austria (30 samples), Bulgaria (30 samples), the

Nowadays, according to the EC regulations in force, we can obtain milk for human consumption from all mammalian species, without exception, provided that the animals are reared and milked when they are in good health and nutrition conditions (EC Regulation 853/04). In developed countries the milk that humans use as food is almost always cow's milk; much more seldom we also consume sheep's, goat's and buffalo milk. In poor developing countries, in addition to cow's milk, buffalo, sheep's and goat's milk is also regularly drunk. In different geographical areas and with varying eating habits, along with "traditional" dairy species, man also employs various other animal species (camels, dromedaries, horses, reindeer, etc.) as a source of milk. Thanks to its chemical composition, milk is an almost ideal food for humans and for this reason it is part of the daily ration of most of the world's population. Over time the different peoples on Earth have developed a remarkable range of food products that use milk as raw material: in the world today there are approximately 1,600 different types of cheeses and over 100 different types of fermented dairy products. From raw milk many kind of milk products are obtained, such as pasteurized or UHT milk, cheese, fermented milks and probiotics, ice cream, butter, ricotta and milk drinks or whey. This wide range of food is obtained by subjecting raw milk to one or more technical processes that change the components of milk and its rheological properties to a greater or lesser extent. These "treatments" may be the addition of salt and the removal of water (*seasoning*) or the addition of natural enzymes and/or milk ferments that trigger these complex biochemical processes that we call *ageing* of cheese or fermented milks. *Salmonella*, as well as other pathogenic agents of foodborne disease, can contaminate raw milk: (1) directly inside the mammary gland (very rare event); (2) during milking, because the bacteria are often present in the faeces of milk animals and on their coat; (3) after milking, because salmonellae can contaminate work surfaces with which the raw milk comes into contact; (4) in subsequent phases, still due to the presence of *Salmonella* on work surfaces and/or cross contamination. The fate of salmonellae in milk and milk products widely depends on the antimicrobial effects the different transformation processes may have on the bacteria, as bactericidal effect or more simply bacteriostatic effect. This explains why in developed countries cases of human salmonellosis caused by the consumption of dairy products and milk are much rarer than those caused by the consumption of fresh meat or fish products (Jayarao & Henning, 2001). Raw milk, of course, represents an exception: in recent years it has caused a number of fairly many outbreaks of human salmonellosis (Newkirk *et al*., 2011). We must not forget that in recent years in many European states the consumption of raw milk purchased directly from the dairy by means of automatic vending machines has greatly increased. The following study will, therefore, give information about the possible presence of *Salmonella* in raw milk and different products that are derived from it, focusing above all on the possibility for *Salmonella* to multiply in different dairy products. According to the epidemiology data of foodborne illnesses provided each year by the EFSA, milk and milk products are not by far the greatest sources of danger for consumers. Like in previous years, in 2009 too there were few cases of *Salmonella* detected in cow's milk. Only three EU Member States conducted specific tests on raw milk sold in vending machines: Austria (71 samples tested), Germany (173 samples) and Hungary (50 samples). *Salmonella* was never detected in any of these samples. On the other hand, as regards pasteurized or UHT milk, seven states reported data: Austria (30 samples), Bulgaria (30 samples), the

