**4.** *Salmonella* **outbreaks**

Disease surveillance reports frequently identify poultry, meat and milk products as the main vehicles in salmonellosis outbreaks. However, in recent years foodborne illness outbreaks have been increasingly associated with greater consumption of fresh fruits and vegetables (CDC, 2009). *Salmonella* is responsible for frequent foodborne illness outbreaks in the

The Role of Foods in *Salmonella* Infections 29

transmission can occur if hands contaminated with infected fecal matter come in contact with food which is then consumed without adequate cooking, often after an intervening period in which microbial growth occurs. Exactly this chain of events led to a major outbreak affecting an international airline in 1984. A total of 631 passengers were infected after eating food containing an aspic glaze prepared by a food service worker who returned to work after a bout of salmonellosis but was still excreting *Salmonella* Enteritidis PT4. The serotype Typhimurium has participated in most recent outbreaks, although it is likely that this serotype's involvement in salmonellosis cases worldwide is far greater than reported. *Salmonella* surveillance sensitivity may vary widely between countries but it is still crucial to identifying trends and detecting outbreaks. Surveillance which includes serotyping is particularly useful for this purpose. Available data suggest that the incidence of *Salmonella*  infections has increased over the last twenty years, that new *Salmonella* serotypes often emerge in several countries at near the same time, and that multi-state or international outbreaks call for a coordinated response. In response, several national and international networks currently address the problem of emerging *Salmonella* infections. An important objective in preventing *Salmonella* outbreaks is improvement and enhancement of

Country Food vehicle Serotypes Number

United States Papaya Agona 99 CDC, 2011a

United States Turkey Burgers Hadar 12 CDC, 2011c United States Cantaloupe Panama 20 CDC, 2011d United States Alfalfa sprouts I 4,[5],12:i:- 140 CDC, 2011e Denmark Salami Typhimurium 20 Kuhn et al, 2011

spicy sprouts Enteritidis <sup>25</sup> CDC, 2011b

prepared salads Typhimurium <sup>179</sup> Boxall et al.,

mayonnaise Typhimurium <sup>87</sup> Jardine et al,

products Agona 163 Nicolay et al.,

school Enteritidis <sup>18</sup> Niehaus et al.,

of cases Reference

2011

2011

2011

2011

surveillance, including serotyping*.*

United States Alfalfa sprouts and

England Sandwiches and

South Africa Food served in a

Pre-cooked meat

Australia Raw egg

Ireland, United Kingdom (England, Wales, Northern Ireland, Scotland), France, Luxembourg, Sweden, Finland, Austria

developed world, and *Salmonella* outbreaks have been associated with different *Salmonella* serovars (Table 2). Over 2000 *Salmonella* serotypes are known, but only a small fraction of these are commonly associated with foodborne illness. Which serotypes cause illness is influenced by serotype geographical distribution and serovar or strain pathogenicity. In the US, *Salmonella* Typhimurium has been considered the principal causative agent of foodborne salmonellosis, but both *S*. Typhimurium and Enteritidis have been increasingly identified in foodborne salmonellosis since the 1980s (Table 2); the exact cause of the predominance of these *Salmonella* serotypes is not yet clearly understood.

Most developed countries have laboratory-based *Salmonella* infection surveillance programs, and many countries have systems for recording outbreaks and notification systems where clinicians submit data on patients with *Salmonella* infections to national public health institutions. Official *Salmonella* infection numbers are usually derived from laboratory-based surveillance in which clinical microbiology laboratories report positive findings and, in some countries, submit *Salmonella* isolates to national reference laboratories for serotyping and other characterization. These data are necessary for measuring trends over time and detecting outbreaks. However, official figures do not quantify the burden of illness, and degree of surveillance differs between countries. Moreover, reported incidence is a composite measure of several factors, including true *Salmonella* infection incidence, the health-care seeking behavior of patients with gastroenteritis, and the likelihood that the physician requests a stool culture. Furthermore, access to laboratories and microbiological methods varies widely, as does the precision of findings reported to public health authorities. Finally, comparisons between different geographical areas can be difficult because public health jurisdictions with a tradition of active case-searching as part of outbreak investigations or extensive testing of contacts of known patients or food-handlers are likely to report higher numbers of infections than jurisdictions with only passive surveillance. As a result, the precise incidence of *Salmonella* food poisoning in all countries is not known, since small outbreaks often remain unreported.

