Qualitative Risk Assessment of Sprouted Seeds

1.INTRODUCTION

In recent years, raw sprouts have become a commonly consumed food item in North America and Europe.  Unfortunately, sprouts have also been found to be vehicles for food poisoning.  In 1973, in Texas, Bacillus cereus food poisoning resulted from the consumption of contaminated homegrown sprouts.  Since 1988, at least 14 Salmonella food poisoning outbreaks, 12 of which have occurred within the last four years, were reported in Canada, the US and the UK, again resulting from the consumption of sprouts.  Worldwide, at least 4 known foodborne outbreaks ofEscherichia coli O157:H7 due to sprouts have also been reported.

Contamination can occur during the growing, harvesting, storage and distribution of seeds and/or beans.  As a raw agricultural product, seeds/beans in the field or in storage, are frequently exposed to pathogenic bacteria, such as Salmonellaand E. coli O157:H7, from various sources (eg:  animals, birds, insects and domestic agricultural waste).  Even if the initial pathogenic contamination on or in the seed is low, the sprouting conditions under which the seeds/beans are kept to foster germination and sprout growth (soaking viable seed in water and then placing the seeds/beans in a warm, humid environment for an average of 3 to 7 days) are optimal for the growth of these bacteria.  Their numbers can quickly reach levels high enough to cause foodborne illness.

Seed producers may also not know whether their seeds are destined for food use and/or are not aware of the potential health risks.  Therefore, they may have little incentive to follow Good Agricultural Practices (GAP).  Furthermore, seed processing, shipping and selling practices often involve the pooling of multiple seed lots from different origins, making it difficult to trace back the origin of the contaminated lots, as well as providing excellent opportunities for cross-contamination.  Consumers may not wash and/or cook sprouted seeds/beans prior to consumption, since these steps may damage their consistency or shorten their shelf life.

            Activities have been carried out by a number of organizations to address the problem of Salmonella food poisoning from alfalfa sprouts.  For example, the International Sprouts Growers Association (ISGA) has developed a document called “Guide to Hygienic Code of Practice in the Production and Sale of Sprouted Beans and Seeds“;  the B.C. Ministry of Health prepared a document entitled “Interim Guidelines for the Production of Raw Seed Sprouts“; and Campden Chorleywood Food Research Association, U.K., has proposed a GMP in Technical Manual #25 (Sept. 1989) entitled “Guidelines for the hygienic manufacture, distribution and retail sale of sprouted seeds with particular reference to mung beans“.  In 1999, the National Advisory Committee on Microbiological Criteria of Foods published a document entitled: “Microbiological Safety Evaluations and Recommendations on Sprouted Seeds“.  The United States Food and Drug Administration (USFDA) has recommended the use of 20,000 ppm of calcium hypochlorite for the treatment of seeds prior to sprouting.  The Codex Alimentarius has included an annex on sprouted seeds/beans within its Code of Hygienic Practice for the Primary Production and Packing of Fresh Fruits and Vegetables.  CFIA is currently finalizing a Code of Practice, for use by sprout manufacturers, for the Hygienic Production of Sprouted Seeds and Beans.  The draft of this code was used as a basis for the Codex annex on sprouted seeds/beans.

The key aspect of sprouted seeds/beans that increases the risk of foodborne disease compared to other fresh produce is the sprouting conditions which are ideal for the growth of pathogenic bacteria that may already be present on the seeds.

2.         PURPOSE

The purpose of this Qualitative Risk Assessment is to identify the microbial hazard(s) of concern linked to the consumption of raw sprouted seeds/beans and the likelihood of their ingestion via this food commodity.  It will also be used to determine the gaps in data with respect to the contamination of sprouted seeds/beans and the risk (severity of illness) to Canadian consumers.  This Risk Assessment will form the basis for risk management decisions regarding raw sprouted seeds and beans.

3.         SCOPE

            This Qualitative Risk Assessment will be concerned with raw sprouted seeds and beans such as, but not limited to: alfalfa, mung, mustard, cress, rose, radish, snow pea, clover and soy.  The hazards of concern are microbiological in nature and are limited to bacteria.  This risk assessment will not address viruses, nor parasites.  The format follows the principles set out by the Codex Alimentarius Commission, Draft Principles and Guidelines for the Conduct of Microbiological Risk Assessment.

The data presented in this risk assessment originate from national and international outbreak data as well as Canadian survey data supplied through the Canadian Food Inspection Agency’s (CFIA) monitoring and surveillance program. 

4.         RISK ASSESSMENT

4.1       HAZARD IDENTIFICATION

The purpose of this section is to identify the microorganisms of concern with this food commodity.  The following will predominantly be a qualitative process.  Information on these hazards were obtained from scientific literature, government agencies and expert opinions.   Scientific literature and microbiological surveys have shown the presence of a variety of foodborne pathogens in sprouted seeds/beans, including Salmonella species, Escherichia coli O157:H7, Listeria monocytogenes,Klebsiella pneumoniae and Bacillus cereus.

4.1.1    Salmonella spp.

            4.1.1.1  Biology

            Salmonella is a gram-negative, nonsporeforming rod-shaped, motile bacterium (D’Aoust, 1997b; USFDA, 1992a). It is commonly found in raw meat, poultry and seafood (USFDA, 1992a; D’Aoust, 1999).  Environmental sources of the organism include, but are not limited to water, soil, insects, factory surfaces, kitchen surfaces, animal feces (USFDA, 1992a).  All known species of Salmonella are pathogenic to humans (Doyle & Cliver, 1990).

4.1.1.2 Modes of Transmission

The consumption of inadequately cooked meat, poultry and egg products, or improperly refrigerated dairy, fresh produce and other food contaminated with Salmonella can result in illness (CDC, 1998).

Person-to-person transmission is also possible.  Salmonella is usually transmitted by the fecal-oral route, remaining in the intestinal tract and other internal organs for variable periods of time (D’Aoust, 1999).  Individuals may also become chronic carriers, meaning they asymptomatically carry Salmonella and may transmit the organism to other individuals via food or personal contact.  Transmission may also occur by handling pet turtles, baby chicks, frogs and snails that harbor Salmonella (CDC, 1998).  Salmonella may also survive extended periods of time in soil and water contaminated with fecal matter.