**7.** *Salmonella* **in raw milk and milk products** 

Czech Republic (135 samples), Germany (980 samples), Greece (26 samples), Hungary (85 samples) and Romania (57 samples). Again, none of the samples tested contained *Salmonella*. Italy reported that out of a total of 928 samples of cow's milk, 3 were positive for *Salmonella* and that 5 samples out of a total of 5,799 samples of milk from "other unspecified species" also tested positive for the pathogen. 11 member countries supplied EFSA with results of their investigations on cheeses, for a total of 23,023 samples analyzed. In the great majority of cases, the cheese samples proved to be negative for *Salmonella*, with the exception of Spain (4 positive samples out of a total of 424 samples tested), Portugal (2 positive samples out of 181 analyzed) and Italy (2 positive out of a total of 1,879 samples tested). As far as we know from the EFSA report, all the cheeses tested positive were semi-hard cheeses, and only semimature, and made from raw or thermised milk (i.e. heated to a temperature between 45 °C and 54 °C, no more). As far as butter is concerned, 7 member states communicated the results of their inspections; no case revealed the presence of *Salmonella*. Besides cheese, the only other product derived from milk which pointed out the presence of *Salmonella* was ice cream. Spain, Hungary and Germany reported the presence of the bacterium respectively in 13 samples out of 305 samples analyzed, 1 out of 140 and 1 in 2,626 samples, always taken in the production facilities. The presence of salmonellae in raw milk is widely documented in the literature, both in the collection tanks on the farms, and in the storage tanks in the food factories (Donaghy *et al*., 2004; Tondo *et al*., 2000). *Salmonella* may be present in raw milk ever since milking because the bacterium is present in the mammary gland, but this occurs very rarely. Mastitis due to *Salmonella* is a very rare condition in dairy cows, but it is reported. We know that different *Salmonella* serotypes can colonize the mammary gland and lead to the excretion, at the same time as the milk, of bacterial loads that can extend up to 3.3 log10 cfu/ml (Fontaine *et al*., 1980). Furthermore, *Salmonella* can pass from animal to animal at the time of milking, both through the milker's hands, and through polluted parts of the milking machines (Bergonier *et al*., 2003; Vautor *et al*., 2003; Zadoks *et al*., 2002; Zschöck *et al*., 2000). Much more often, however, salmonellae contaminate raw milk in the stages that follow the milking process, because the bacteria may be present on the various surfaces that come into contact with the milk being collected. In particular, *Salmonella* (such as *Listeria monocytogenes* and verotoxigenic strains of *E. coli*) can enter the milk through the traces of animal faeces in the environment (Van Kessel *et al*., 2004). This factor of pollution, in turn, is influenced by the prevalence among dairy cows of *Salmonella* healthy carriers, which can evacuate various loads of the pathogen more or less frequently. In this regard, it is estimated that the U.S. dairy cows can be healthy carriers of *Salmonella* in their faeces with a prevalence that ranges from a minimum of 2% to a maximum of 27.5% of the animals tested (Kabagambe *et al*., 2000; Losinger *et al*., 1995, Wells *et al*., 2001). What can be the prevalence of a batch of raw milk tested positive for *Salmonella* ever since the milking phase? In view of the data that we can gather from the literature, we can estimate that the batches of raw milk straight after milking can be positive for *Salmonella* from a minimum of 2.6% to a maximum of 25.3% (Jayarao & Henning, 2001; Murinda *et al*., 2002; Zhao *et al*. 2002). Compared to other pathogenic microorganisms such as *L. monocytogenes*, salmonellae are not very resistant in the outside, so it is not very frequent for the work surfaces in the production plants to transfer salmonellae to the product. Nevertheless, in theory, *Salmonella* can survive for long on any work surface and then pollute the cheese curd which is meant to become cheese. This justifies the episodes of foodborne infection caused by processed dairy products, such as

Food as Cause of Human Salmonellosis 65

 the type of raw material: cheese made with raw milk may contain salmonellae still alive and vital, while it is hard for those made with pasteurized milk to still shelter the

 the microbiological quality of milk used to make cheese. Cheeses made with raw milk are not necessarily infected with *Salmonella*, if good hygiene conditions are maintained

As with many other types of foodstuffs, salmonellae can contaminate cheese coming from: raw materials used in production, most likely from raw milk and less likely from other

work surfaces in the cheese factories, including the air that circulates in various

packaging materials in which is wrapped the finished product ready for sale (Temelli *et* 

As regards in particular brines used to salt the cheese, Ingham *et al*. (2000) conducted experimental inoculation tests with *Salmonella* ser. Typhimurium to test the viability of the pathogen in the cheeses' brines. The researchers experimentally inoculated two cultures with *S*. Typhimurium and *E. coli* O157, mixed together, in three different brines containing 23% salt, with the addition of 2% of flour. The brines were then stored at 8 °C and 15 °C for 28 days. The same cultures were also inoculated into brines offered for sale, and then stored at 4 °C and 13 °C for 35 days. The load of the two pathogens immediately underwent a gradual decline during storage, but it is significant that the reduction was less noticeable in the brines stored at 4 °C compared to the ones stored at 13 ° or 15 °C. This study shows that *Salmonella* may still survive in saline solutions used for salting cheese, although with very