*Salmonella* spp. and *S.* Typhi infections are endemic in many developing countries. In Mexico, there were 709,278 salmonellosis cases and 228,206 typhoid fever cases reported from 2004-2009 (Secretaría de Salud, 2011). In addition, *S.* Gaminara and *S.* Montevideo have been associated with several cases of human illness in Mexico (Gutiérrez-Cogco et al., 2000). A certain proportion of salmonellosis and typhoid fever cases in Mexico may be associated with consumption of raw vegetables exposed to fecal contamination, probably due to the continued but limited practice of irrigating vegetable crops with untreated wastewater.

Centers for Disease Control and Prevention (CDC) data for the US indicate that over 40,000 salmonellosis cases occur annually, with about 500 resulting deaths. As is the case for staphylococcal gastroenteritis, the largest salmonellosis outbreaks typically occur at banquets or similar functions. However, the two largest recorded salmonellosis outbreaks occurred under rather unusual circumstances. The largest occurred in 1994 and involved over 224,000 cases in 41 states. The serovar was *S.* Enteritidis and the vehicle food was ice cream produced from milk transported in tanker trucks which had previously hauled liquid eggs. The second largest occurred in 1985 and involved nearly 200,000 cases. *S.*  Typhimurium was the etiological agent and the vehicle was 2% milk produced by a single dairy plant in Illinois. The third largest outbreak occurred in 1974 on the Navajo Indian Reservation, when 3,400 persons became ill with the *S*. Newport serovar. Human carriers are generally less important than animals in transmission of salmonellosis. Human

developed world, and *Salmonella* outbreaks have been associated with different *Salmonella* serovars (Table 2). Over 2000 *Salmonella* serotypes are known, but only a small fraction of these are commonly associated with foodborne illness. Which serotypes cause illness is influenced by serotype geographical distribution and serovar or strain pathogenicity. In the US, *Salmonella* Typhimurium has been considered the principal causative agent of foodborne salmonellosis, but both *S*. Typhimurium and Enteritidis have been increasingly identified in foodborne salmonellosis since the 1980s (Table 2); the exact cause of the predominance of

Most developed countries have laboratory-based *Salmonella* infection surveillance programs, and many countries have systems for recording outbreaks and notification systems where clinicians submit data on patients with *Salmonella* infections to national public health institutions. Official *Salmonella* infection numbers are usually derived from laboratory-based surveillance in which clinical microbiology laboratories report positive findings and, in some countries, submit *Salmonella* isolates to national reference laboratories for serotyping and other characterization. These data are necessary for measuring trends over time and detecting outbreaks. However, official figures do not quantify the burden of illness, and degree of surveillance differs between countries. Moreover, reported incidence is a composite measure of several factors, including true *Salmonella* infection incidence, the health-care seeking behavior of patients with gastroenteritis, and the likelihood that the physician requests a stool culture. Furthermore, access to laboratories and microbiological methods varies widely, as does the precision of findings reported to public health authorities. Finally, comparisons between different geographical areas can be difficult because public health jurisdictions with a tradition of active case-searching as part of outbreak investigations or extensive testing of contacts of known patients or food-handlers are likely to report higher numbers of infections than jurisdictions with only passive surveillance. As a result, the precise incidence of *Salmonella* food poisoning in all countries is

*Salmonella* spp. and *S.* Typhi infections are endemic in many developing countries. In Mexico, there were 709,278 salmonellosis cases and 228,206 typhoid fever cases reported from 2004-2009 (Secretaría de Salud, 2011). In addition, *S.* Gaminara and *S.* Montevideo have been associated with several cases of human illness in Mexico (Gutiérrez-Cogco et al., 2000). A certain proportion of salmonellosis and typhoid fever cases in Mexico may be associated with consumption of raw vegetables exposed to fecal contamination, probably due to the continued but limited practice of irrigating vegetable crops with untreated

Centers for Disease Control and Prevention (CDC) data for the US indicate that over 40,000 salmonellosis cases occur annually, with about 500 resulting deaths. As is the case for staphylococcal gastroenteritis, the largest salmonellosis outbreaks typically occur at banquets or similar functions. However, the two largest recorded salmonellosis outbreaks occurred under rather unusual circumstances. The largest occurred in 1994 and involved over 224,000 cases in 41 states. The serovar was *S.* Enteritidis and the vehicle food was ice cream produced from milk transported in tanker trucks which had previously hauled liquid eggs. The second largest occurred in 1985 and involved nearly 200,000 cases. *S.*  Typhimurium was the etiological agent and the vehicle was 2% milk produced by a single dairy plant in Illinois. The third largest outbreak occurred in 1974 on the Navajo Indian Reservation, when 3,400 persons became ill with the *S*. Newport serovar. Human carriers are generally less important than animals in transmission of salmonellosis. Human

these *Salmonella* serotypes is not yet clearly understood.