4.1.1.3  Epidemiology and Relevant Outbreaks

All age groups are susceptible to Salmonella, but symptoms can be more severe in the elderly, the young, and immunocompromised persons (CDC, 1998).  Outbreaks of Salmonella infection associated with sprouted seeds/beans have been reported in various parts of the world including United Sates, United Kingdom, Europe and Canada (Inami and Moler, 1999; Van Beneden, 1999; Puohiniemi, et al., 1997; D’Aoust, 1999). See Appendix A for a detailed description of the Salmonellaoutbreaks associated with sprouted seeds/beans.

Salmonella Saint-Paul, Virchow and Gold-Coast

In 1988, an outbreak consisting of 143 cases in the Oxford Region of the UK was caused by the consumption of raw mung bean sprouts contaminated withSalmonella Saint-Paul (O’Mahony et al., 1990).  S. Saint-Paul was isolated from mung bean sprouts obtained from various retail stores, and from mung bean seeds on the premises of the manufacturer.  S.Virchow was associated with 7 cases of foodborne infection due to the consumption of raw mung bean sprouts produced from seeds imported from Thailand and Australia (Jerngklinchan & Saitanu, 1993).  The year after, also in the UK, an outbreak of 31 culture-confirmed cases was associated with eating mustard cress sprouts contaminated with S. Gold-Coast.  The mustard cress sprouts involved in the outbreak were grown from imported seeds originating from the Netherlands (Joce et al., 1990).

            Salmonella Bovismorbificans

More recently, in 1994, 2 outbreaks, both due to S. Bovismorbificans, were linked to alfalfa sprouts.  A total of 492 cases were reported, 282 cases in Sweden and 210 cases in Finland (Ponka et al., 1995; Puohiniemi et al., 1997).  Investigations revealed that the implicated alfalfa sprouts were grown from alfalfa seeds imported from Australia (Puohiniemi et al., 1997).

            Salmonella Stanley

In the spring of 1995, an international outbreak of S. Stanley was again caused by alfalfa sprouts grown from contaminated seeds.  A total of 242 cases were culture-confirmed in the U.S. (17 states) and in Finland.  Molecular characterization of human patient isolates revealed that the same outbreak strain infected cases in U.S. and Finland (Mahon et al., 1997).  Follow-up investigation indicated the sprouts from both countries were grown from contaminated seeds obtained from a single supplier in the Netherlands, but due to the mixing of the seed lots, the source of the seed could not be determined (NACMCF, 1999).  This suggests that seeds may have been contaminated during the growing, harvesting and/or processing (Taormina et al., 1999).

Salmonella Newport

In 1995, the consumption of alfalfa sprouts was associated with an outbreak of S. Newport in Denmark.  Approximately 150 clinical cases were identified with this outbreak.  In late 1995 and early 1996, the province of British Columbia and the state of Oregon also experienced an outbreak of S. Newport infections associated with the consumption of eating contaminated alfalfa sprouts.  One hundred and thirty-three cases were culture-confirmed (Van Beneden et al., 1999).  The S. Newport isolates obtained from the outbreaks in Denmark, the state of Oregon, and the province of British Columbia showed identical PFGE pattern, that is they were found to be indistinguishable from each other (Aabo & Baggesen, 1997).  In addition, the implicated seeds were from the same shipper implicated in the S. Stanley outbreak (Taormina, et al., 1999).  Subsequent to these outbreaks, in 1996 the province of Quebec also reported 60 cases of S. Newport infection epidemiologically-linked to the consumption of contaminated alfalfa sprouts (Foodborne outbreaks in Canada linked to produce, In Press).  Phage-typing, antibiograms and PFGE pattern of human and alfalfa sprout isolates indicated that the  S. Newport outbreaks in Denmark, British Columbia & Oregon and Quebec were all linked (Foodborne outbreaks in Canada linked to produce, In Press).

            Salmonella Montevideo and Salmonella Meleagridis

In Nevada and California, 1996, approximately 500 culture-confirmed cases of S. Montevideo and S. Meleagridis infections due to alfalfa sprouts were reported (NACMCF, 1999).  Investigations that were conducted at the sprouting facility revealed unsanitary sprouting practices and poor employee hygiene (Taormina et al., 1999).

Contaminated alfalfa sprouts were responsible for one further Canadian outbreak in the fall of 1997 (Buck et al., 1998).  In this episode, 78 cases of S. Meleagridis infections were reported in three provinces (Alberta, Ontario & Saskatchewan) of which 43 occurred in Alberta.  An Alberta production plant (as well as other plants belonging to the same company in Ontario and Saskatchewan) used imported alfalfa seeds from one seed lot (Buck et al., 1998). The company marketed “organically grown” produce; alfalfa seeds did not receive a chlorine pre-soak prior to sprouting (Buck et al., 1998). S. Meleagridis isolates from the three provinces were of the identical phage-type (Buck et al., 1998).  The final count at the end of the outbreak was 124 cases (Foodborne outbreaks in Canada linked to produce, In Press)

            Salmonella Infantis and Salmonella Anatum

In 1997, alfalfa, china rose radish and snow pea sprouts were implicated in a S. Infantis and S. Anatum outbreak in Kansas and Missouri (NACMCF, 1999).  One-hundred and nine reported cases were culture-confirmed (Glynn et al., 1998; NACMCF, 1999).  The implicated sprouts were produced at a single sprout manufacturing facility, from seeds that came from various local farms (Pezzino et al., 1998).  Isolates from the sprouted seeds and patients has the same PFGE pattern (Pezzino et al., 1998).

            Salmonella Senftenberg

In late 1997 through to July 1998, two clusters of S. Senftenberg infections occurred in Nevada and California (NACMCF, 1999).  An alfalfa/clover sprout mixture from a single local sprout manufacturer was found to be the source of the 60 culture-confirmed cases.  Since the alfalfa/clover sprouts were always mixed before sale, it was not possible to determine which type of sprout caused the outbreak.  The clover sprouts may have been more likely the cause of the outbreak since during the entire outbreak period, the clover seeds used for sprouting were all from the same production lot, whereas the source of the alfalfa seed changed in March 1998 (NACMCF, 1999).

            Salmonella Havana and Salmonella Cubana

In May 1998, the consumption of alfalfa sprouts was linked to an outbreak of S. Havana in Arizona and California, causing 18 cases of human illness.  The implicated alfalfa sprouts were produced by one large California sprout manufacturer and one small manufacturer.