Compared to other pathogens such as *L. monocytogenes* and *Staphylococcus aureus*, *Salmonella* is much less often blamed as a source of illness due to the consumption of cheese. As a result, we do not have precise data as to the actual prevalence of *Salmonella* in cheese. We can, however, find some data on the persistence of salmonellae in cheese sold in retail food stores. The pathogen was detected in Turkey in various kind of cheese produced mainly in an artisanal manner with raw cow's, ewe's and/or goat's milk (Colak *et al*., 2000; Hayaloglu & Kirbag, 2007; Tekinşen & Özdemir, 2006), always in very low prevalence of the samples analyzed. On the other hand, we also have data documenting how salmonellae, potentially present in raw milk and/or in environments where milk and cheese are produced, are not so detectable in the dairy products offered for sale. For example, in Spain Cabedo *et al*. (2008) conducted a large study to test the microbiological quality of the cheeses of their land: they never detected *Salmonella* in any of the samples they analysed. In Britain, two studies conducted by Little *et al*. (2008) first in 2004 and then in 2005, showed that a total of 4,437 samples of various types of cheeses (fresh, semi-mature and mature, made with raw or

during the milking process and the ensuing manufacturing process.

ingredients such as lactic acid starter and salt, salt solutions (brine) used for salting certain products,

pasteurized milk) never showed the presence of *Salmonella*.

pathogen, unless the contamination occurred after the pasteurization process, the duration and type of ageing: in cheeses which mature for a short time, *Salmonella* is more likely to survive, because the maturing biochemical processes that have a good antimicrobial effect against pathogen are not yet established in the substrate. In cheeses that mature for over 60 days, on the contrary, the characteristics of the substrate that are obtained as a result of aging make the product unfit for the reproduction and survival

of salmonella,

environments,

*al*., 2006).

small loads.

milk powder and cheeses made with pasteurized milk. **Fermented milks** can be divided into two kinds: (i) acid, if their production is based on homolactic fermentation, (ii) acidalcoholic, if the starter strains used for fermentation are of the heterofermentative type. In case (i) the product will only be acid, while in case (ii) besides the presence of acid there is a fair amount of ethyl alcohol which enhances the food's antimicrobial effect against *Salmonella*. Their production process usually starts from pasteurized milk. Furthermore, milk is caused to coagulate by using acid, by adding selected milk ferments that produce large amounts of lactic acid or other organic acids and possibly ethyl alcohol, with a drastic drop in the substrate's pH which makes the casein coagulate. The presence of high loads of lactic acid bacteria, coupled with low pH levels (4.0 to 4.1 on average) and Aw mean that yogurt and other fermented milk products are a very unfit food matrix for allowing the growth and even the survival of *Salmonella*.

**Cheese** is among the foods which are less likely to cause salmonellosis in humans due to their production process (Little *et al*., 2008). Nevertheless, in 2008 it was responsible for 0.4% of all episodes of illness reported in the EU (EFSA, 2010). In addition, several cases of salmonellosis caused by the consumption of cheese contaminated with *Salmonella enterica* are reported in the bibliography. The problem is that despite the fact that the production process poses several obstacles to the survival and multiplication of salmonellae, we eat cheese without further heat processing. Moreover, cheese often does not carry pathogenic microorganisms in its inside, but rather on its surface. This may result in the transfer of *Salmonella* and other pathogens to domestic working environments, thus favouring cross contamination, which in turn enables the outbreak of foodborne illnesses (Kousta *et al*., 2010). The bibliography gives at least a dozen episodes of salmonellosis caused by the consumption of cheeses made not only with raw milk, but also with pasteurized milk. This means that in many cases the milk used to produce cheese is contaminated with *Salmonella* "after" its pasteurization, since this is largely able to inactivate very high loads of the bacteria. Nowadays, HTST pasteurization is often used in the dairy industry (at least 72 °C for at least 15 seconds) and it can produce a drop of about 6 LOG-degrees in the original load of *Salmonella*, as demonstrated by accurate experimental investigations (D'Aoust *et al*., 1988; D'Aoust *et al*., 1987; Farber *et al*., 1988). In particular, these studies showed that *Salmonella* can still be detected in milk heated up to 67.5 °C for 15 seconds, but not at higher temperatures. We need not forget, though, that *Salmonella*, just like *Listeria monocytogenes*, can penetrate into the milk somatic cells that can provide it with a slight protection against the effects of heat. It is not, therefore, possible to exclude *a priori* that in normally pasteurized milk it may still be possible to detect some salmonellae which survived the treatment itself, if it was not carried out at temperatures above 72 °C. In the past decades, salmonellae have caused a series of outbreaks of illness caused by the consumption of various types of cheese. As mentioned before, we can find several references in the literature to outbreaks of salmonellosis caused by foods that contain very low numbers of *Salmonella*. According to D'Aoust (1985) and Ratnam & March (1986), the literature documents cases of salmonellosis caused by Cheddar cheese in which the estimated infectious load proved to be under 10 cfu of *Salmonella*/g of food.