not known, since small outbreaks often remain unreported.

wastewater.

transmission can occur if hands contaminated with infected fecal matter come in contact with food which is then consumed without adequate cooking, often after an intervening period in which microbial growth occurs. Exactly this chain of events led to a major outbreak affecting an international airline in 1984. A total of 631 passengers were infected after eating food containing an aspic glaze prepared by a food service worker who returned to work after a bout of salmonellosis but was still excreting *Salmonella* Enteritidis PT4. The serotype Typhimurium has participated in most recent outbreaks, although it is likely that this serotype's involvement in salmonellosis cases worldwide is far greater than reported. *Salmonella* surveillance sensitivity may vary widely between countries but it is still crucial to identifying trends and detecting outbreaks. Surveillance which includes serotyping is particularly useful for this purpose. Available data suggest that the incidence of *Salmonella*  infections has increased over the last twenty years, that new *Salmonella* serotypes often emerge in several countries at near the same time, and that multi-state or international outbreaks call for a coordinated response. In response, several national and international networks currently address the problem of emerging *Salmonella* infections. An important objective in preventing *Salmonella* outbreaks is improvement and enhancement of surveillance, including serotyping*.*


The Role of Foods in *Salmonella* Infections 31

Australia Eggs Typhimurium <sup>19</sup> Slinko et al.,

Mauritius Marlin mousse Typhimurium 53 Issack et al.,

Pakistan Drinking water *S. typhi* <sup>300</sup> Farooqui et al.,

France Goat's cheese Muenster 25 Van Cauteren

United States Frozen Pot Pies I 4,5,12:i:-\* 401 CDC, 2008b United States Fruit salad Litchfield 30 CDC, 2008c

Wales Fresh basil Senftenberg 32 Pezzoli et al.,

Norway Rucola lettuce Thompson 21 Nygård et al.,

Bulgaria Minced meat Typhimurium 22 Pekova et al.,

Switzerland Soft cheese Stanley 82 Pastore et al.,

Denmark Pork products Typhimurium 1,054 Ethelberg et al.,

Japan Snapping turtle Typhimurium 4 Fukushima

Ireland Meat products Agona 119 O'Flanagan

Table 2. Recent reported *Salmonella* outbreaks, including country (ies) affected, food vehicle

Australia Chocolate mousse Typhimurium 8

Unpasteurized Mexican-style aged

cheese

United States Jalapeño peppers Saintpaul at least

pork products Typhimurium 41 Bruun et al.,

raw milk Montevideo 23 Dominguez

pesto Anatum At least 4 Pakalniskiene

from raw milk Typhimurium 224 Van Duynhoven

of cases Reference

2009

2009

2009

2009

et al., 2009

et al., 2009

et al., 2009

et al., 2009

Witteveen et al.,

Roberts-

2009

2008

2008

2008

2008

2008

et al., 2008

et al., 2008

1,442 CDC, 2008a

Newport 85 CDC, 2008d

Country Food vehicle Serotypes Number

Pork meat and

France Cheese made from

Denmark Pasta salad with

Netherlands Hard cheese made

Denmark, Norway and Sweden

United States

England and

and serovar.



Japan Boxed lunches Braenderup <sup>176</sup> Mizoguchi

England Multiples foods Enteritidis <sup>63</sup> Janmohamed

frozen entrée Chester <sup>44</sup> CDC, 2010b

fruit pulp Typhi <sup>9</sup> CDC, 2010c

chicken meat Montevideo 58 Patel et al, 2010

raw egg Typhimurium <sup>20</sup> Reynolds et al.,

Montevideo 272 CDC, 2010e

Typhimurium <sup>9</sup> CDC, 2010f

Typhimurium 23 Whelan et al.,

and Saintpaul <sup>152</sup> Jain et al., 2009

Typhimurium <sup>69</sup> Sotir et al., 2009

Enteritidis 8 Much et al., 2009

United States Shell Eggs Enteritidis 1,939 CDC, 2010a

United States Alfalfa Sprout Newport 44 CDC, 2010d

Netherlands Fresh fruit juice Panama 33 Noël et al., 2010 France Dried pork sausage 4,12:i:- 90 Bone et al., 2010 China Water *S.* Paratyphi A 267 Yang et al., 2010

Kingdom Raw bean sprouts Bareilly <sup>231</sup> Cleary et al.,

New Zealand Watermelon Typhimurium 15 McCallum et al.,

United States Alfalfa Sprouts Saintpaul 228 CDC, 2009a United States Peanut butter Typhimurium 529 CDC, 2009b