From May through to August 1998, 22 cases of S. Cubana occurred in Arizona, California, Maryland, New Mexico and Utah, all linked to the consumption of alfalfa sprouts (NACMCF, 1999).  The alfalfa sprouts originated from the same large California sprout manufacturer identified in the S. Havana outbreak (NACMCF, 1999).  The same seed lot was used by the large manufacturer to produce the sprouts associated with both outbreaks (NACMCF, 1999).  Analysis of the implicated seed lot yielded S. Havana, S. Cubana and S. Tennessee (NACMCF, 1999).

            Salmonella Mbandaka

From January to March 1999, Oregon, Washington, Idaho, and California experienced an outbreak of S. Mbandaka outbreak.  A total of 75 cases were reported(NACMCF, 1999) of which 34 were culture-confirmed (Johnson, 1999).  S.Mbandaka was isolated from the implicated alfalfa seed lot, from alfalfa sprouts obtained from the sprout manufacturer in Washington as well as from aseptically sprouted alfalfa sprouts (NACMCF, 1999).

Salmonella Java

During August and September of 1999, cases of S. Java in Canada were linked to the consumption of alfalfa sprouts.  A total of 61 cases were reported in Alberta (48), British Columbia (9), Saskatchewan (3) and Manitoba (1).  Of the 48 cases in Alberta, 45 were lab-confirmed 40 of which have the same PFGE pattern (Health Canada, 2001).  No further information is available on this outbreak at this time.

Salmonella St-Paul

Also in 1999, an outbreak of S. St-Paul occurred in California due to the consumption  clover sprouts.  Twenty-six culture-confirmed cases were identified (Johnson, 1999).  The clover sprouts were produced by a sprout manufacturer in California (Johnson, 1999).  No further information concerning this outbreak is known at this time.

Salmonella Typhimurium

In Colorado, 1999, 90 culture-confirmed cases of S. Typhimurium were linked to the consumption of clover sprouts (Johnson, 1999).  The same implicated seed lot was used by two sprout manufacturers in Colorado (Johnson, 1999).  S. Typhimurium isolates obtained from case-patients and from the bin used to transfer the sprouts had the same PFGE pattern (Johnson, 1999).

Salmonella Muenchen

Between September and October 1999, 51 laboratory-confirmed cases of S. Muenchen linked to the consumption of alfalfa sprouts were reported in the state of Wisconsin (Wisconsin Department of Health & Family Services, 1999).  Additional illnesses caused by S.  Muenchen in Arkansas, Idaho, Michigan, Missouri, Nevada and Washington seemed identical to those causing illness in Wisconsin (Wisconsin Department of Health & Family Services, 1999).  Investigations indicated that alfalfa seed from a single seed lot appeared to be associated with the illnesses in Wisconsin (Wisconsin Department of Health & Family Services, 1999).

Salmonella Enteritidis

Between late March to early April 2000, the first reported outbreak of salmonellosis associated with raw mung bean sprouts was reported in North America.  Forty-five cases of S. Enteritidis in the U.S. have been confirmed in Sacramento, Placer and Yolo counties of California.

Salmonella Enteriditis PT11

In Canada, between April to June 2000, Alberta and Saskatchewan experienced an outbreak of S. Enteritidis PT11.  Epidemiological investigations indicated that at least 10 cases of illness were associated with the consumption of mung bean sprouts produced from 2 different sprout manufacturers who used seeds from the same supplier in British Columbia.

Salmonella Enteritidis PT4b

Between November and December 2000, 25 cases of S. Enteritidis PT4b linked to the consumption of bean sprouts were reported in the Netherlands (Food Safety & Security, 2001; SproutNet, 2001a).  S. Enteritidis was isolated by the sprout manufacturer through quality-control sampling (Food Safety & Security, 2001; SproutNet, 2001a).

Salmonella Enteritidis PT913

In February 2001, 46 lab-confirmed cases of S. Enteritidis PT913 linked to bean sprouts, were reported (BMH, person. commun.).  Of the 46 cases, 41 resided in the Capital Region of Alberta (Edmonton), 2 in Calgary and 3 in British Columbia.  All the non-Capital region cases visited Edmonton during the incubation period, between February 03 and 10, 2001.  Prior to this outbreak, sprout samples taken revealed high fecal coliform counts.  No further information concerning this outbreak is known at this time.

4.1.2    Escherichia coli O157:H7

            4.1.2.1  Biology

            E. coli is a gram-negative bacterium that is a normal inhabitant of the intestines of all animals, including humans. When aerobic culture methods are used, E. coli is the dominant species found in human feces. Normally E. coli serves a useful function in the body by suppressing the growth of harmful bacterial species and by making nutrients available to the host.  A few strains of E. coli, however, are pathogenic.  Currently, there are 4 recognized categories of enterovirulent E. coli(EEC) that cause gastroenteritis in humans. These are the enterotoxigenic E. coli(ETEC), enteropathogenic E. coli (EPEC), enteroinvasive E. coli (EIEC) and verocytotoxin-producing E. coli (VTEC) (Willshaw, 2000).  Enterohemorrhagic E. coli(EHEC), a subset of VTEC, are strains that can cause bloody diarrhea in humans (Willshaw, 2000).  EHEC possess virulence factors in addition to the verocytotoxin (Willshaw, 2000).  Among the EHEC is a strain designated E. coli O157:H7, that was found to be of most concern in sprouted seeds/beans.  Three syndromes, hemorrhagic colitis (HC), hemolytic uremic syndrome (HUS) and thrombotic thrombocytopenic purpura (TTP) have been linked to E. coli O157:H7.

            4.1.2.2  Modes of Transmission

The intestinal tract of animals, primarily cattle, are the principal reservoirs for E. coli O157:H7 (Willshaw, 2000).  Transmission by contaminated water supplies, person-to-person transmission, as well as direct animal-to-human transmission, have also been documented  (Karmali, 1989; Griffin and Tauxe, 1991; Renwick et al., 1993; Willshaw, 2000).

Most outbreaks and sporadic cases of E. coli 0157:H7 have been associated with the consumption of foods of bovine origin including ground beef and raw milk.  However, in recent years, raw vegetables (USFDA, 1992b) as well as drinking water have also been implicated. 