From the data we possess, we can therefore sum up that *Salmonella* may still be present in cheeses for human consumption, but with a prevalence which varies widely depending on several factors:

milk powder and cheeses made with pasteurized milk. **Fermented milks** can be divided into two kinds: (i) acid, if their production is based on homolactic fermentation, (ii) acidalcoholic, if the starter strains used for fermentation are of the heterofermentative type. In case (i) the product will only be acid, while in case (ii) besides the presence of acid there is a fair amount of ethyl alcohol which enhances the food's antimicrobial effect against *Salmonella*. Their production process usually starts from pasteurized milk. Furthermore, milk is caused to coagulate by using acid, by adding selected milk ferments that produce large amounts of lactic acid or other organic acids and possibly ethyl alcohol, with a drastic drop in the substrate's pH which makes the casein coagulate. The presence of high loads of lactic acid bacteria, coupled with low pH levels (4.0 to 4.1 on average) and Aw mean that yogurt and other fermented milk products are a very unfit food matrix for allowing the

**Cheese** is among the foods which are less likely to cause salmonellosis in humans due to their production process (Little *et al*., 2008). Nevertheless, in 2008 it was responsible for 0.4% of all episodes of illness reported in the EU (EFSA, 2010). In addition, several cases of salmonellosis caused by the consumption of cheese contaminated with *Salmonella enterica* are reported in the bibliography. The problem is that despite the fact that the production process poses several obstacles to the survival and multiplication of salmonellae, we eat cheese without further heat processing. Moreover, cheese often does not carry pathogenic microorganisms in its inside, but rather on its surface. This may result in the transfer of *Salmonella* and other pathogens to domestic working environments, thus favouring cross contamination, which in turn enables the outbreak of foodborne illnesses (Kousta *et al*., 2010). The bibliography gives at least a dozen episodes of salmonellosis caused by the consumption of cheeses made not only with raw milk, but also with pasteurized milk. This means that in many cases the milk used to produce cheese is contaminated with *Salmonella* "after" its pasteurization, since this is largely able to inactivate very high loads of the bacteria. Nowadays, HTST pasteurization is often used in the dairy industry (at least 72 °C for at least 15 seconds) and it can produce a drop of about 6 LOG-degrees in the original load of *Salmonella*, as demonstrated by accurate experimental investigations (D'Aoust *et al*., 1988; D'Aoust *et al*., 1987; Farber *et al*., 1988). In particular, these studies showed that *Salmonella* can still be detected in milk heated up to 67.5 °C for 15 seconds, but not at higher temperatures. We need not forget, though, that *Salmonella*, just like *Listeria monocytogenes*, can penetrate into the milk somatic cells that can provide it with a slight protection against the effects of heat. It is not, therefore, possible to exclude *a priori* that in normally pasteurized milk it may still be possible to detect some salmonellae which survived the treatment itself, if it was not carried out at temperatures above 72 °C. In the past decades, salmonellae have caused a series of outbreaks of illness caused by the consumption of various types of cheese. As mentioned before, we can find several references in the literature to outbreaks of salmonellosis caused by foods that contain very low numbers of *Salmonella*. According to D'Aoust (1985) and Ratnam & March (1986), the literature documents cases of salmonellosis caused by Cheddar cheese in which the estimated infectious load proved to be

From the data we possess, we can therefore sum up that *Salmonella* may still be present in cheeses for human consumption, but with a prevalence which varies widely depending on

growth and even the survival of *Salmonella*.

under 10 cfu of *Salmonella*/g of food.

several factors:


As regards in particular brines used to salt the cheese, Ingham *et al*. (2000) conducted experimental inoculation tests with *Salmonella* ser. Typhimurium to test the viability of the pathogen in the cheeses' brines. The researchers experimentally inoculated two cultures with *S*. Typhimurium and *E. coli* O157, mixed together, in three different brines containing 23% salt, with the addition of 2% of flour. The brines were then stored at 8 °C and 15 °C for 28 days. The same cultures were also inoculated into brines offered for sale, and then stored at 4 °C and 13 °C for 35 days. The load of the two pathogens immediately underwent a gradual decline during storage, but it is significant that the reduction was less noticeable in the brines stored at 4 °C compared to the ones stored at 13 ° or 15 °C. This study shows that *Salmonella* may still survive in saline solutions used for salting cheese, although with very small loads.