Typhimurium

Wandsworth,

Australia Eggs Typhimurium 22 Dyda et al., 2009

Australia Papaya Litchfield 26 Gibbs et al., 2009

of cases Reference

et al., 2011

et al., 2011

2010

2010

2010

2010

2010

Rodriguez-Urrego et al.,

Country Food vehicle Serotypes Number

United States Cheesy chicken rice

Red and Black Pepper/Italian-Style Meats

United States Potato salad Schwarzengrund,

United States Frozen mamey

United States Cilantro and

Raw or

products

Australia Dessert containing

United States Unpasteurized

orange juice

food

eggs

Vegetable-coated ready-to-eat snack

Bread dumpling loaf prepared with

undercooked beef

Spain Infant formula Kedougou 42

United States

United

Netherlands

United States

Australia


Table 2. Recent reported *Salmonella* outbreaks, including country (ies) affected, food vehicle and serovar.

The Role of Foods in *Salmonella* Infections 33

also dictated by its metabolic capabilities. However, the manifestations of these capabilities can be greatly influenced by intrinsic (e.g. vegetable moisture surface) and extrinsic ecological factors naturally present in the raw produce or imposed at one or more points during production, processing and distribution (Harris et al., 2003)*. Salmonella* strains may be able to enter a viable but nonculturable state (VBNC) on the surface of fruit and vegetables, resulting in underestimation of viable population size by direct plating on culture medium. Brandl and Mandrell (2002), suggested that *S.* Thompson may enter into a VBNC state on *Cilantro phyllosphere* due to exposure to dry pre-harvest conditions on the plant surface. Improved understanding of microbial ecosystems on the surface of foods such as raw fruits and vegetables would be extremely useful in developing strategies to minimize contamination, prevent pathogen growth, and kill or remove pathogens at different stages in production, processing, marketing and preparation for consumption. Food ecosystems are extremely diverse and complex. *Salmonella* survival and/or growth on foods are influenced by the organism, produce item and environmental conditions in the field and post-harvest, including storage conditions. For many years, the interaction of *Salmonella* with animal hosts and animal-origin foods has received intense attention. In contrast, little research has been done on the interaction between *Salmonella* spp. and fruits and vegetables, and more specifically on its frequency and behavior in fruits and vegetables which may pose a special risk to humans [e.g. radish root (*Raphanus sativus*), beetroot (*Beta vulgaris* var. *conditiva*), jicama (*Pachyrhizus erosus*), loroco (*Fernaldia pandurata*), prickly pear (*Opuntia* spp.), zucchini squash (*Cucurbita pepo*), chili peppers (Jalapeño and Serrano peppers) and others]. It is particularly urgent to study fruits and vegetables not previously considered health hazards and those with the potential

to function as pathogen microorganism vehicles but are as yet unidentified.

species or strains (Takeuchi et al., 2000).

In a recent *Salmonella* outbreak in the US, jalapeño and serrano peppers were the food vehicle and the isolated serovar was Saintpaul (CDC, 2008). It affected at least 1,442 persons in 43 states, the District of Columbia and Canada, and was traced back to distributors in the United States which had received produce grown and packed in Mexico. The outbreak strain was isolated from samples of jalapeño peppers collected from a US warehouse and a patient's home, as well as from samples of serrano peppers and water collected from a farm in Mexico. We have studied the behavior of *Salmonella* serotypes in zucchini squash and chili peppers. In zucchini, we tested the behavior of four *Salmonella* serotypes (Typhimurium, Typhi, Gaminara and Montevideo) and a cocktail of three *Escherichia coli* strains on whole and sliced zucchini squash at 25±2 and 3-5 °C. No growth was observed for any of the tested microorganisms or the cocktail on whole fruit stored at 25±2 or 3-5 °C. After 15 days at 25±2 °C, the tested *Salmonella* serotypes had decreased from an initial inoculum level of 7 log CFU to <1 log and at 3-5 °C they decreased to approximately 2 log (Figure 1). Among the *E. coli* strains, survival was significantly higher than for the *Salmonella* strains at the same times and temperatures: after 15 days at 25±2 °C, *E. coli* cocktail strains had decreased to 3.4 log CFU/fruit and at 3-5 °C they decreased to 3.6 log CFU/fruit (Figure 1). The observed differences in survival between the *Salmonella* and *E. coli* strains on zucchini squash fruit could be due to factors such as the area inoculated, fruit ripeness and physical and chemical characteristics of the studied fruit and strains. Different strains of *E. coli* O157:H7, *Pseudomonas*, *Salmonella*, and *Listeria monocytogenes* attach to different regions of cut lettuce leaves, indicating different and specific attachment mechanisms among different