4.1.2.3  Epidemiology and Relevant Outbreaks

All people are believed to be susceptible to hemorrhagic colitis, but the young, elderly and immunocompromised are more sensitive (Griffin and Tauxe, 1991).  Between 5-16% of HC patients may develop HUS (Karmali et al., 1983; Cassin et al., 1998; USDA, 1999) of which 5-12% usually die (Cassin et al., 1998).  The mortality rate in the elderly can be as high as 50% (USFDA, 1992b).

Hemorrhagic colitis infections have not been commonly reported, but this is probably not reflective of its true frequency.  In the Pacific Northwest, E. coliO157:H7 is thought to be second only to Salmonella as a cause of bacterial diarrhea.  Individuals having symptoms of profuse bloody diarrhea will probably seek medical attention and the information regarding those cases is reported and compiled.  Less severe cases, however, may go unreported (USFDA, 1992b).  Outbreaks of E. coliO157:H7 infection associated with sprouted seeds/beans have been reported in the United States and in Japan.

E. coli O157:H7 in Japan (1996)

Between 1996 and 1997, a total of five E. coli O157:H7 outbreaks, linked to white radish sprouts, occurred in Japan.  In July 1996, white radish sprouts were epidemiologically linked to the world’s largest outbreak of E. coli O157:H7.  In total, 6,561 cases among 62 public elementary schools in the city of Sakai were reported (Ministry of Health and Welfare of Japan, 1997; CSR, 1996).  On the 8^th or 9^th of July, depending on the school, white radish sprouts were served in the children’s school lunches (FSNET, August 30 1996).  Six thousand three-hundred and nine school children, 92 teachers and staff, and 160 family members were affected (CSR, 1996).   That same month, 47 cases of E. coli O157:H7 were reported among factory workers in Kyoto, Japan (Watanabe, 1999).  On July 11, 1996 the factory cafeteria served white radish sprouts, purchased from the same sprout manufacturer that was implicated in the Sakai city school outbreak (Watanabe, 1999).  Isolates from the Sakai city school outbreak and the factory outbreak had indistinguishable PFGE and RAPD patterns (Watanabe, 1999).  Also in 1996, another outbreak of E. coli O157:H7 was linked to the consumption of white radish sprouts in Horbikino, Japan (FSNET, August 30 1996).  A total of 98 cases were reported among an elderly home and 3 smaller outbreaks in a nearby region (FSNET, August 30 1996).

E. coli O157:H7 in Japan (1997)

In March 1997, an outbreak of 96 cases, linked to white radish sprouts was reported in Tokyo, Yokohama, Nagoya (LCDC, 1997).  Investigations revealed that the majority of E. coli O157:H7 isolates obtained in this outbreak from patients and asymptomatic carriers had the same PFGE patterns (LCDC, 1997).  Another outbreak of E. coli O157:H7, involving 126 individuals, also originating from white radish sprouts was reported that same year (Gutierrez, 1997).

 E. coli O157 in the US (1997-1998)

IN 1997, the Centers of Disease Control and Prevention (June-July1997) reported  that in Michigan and Virginia, simultaneous outbreaks of E. coli O157:H7, resulting in a total of 108 epidemiologically-linked cases, were independently associated with eating alfalfa sprouts grown from the same lot of seed.  The outbreak strains from Michigan and Virginia were found to be indistinguishable.  Although not confirmed, it was suspected that infection originated from contaminated seeds rather than contamination during the sprouting process.  In June of 1998, an E. coli:NM outbreak linked to the consumption of alfalfa/clover sprout mixture occurred in Nevada and California.  Eight culture-confirmed cases linked the outbreak to the same sprout manufacturer implicated in the 1997-1998 outbreak ofSalmonella Senftenberg in California and Nevada (NACMCF, 1999).

4.1.3    Listeria monocytogenes

            4.1.3.1  Biology

Listeria monocytogenes (LM) is a short, gram-positive, motile, non-sporeforming pathogen that is hardy and fairly resistant to freezing, drying and heat (Farber & Peterkin, 1991; USFDA, 1992c; ).  LM is widely distributed in soil, water and plant vegetation where it can persist for a long period of time.  The ability of LM to grow at low temperatures permits it to multiply in refrigerated foods (Farber & Peterkin, 1991; USFDA, 1992c).  Illness due to the consumption of foods contaminated with Listeria monocytogenes is called listeriosis.

The manifestations of listeriosis include septicemia (bacteria in the bloodstream), meningitis (brain infection), encephalitis (brain inflammation) and infection in pregnant women (which may result in spontaneous abortion or stillbirth).  The symptoms due to listeriosis are variable, depending on the individuals’ age and susceptibility (Farber & Peterkin, 1991). 

4.1.3.2  Modes of transmission

Several modes of transmission of this organism have been identified, such as mother to fetus infection in utero or infection at childbirth, infant to infant, animal to human, and most importantly,  foodborne transmission (Farber & Peterkin, 1991).  Human listeriosis has been associated with foods such as raw milk, pasteurized milk, cheeses, coleslaw, vegetables, smoked salmon, chicken, and hot dogs (Farber, 1999). 

            Few data are reported on the occurrence of L. monocytogenes on fresh sprouted beans and seeds.  Arumugaswamy et al. (1994) reported that a high percentage of beans sprouts analysed in Malaysia were positive for L. monocytogenes. 

4.1.3.3  Epidemiology and Relevant Outbreaks

The highest incidence of listeriosis is found among the neonates, elderly and  immunocompromised individuals (Farber & Peterkin, 1991).  There is also a significant proportion of cases of  listeriosis in pregnant women (Farber & Peterkin, 1991).

Based on epidemiological and laboratory evidence, several foodborne outbreaks have been documented in North America.  Though no outbreaks due to sprouted seeds/beans have been reported, L. monocytogenes has caused an outbreak involving cabbage, a raw vegetable commodity.  In 1981, Nova Scotia experienced 41 cases of listeriosis.  All cases resulted from the consumption of coleslaw made from cabbages that were believed to have been contaminated with untreated sheep manure (Schlech WF 3d et al., 1983).

4.1.4    Klebsiella pneumoniae

            4.1.4.1  Biology

Klebsiella pneumoniae is a large, gram-negative, non-spore forming, non-motile bacterium.  An opportunistic human pathogen of increasing importance in hospital environments (Bagley, 1985), it is carried in the intestinal tract of 30 to 40% of humans and animals (Bagley & Seidler, 1977).  Intestinal colonization of  K. pneumoniae in humans has been attributed to the ingestion of contaminated foods, including raw vegetables (Shooter et al., 1971). Bagley & Seidler (1977) state that, 85% of fecal coliform positive K. pneumoniae isolated from the environment are of pathogenic origin.  Since pathogenic strains of K. pneumoniae have been shown experimentally to multiply on raw vegetable surfaces, plants are therefore considered as a good reservoir (Brown et al., 1973).