Compared to other pathogens such as *L. monocytogenes* and *Staphylococcus aureus*, *Salmonella* is much less often blamed as a source of illness due to the consumption of cheese. As a result, we do not have precise data as to the actual prevalence of *Salmonella* in cheese. We can, however, find some data on the persistence of salmonellae in cheese sold in retail food stores. The pathogen was detected in Turkey in various kind of cheese produced mainly in an artisanal manner with raw cow's, ewe's and/or goat's milk (Colak *et al*., 2000; Hayaloglu & Kirbag, 2007; Tekinşen & Özdemir, 2006), always in very low prevalence of the samples analyzed. On the other hand, we also have data documenting how salmonellae, potentially present in raw milk and/or in environments where milk and cheese are produced, are not so detectable in the dairy products offered for sale. For example, in Spain Cabedo *et al*. (2008) conducted a large study to test the microbiological quality of the cheeses of their land: they never detected *Salmonella* in any of the samples they analysed. In Britain, two studies conducted by Little *et al*. (2008) first in 2004 and then in 2005, showed that a total of 4,437 samples of various types of cheeses (fresh, semi-mature and mature, made with raw or pasteurized milk) never showed the presence of *Salmonella*.

Food as Cause of Human Salmonellosis 67

those who produce ice cream from raw milk. In recent years, in fact, this habit seems to have come back into fashion, under the pressure from consumers who take great pleasure in consuming food products from raw materials treated as little as possible. Regarding ice cream too, the EU has set specific criteria for *Salmonella*, which must be "absent" in 125 g of product. This law does not apply to ice creams "whose manufacturing process or composition properties eliminate the risk of *Salmonella*" as required by Regulation 2073/05.

All this makes it difficult to control and prevent these toxi-infections; as a result, it is necessary for epidemiologists, clinicians and microbiologists as well as veterinarians to collaborate in order to launch an integrated approach to solve the problem. In order to prevent the occurrence of salmonellosis, it is therefore essential to know which animals and/or which foods most frequently carry the pathogens which have led to sporadic cases or episodes of disease in humans. Epidemiological data should then be given special attention and consideration by meat producers and in general by anyone whose role it is to carry out investigations on food, as they can provide useful information regarding changes

Berger C. N., Sodha S. V., Shaw R. K., Griffin P M., Pink D., Hand P.& Frankel G. (2010).

Bergonier D., De Cremoux R., Rupp R., Lagriffoul G.& Berthelot X. (2003). Mastitis of dairy

Bozzo G. (2008). Uova e ovoprodotti, In: *Igiene e tecnologie degli alimenti di origine animale*,

Cabedo L., Barrot L.P.& Teixidó A.T. (2008). Prevalence of *Listeria monocytogenes* and

Cantoni C. & Bersani C. (2010). *E. coli* O157:H7, non O157:H e *Salmonella enterica*. *Qualità e* 

Cantoni C. & Ripamonti B. (1998). La presenza di *Salmonella* negli alimenti e sua

Castillo N.A., de Moreno de LeBlanc A., Maldonado Galdeano C. & Perdigón G. (2011).

mechanisms involved, *Food Research International*, Vol. x, No. x, pp. xxx-xxx. Catellani P., Giaccone V. & Novelli E. (2005). Studio della flora microbica nei vegetali di IV

Colak H., Hampikyan H., Bingol E.B. & Ulusoy B. (2007). Prevalence of *L. monocytogenes* and *Salmonella* spp. in Tulum cheese, *Food Control*, Vol. 18, No.5, pp. 576-579 D'Aoust J.Y. (1985). Infective dose of *Salmonella* typhimurium in Cheddar cheese, *American* 

Fresh fruit and vegetables as vehicles for the transmission of human pathogens.

Colavita G., pp. 321-328, Le Point Vétérinaire Italie srl, ISBN 978-88-95995-47-2,

*Salmonella* in ready-to-eat Food in Catalonia, Spain, *Journal of Food Protection*, Vol.