            K. pneumoniae can cause illness in infants, elderly and immunocompromised individuals (Highsmith & Jarvis, 1985), but it has not been shown to be pathogenic for healthy individuals (Patterson & Woodburn, 1980).

            4.1.4.2  Modes of Transmission

Person-to-person spread through the fecal-oral route as well as coughing are the most common modes of transmission of Klebsiella spp.  Klebsiella is most commonly spread in hospitals and nursing homes among immunocompromised individuals.

Park and Sanders (1990) stated that the major transmission route of K. pneumoniae to the alfalfa and mung seeds is likely through soil, since the amounts of soil on the surface of the seeds roughly parallelled the counts of K. pneumoniae, and the bacteria were readily isolated from soil which had been separated from seeds. 

4.1.4.3  Epidemiology and Relevant Outbreaks

Foods that are frequently contaminated with Klebsiella spp. include meats, milk and dairy products, fresh vegetables, salads and drinking water.  In a study done by Patterson and Woodburn (1980), large populations (10⁶ CFU/g) of K. pneumoniae were found on alfalfa and bean sprouts at the retail level in the state of Oregon.  Similar results were obtained from vegetable sprouts sold in Central Canada (Park et al., 1981), and in the state of New York (Splittstoesser et al., 1983).  Skrowronek et al (1998) stated that microbial populations within the genus Enterobacteriaceae, such as Klebsiella pneumoniae, exceeding 5 x 10⁶ CFU/g may display toxic metabolites.

In the past, K. pneumoniae was not recognized as a foodborne pathogen and may have been overlooked during foodborne investigations (Sabota et al., 1998). 

4.1.5    Bacillus cereus

4.1.5.1 Biology

B. cereus is a gram-positive, sporeforming, motile facultative anaerobic foodborne pathogen that is widely found in nature.  Two distinct illnesses have been identified (Farber 1989; Johnson 1999).  The diarrheal form of foodborne illness has symptoms that mimic Clostridium perfringens foodborne illness (USFDA, 1992d).  The second type of food poisoning, the emetic form (vomiting-type), mimics the Staphylococcus aureus food poisoning symptoms (USFDA, 1992d).

            4.1.5.2  Modes of Transmission

Due to the widespread contamination of our food supply with B. cereus, its ingestion may be regarded as inevitable. Food vehicles are quite varied, ranging from vegetables and salads to meat dishes (Johnson, 1990).  Emetic illnesses seem to be almost exclusively associated with rice (Johnson, 1990).

4.1.5.3  Epidemiology and Relevant Outbreaks

B. cereus has been recognized as a food poisoning organism for quite a long time (Raevuori et al., 1976). The increasing awareness and recognition of B. cereusassociated illness has resulted in a substantial increase in the number of reports of the ‘diarrheal syndrome’ type food poisoning.  The symptomatic similarities between B. cereus food poisoning and the foodborne illnesses caused by Clostridium perfringens and Staphylococcus aureus may have caused B. cereus outbreaks in the past to be unreported and/or misdiagnosed (USFDA, 1992d).  One food poisoning outbreak due to B. cereus in sprouts has been reported.  In Houston, Texas, 1973, an outbreak of B. cereus food poisoning resulting from contaminated homegrown vegetable sprouts was reported (Portnoy et al, 1976).  Four persons developed gastrointestinal illness after the consumption of raw soy, mustard and cress sprouts contaminated with B. cereus during germination in a commercially available seed sprouting kit.

4.1.6    Other pathogens

Other pathogens can also be a cause of concern if found in sprouted seeds or beans.  Yersinia enterocolitica is commonly isolated from different environments such as lakes, rivers, wells, and soil (Kapperud, 1991) Y. enterocolitica can also be found in diverse foods of animal origin, such as pork, beef, poultry, and dairy products.  In 1989, Cover & Aber reported that yersiniosis was associated with eating non-commercially produced bean sprouts that were grown using pond water.  This bacterium can likely grow during sprout production, but no specific data are available.  Shigella is another pathogen to consider because of its low infectious dose (10-100 organisms), its dissemination in fecally-contaminated water and its ability to proliferate in vegetables (Rafii et al., 1995).

4.1.7    Conclusions on the Hazard Identification

Several microbial agents were evaluated to determine the potential human health hazard associated with sprouted seeds and beans.  E. coli O157:H7, Salmonella spp., Bacillus cereus, Listeria monocytogenes and Klebsiella pneumoniaehave been isolated from fresh sprouted beans and seeds.  Home grown raw vegetable seed sprouts contaminated with B. cereus have been associated with one outbreak of gastrointestinal illness.  In  recent years, Salmonella has been increasingly linked to human illness from commercially produced fresh sprouted beans and seeds in Canada, the U.S., Europe and the UK.  E. coli O157:H7 has also been associated with sprout-borne outbreaks in the U.S. and Japan.  Based on the most recent outbreaks, the microbial hazards of major concern are Salmonella and E. coli O157:H7.

4.2  EXPOSURE ASSESSMENT

The following Exposure Assessment will serve to provide an estimate of the anticipated human exposure to Salmonella and E. coli O157:H7 via sprouted seeds/beans.  This will be accomplished by predicting possible exposure to both pathogens during seed/beans and sprouted seeds/beans production, the likelihood of survival and growth of Salmonella and E. coli O157:H7 on/in the seeds/beans and sprouted seeds/beans, and the patterns of consumption of sprouted seeds/beans in Canada.  Levels of human exposure to Salmonella and E. coli O157:H7 are also dependent on the controls that exist throughout the production and processing of seeds/beans and sprouted seeds/beans. 

4.2.1    Potential Exposure of Seeds/Beans and Sprouted seeds/beans to Salmonellaand E. coli O157:H7

            4.2.1.1 Potential Exposure during Seed Production

There is little information available on how seeds become contaminated with bacterial pathogens, though epidemiological investigations suggest that they are the likely source in most, if not all, sprout-associated outbreaks (Puohiniemi et al., 1997; CDC, 1997; Mahon et al., 1997).  Dried seeds and beans may be exposed to Salmonella and E. coli O157:H7 anywhere during the growing, harvesting, cleaning and  transportation from the seed grower to the sprout manufacturer.  Since seeds and beans are considered raw agricultural commodities, they may be exposed to high levels of Salmonella and E. coli O157:H7 from the field through contaminated agricultural water, improperly managed as well as improperly treated animal manure, contact with wild animals, location of agricultural fields near animal rearing facilities and inadequate worker hygiene.  The equipment used in post-harvesting, cleaning and packing of the dried seeds and beans could also be contaminated with rodent or animal manure and urine, as well as be contaminated from previous contaminated lots.  During transport, seeds can also become contaminated from the transport vessel.  In the US, the National Advisory Committee on Microbiological Criteria for Foods, NACMCF, (1999) also suggests that damage to seeds could aggravate contamination by making removal of pathogenic microorganisms during subsequent steps more difficult.  Seed processing, shipping and selling practices often involve mixing multiple seed lots of different origins, therefore complicating traceback and providing greater opportunity for cross-contamination (NACMCF, 1999).

4.2.1.2 Potential Exposure during Sprout Manufacture

Although the seed appears to be the primary source of contamination in the sprout-associated foodborne illness outbreaks, practices at the sprouting establishment may increase or decrease the extent of the microbial hazards.  The manufacturing of sprouted seeds/beans involves a fair amount of water during sprouting. Therefore, the use of untreated or non-potable water can be a source of Salmonella and/or E. coli O157:H7.  These pathogens may also be introduced into a sprouting facility via previous lot of contaminated seeds and/or beans, an unsanitary harvesting/packaging facility, as well as unsanitary transportation practices.  These organisms can also persist due to poor sanitation and inadequate hygienic practices.

Frequent failures to isolate pathogens from implicated seeds suggest that seed contamination may be present at very low levels, or unequally distributed within the seed lot.

4.2.2    Likelihood of Survival and Growth of Salmonella and E. coli O157:H7

4.2.2.1 Likelihood of Survival and Growth in Seeds

Various studies have revealed that prior to sprouting, seeds are capable of harbouring significant levels of microorganisms (Andrew et al., 1979; Andrew et al., 1982; Prokopowick & Blank, 1991; Piernas & Guiraud, 1997).  It is important to keep in mind that, although the presence of pathogenic microorganisms, such as Salmonella and E. coli O157:H7, may be present on seeds at low levels, the environmental conditions and nutrients present during sprout production are ideal for the growth of these pathogens. In addition, once present on or in the seeds, Salmonella and/or E. coli O157:H7 are capable of surviving for extended periods of time.   Therefore, if poor sanitation and inadequate hygiene are practised, both Salmonella and E. coliO157:H7 can persist for long periods of time.

            4.2.2.2 Likelihood of Survival and Growth in Sprouted seeds/beans

Sprouted seeds are produced by soaking seeds in water and then placing them in a warm humid environment for an average of 3 to 7 days to foster germination and sprout growth (NACMCF, 1999).  If Salmonella and/or E. coli are present on or in the seeds, the sprouting conditions used (i.e.: time, temperature, water activity, pH, and nutrients) provide ideal conditions for their  growth.  Jaquette et al. (1996) found that S. Stanley increased by 2.5 logs in 24 h during the germination of alfalfa seed and by another 1 log during the first 18 h of sprouting.  When S. Anatum and S. Montevideo were pooled and inoculated onto mung beans, they increased in number by 4.5 logs, while pooled S. Eimsbuettel and S. Poona, inoculated onto alfalfa seeds, increased by 3-4 logs during sprouting (Andrew et al., 1982).  E. coli O157:H7, inoculated onto radish seeds, increased in number  by 4-5 logs during the first 24 h of sprouting (Hara-Kudo, 1997).  Also, during radish sprout production, Bari et al. (1999) observed that within 2 days, E. coli O157:H7 increased in number from 10² to 10⁶ – 10⁷ cfu/ml.  After 6 days, E. coli O157:H7 had increased in number and remained constant at 10⁸cfu/ml.  Ingram et al. (1998) reported that populations of E. coli O157:H7, when inoculated onto alfalfa seeds  increased to 10⁷cfu/g in/on the alfalfa sprouts.

4.2.3    Process Control

Process control in the production of seeds/beans and sprouted seeds/beans, is any practice or form of treatment(s) during the production of seeds/beans and/or sprouted seeds/beans that will reduce the likelihood of their exposure to pathogen(s), particularly Salmonella and E. coli O157:H7.  As mentioned in sections 4.2.1.1 and 4.2.1.2 seeds/beans and/or sprouted seeds/beans, respectively, can be exposed to Salmonella and/or E. coli O157:H7 through untreated water, improperly managed animal manure, contact with livestock, wild animals, rodents and pests, and inadequate worker hygiene.  Previously contaminated lots, unsanitary harvesting and packaging, as well as unsanitary transportation measures can also cause be sources of Salmonella and E. coli O157:H7.

Presently, not much is known regarding the controls taken by the seed and sprout manufacturers to reduce the potential exposure of seed/bean and sprouted seeds/beans to Salmonella and E. coli O157:H7.  The focus of the following process controls are on risk reduction and not on risk elimination because, the currently available technologies with respect to the production of sprouted seeds/beans, cannot eliminate all potential food safety hazards associated with seeds/beans and sprouted seeds/beans.

            4.2.3.1 Process Control for Seed Producers

Seed producers may not necessarily know whether their seed is destined for food use and/or may not be aware of the potential health risk, therefore having little incentive in following Good Agricultural Practices (GAP).  Furthermore, seed processing, shipping and selling practices often involve the pooling of multiple seed lots from different origins, making it difficult to trace back the origin of the contaminated lot, as well as providing excellent opportunities for cross-contamination.

Livestock and Wildlife

Animal feces are well known sources of both Salmonella and E. coli O157:H7.  While it is not possible to completely exclude all animals from the seed production fields, seed producers may consider preventing livestock from entering by using physical barriers.  If seed production areas are adjacent to wooded areas, open meadows, and waterways, seed producers may consider deterring or redirecting wildlife to other areas, such as seed production areas not destined for sprouting.

During heavy rains, seed producers may also want to have ditches or buffer areas  to prevent animal waste from adjacent facilities from contaminating the seed production field.