Probiotics: an alternative strategy for combating salmonellosis. Immune

gamma. Nota preliminare. *Proceedings of S.I.S.Vet*, Viareggio, Italy, 21-24

or additions to be made to the eradication plans against *Salmonella*.

*Environmental Microbiology,* Vol. 12, No.9, pp. 2385-2397

small ruminants, *Veterinary Research*, Vol. 34, No.5, pp. 689-716

*Sicurezza Alimentare*, Vol. 3, No.14, pp. 47-53, ISSN 2038-9760

patogenicità. *Tecnologie alimentari*, Vol. 9, No.5, pp. 76-83

*Journal of Epidemiology*, Vol. 122, No.4, pp. 717-720

**8. Conclusion** 

**9. References** 

Milano, Italy

71, No. 4, pp. 855-859

September, 2005

**Butter** is produced by the mechanical churning of the cream obtained after centrifugation of cheese whey. It can be *sweet* if the cream is used as it is, or *ripened* if it comes from cream that was first matured with the addition of starter enzymes. In most cases, the raw material for butter is subjected to pasteurization in butter before being processed, but in some cases butter is obtained directly from the cream of raw, unpasteurized milk. It is clear that in this second case *Salmonella* may be present in the butter from the start of the making process because the raw material itself was contaminated. In the case of butter made from pasteurized cream, however, a possible contamination with *Salmonella* cannot be excluded, because the pathogen could infect the finished product through a secondary contamination. In the past decades, in fact, several episodes of human salmonellosis caused by butter contaminated with *Salmonella* occurred, but over the years these episodes have registered a sharp decline, due to the fact that producers dedicate more attention to production hygiene and to the fact that butter is now rarely made with unpasteurized cream. The EU has established with EC Regulation 2073/05 that "cheese, butter and cream made from raw milk or milk subjected to heat treatment at sub-pasteurization temperatures" should not contain even one living cell of *Salmonella* in 125 g (25 g in 5 units of the sample) of product throughout its shelf life. **Dried milk products** as a rule, these foods are products obtained after pasteurized milk is nebulized in towers where a very dry and hot air current circulates, but on the market you can find lyophilised products, i.e. put through the cold-removal of water, not involving the use of high temperatures. The sanitary characteristics of milk powders, therefore, is determined by: (i) the microbiological quality of the raw material, (ii) the conditions of the production process (with or without heat treatment), (iii) the possibility of the dehydrated/lyophilised product to be contaminated with salmonellae after its processing. Salmonellae are sensitive to normal temperatures applied in the production process of dried milk products, so it is logical to expect that such products are rarely at risk of containing *Salmonella*, unless they are contaminated after this process, during packaging or storage. In these cases, dried milk products may be a risk to human health, since salmonellae can survive for months in substrates with low water content, such as bone meal and powdered foods. The possible dangers of these products is also enhanced by the fact that such foods are usually meant for very young children, much more sensitive than adults to even minor loads of *Salmonella*. For this reason, the EU has established by law (EC Regulation 2073/05) that "powdered milk and powdered whey" should not contain even one living cell of *Salmonella* in 125 g of product throughout its shelf life. **Ice cream** is a complex food made of various ingredients, including eggs and milk, where water crystallizes, forming a homogeneous creamy mass, thanks to the high amount of fat. As such, also ice cream can be contaminated with *Salmonella*, if it is contained in the raw milk or appears in the manufacturing process. Over the past decades, in fact, many outbreaks of salmonellosis caused by the consumption of ice cream have been documented, but it was not always possible to establish with certainty whether the pollution came from the raw milk or from the eggs, which are also used raw. For several years now, the use of pasteurized milk and eggs has become a habit for producing ice cream, so the risk of *Salmonella* contamination in these products has been greatly reduced. But we must remember that ice cream, due to its almost always neutral or slightly acidic pH levels and to its high amount of free water (Aw), can be an excellent substrate for the survival and growth of *Salmonella*, if the latter managed to infect it. The risks to public health may be greater for those who produce ice cream from raw milk. In recent years, in fact, this habit seems to have come back into fashion, under the pressure from consumers who take great pleasure in consuming food products from raw materials treated as little as possible. Regarding ice cream too, the EU has set specific criteria for *Salmonella*, which must be "absent" in 125 g of product. This law does not apply to ice creams "whose manufacturing process or composition properties eliminate the risk of *Salmonella*" as required by Regulation 2073/05.