Animal Manure

The use of animal manure and other fecal matter in the production of seeds destined for sprouting represents a significant source of Salmonella and E. coliO157:H7. Seed producers may want to consider treating the animal manure to reduce the levels of both of these pathogens.  Passive treatments of animal manure rely on the passage of time in conjunction with environmental factors (e.g.,  natural temperature and moisture fluctuations and ultraviolet (UV) irradiation).  Seed producers that prefer to use passive treatments for animal manure would have to ensure that the manure is well aged and decomposed before usage.  Active treatments of animal manure involve a greater level of intentional treatment.  Examples of active treatment include, pasteurization, heat drying, anaerobic digestion, alkali stabilization, aerobic digestion, etc.

Harvesting

Contamination with Salmonella and E. coli O157:H7 during the harvesting of seeds/beans may result from contact with soils, fertilizers, water, workers and unsanitary harvesting equipment.  Any surfaces that come into contact with the seeds/beans during harvesting can potentially be sources of contamination.  The general cleanliness of the harvesting equipment will help in reducing the risk of cross-contamination of the seeds.

Packaging

An improperly maintained packaging facility may attract rodents and pests.  Rodents and pests may expose the seeds to Salmonella and E. coli O157:H7.  Daily cleaning in and around the packing facility, will reduce the risk of rodent entry into the facility.

Packaging Equipment

The improperly sealed packaged seeds/beans as well as unclean packaging material may lead to the contamination of the seeds/beans.  The use of clean and properly sealed packaging material is an important factor in reducing the potential exposure of seeds/beans to Salmonella and E. coli O157:H7  contamination. 

Transportation

Microbial cross-contamination from other foods and food sources and contaminated surfaces may occur during the loading, unloading, storage and transportation of seeds.  The regular cleaning, sanitizing and monitoring of transportation containers and vehicles will help reduce the exposure of seeds to contamination by Salmonella and E. coli O157:H7.  Separation of seeds from other foods and non-food sources during transport to the sprout production facility will also help decrease exposure of the seeds to Salmonella and E. coli O157:H7.

            4.2.3.2 Process Controls for Sprout Manufacturers

In the summer of 1999, the Canadian Food Inspection Agency surveyed 82 sprout manufacturing facilities across Canada to determine the environment in which sprouted seeds/beans are produced.  A total of 38 surveys were completed.

Incoming Seeds

It is important that the seeds/beans, which are destined for sprouting, and are coming into the sprout manufacturing facility be: i) from a reputable supplier, ii) of good quality (produced under good agricultural practices) and iii) free of foreign matter.   Thirty-four out of the 38 sprout manufacturers were able to provide information on the origin of the seed, though little information was provided regarding incoming seed bag quality and whether visual and/or superficial inspection of foreign matter was carried out.

Sprouting Establishment

Improperly maintained sprouting facilities may attract rodents and pests, which, in turn, expose the sprouted seeds/beans as well as the equipment used for sprouting to pathogenic bacteria such as Salmonella and E. coli O157:H7.  Out of the 38 sprout manufacturers that CFIA inspected, 25 maintained clean and dry areas for seed storage.  Only 12 out of the 38 manufacturing facilities were using adequate ventilating systems.  Daily cleaning and proper ventilation of the facility will reduce rodent entry and, as such, reduce exposure, survival and growth of Salmonella and E. coli O157:H7.  Taormina & Beuchat (1999) found that storing dry alfalfa seeds at elevated temperatures 25 and 37EC) enhanced the rate of death of E. coli O157:H7 on the seeds.  Very few (5/38) sprout manufacturers had a written sanitation program, and even fewer (2/38) had a written environmental sampling program.

Cross-contamination of the seeds/beans and sprouted seeds/beans from contaminated equipment used in the sprouting facility may also occur.  Out of the 38 sprout facilities surveyed, 21 had separate production areas so as to limit cross contamination between the storage, manufacturing and packing areas.  Daily cleaning and sanitizing of the equipment used in sprouting, can help reduce cross-contamination of seeds/beans and sprouted seeds/beans.

The accessibility and cleanliness of fully functional rest rooms are also important in reducing the risk of introduction of these pathogens into the facility.  Twenty-three of 38 facilities surveyed had well maintained rest rooms, while the other 15 did not.

Worker Hygiene

Worker hygiene during the manufacturing, harvesting, sorting, packing and transporting of sprouted seeds/beans also plays an important role in minimizing the potential exposure of seeds/beans and sprouted seeds/beans to Salmonella and E. coli O157:H7.  Only 2 out of the 38 sprout facilities had a written training program to address worker hygiene. After treating the seeds, sprout manufacturers must not re-contaminate the seeds/beans and/or sprouted seeds/beans during the other steps in the sprouting process.

Seed Decontamination Treatments

The prevention of contamination by Salmonella and E. coli O157:H7, plus the implementation of seed disinfection treatment will significantly reduce the levels of  Salmonella and E. coli O157:H7 on the seeds.  Presently, approximately 45% (17/38) of the surveyed sprout manufacturing facilities use some kind of seed treatment.  Out of the 17 manufacturing facilities, six used various concentrations of hydrogen peroxide, one sprout manufacturer used chlorine concentrations between 2, 000 ppm to 3, 500 ppm, six used 50-250 ppm of chlorine, while four used chlorine concentrations less than 50 ppm.

Once the seed is treated with a disinfectant, it is important to maintain Good Manufacturing Practices so as not to re-contaminate the seeds.  One must keep in mind that different types of seeds vary in sensitivity to antimicrobial treatments, therefore, a treatment that is effective for one type of seed may not be applicable to all types of seeds.  In general, alfalfa seeds are the most difficult to effectively sanitize (NACMCF, 1999).

A successful treatment for seed disinfection would inactivate E. coli O157:H7 and/or Salmonella, while preserving the viability, germination and vigour of the seed.  Research is in progress to find seed disinfection treatments that will reduce the pathogenic loads found on seeds and/or beans. To date, there is no single available treatment that has been identified that can completely eliminate the total bacterial pathogen load on seeds/beans without compromising germination.  For a more detailed overview of seeds/beans/sprouted seeds/beans disinfection treatments, and the research that has been on them thus far, please see Appendix B.

Treatment with Chlorine

Hypochlorite solutions are the oldest and most widely used active chlorine compounds in the field of chemical disinfection (Dychdala, 1983).  Sodium/calcium hypochlorite reacts with water to form hypochlorous acid, which has a bactericidal effect.  This bactericidal effect of hypochlorous acid results from the rapid combination of chlorine compounds with bacterial proteins.  It non-competitively inhibits enzymes, especially those involved with glucose metabolism.  The calcium and sodium forms of hypochlorite appear to be equally effective for inactivating E. coli O157:H7, however, the calcium form is preferable, because it is not as volatile.

Dychdala (1983) also noted that chlorine inactivation of bacteria increased with increasing exposure time, chlorine concentration and temperature, but substantially decreased with increasing pH.  It is important to note that the activity of available (or free) chlorine may be diminished as it comes into contact with organic materials.  Therefore, the effectiveness of chlorine solutions may decrease upon contact with the seeds/beans (Taormina & Beuchat, 1999).

In 1996, the International Sprout Growers Association (ISGA) recommended pre-soaking the seeds, while stirring, in sodium hypochlorite (2, 000 ppm) or calcium hypochlorite (1, 800 ppm) for a maximum of 10 min at 21EC, as well as rinsing them in fresh water, prior to gemination.  Jaquette et al. (1996) suggested that a 2, 000 to 4, 000 ppm chlorine soak treatment could be used on alfalfa seeds prior to germination to reduce populations of S. Stanley and possibly other salmonellae, while not significantly affecting germination.  Beuchat (1997) inoculated alfalfa seeds with a mixture of 5 Salmonella serotypes (S. Stanley, S. Poona, S. Montevideo, S. Harford, S. Newport) and used a variety of treatments including: calcium hypochlorite (1, 800 ppm) or sodium hypochlorite (2, 000 ppm) for 10 min.  He reported that the latter treatment effectively reduced microbial populations by more than 3 logs. Piernas and Guiraud (1997) used a sodium hypochlorite treatment (1, 000 ppm) for decontaminating rice seeds and found that it reduced aerobic plate counts (APCs) on rice seeds by 2 to 3 logs. Taormina & Beuchat (1999) noticed that significant reductions in populations of E. coli O157:H7 occurred when alfalfa seeds were treated with $ 2, 000 ppm of active chlorine, obtained from using calcium hypochlorite.  Weissinger & Beuchat (2000) reported a 0.72 log reduction of Salmonella when alfalfa seeds were soaked in 2 000 ppm chlorine (NaOCl).  Acidified OCl₂ ($ 100 ppm) and acidified NaOCl ($500 ppm) reduced populations of  E. coli O157:H7 by more than 2 logs.   They also observed that the percent reduction of E. coli O157:H7 was greater with the use of  2, 000 ppm and 20, 000 ppm of chlorine as compared to a 200 ppm chlorine solution.  Taormina and Beuchat (1999) concluded that the use of $ 2, 000 ppm (CaOCl₂), $100 ppm (acidified OCl₂) or $ 500 ppm (acidified NaOCl) significantly reduced E. coli O157:H7 populations, with little or no effect on percent seed germination.  In 1999, the USFDA recommended the use of 20, 000 ppm of calcium hypochlorite for 15 min for the treatment of seeds.  Beuchat et al. (2001) reported reductions of 2.3 and 2.6 log₁₀ of Salmonella and E. coli O157:H7, respectively, when alfalfa seeds were soaked for 15 min in 20 000 ppm chlorine (Ca(OCl)₂).  When the seeds were soaked for a longer period of time (30 min) the E. coli OI157H7 populations was reduced by 2.0 logs (Beuchat et al., 2001).  Weissinger & Beuchat (2000) showed that treating alfalfa seeds with 20 000 ppm chlorine ((CaOCl)₂) for 10 min results in a 1.95 log₁₀reduction of Salmonella.  Lang et al. (2000) used 2 000 ppm chlorine to decrease E. coli O157:H7 population on alfalfa seeds by 6.6-6.9 logs.   Beuchat et al. (2001) also noted that differences in effectiveness of chlorine may be attributed to the use of different lots of seeds, Salmonella serotypes, and the procedures used for inoculation, treatment and recovery

            Treatment with Chlorine + Heat

A minimum of 5-log reduction in APCs was achieved when a 5 min treatment of seeds with sodium hypochlorite (1, 000 ppm) was combined with heating at 60EC (Piernas & Guiraud, 1997).  Taormina and Beuchat (1999) reported that treatment with calcium hypochlorite (20, 000 ppm) for 3 min at 55EC eliminated pathogens from three out of three sprout samples, although the pathogens were detected in all samples after cultural enrichment.

            Treatment with Hydrogen Peroxide

The antimicrobial activity of hydrogen peroxide is due to its oxidizing capacity.  ISGA (1996) had recommended the use of 6% hydrogen peroxide soak for a maximum of 10 min at 21EC.  The seeds would then have to be rinsed in fresh water prior to the germination phase.  Beuchat (1997) reported that treatment of alfalfa seeds with 6% hydrogen peroxide for 10 min reduced microbial populations by 3 logs.  Piernas and Guiraud (1997) found that a 1% hydrogen peroxide treatment (10, 000 ppm)  reduced microbial counts on rice seeds by 2 to 3 logs.  Weissinger & Beuchat (2000) reported that Salmonella was reduced by ~ 3 logs when the seeds were treated with 8% hydrogen peroxide.

            Gaseous Acetic Acid

Delaquis et al. (1999) hypothesized that gaseous antimicrobials, including acetic acid, are more likely to come into contact with the microorganisms buried within the cracks and crevices of seeds.  Acetic acid is an inexpensive common food ingredient. The microbicidal effect of this treatment increases with increasing concentration, length of exposure time and temperature. Based on the highest mung bean seed inoculation levels tested, treatment with 242 FL of acetic acid per litre of air at 22EC for 24 h led to reductions of approximately 5, 6 and 4 log cycles for S.Typhimurium, E. coli O157:H7 and L. monocytogenes, respectively (Delaquis et al, 1999).  The recovery of viable L. monocytogenes after treatment suggests that susceptibility to gaseous acetic acid varies from species to species. The germination rates of the mung bean seeds were not significantly affected by fumigation with gaseous acetic acid.

Calcinated Calcium

Calcinated calcium is produced from natural oyster shells that are ground with their pearl layer being treated electrically with ohmic heating (220V, 60-100A for 10-60 min).  The shells are crushed into powder and used as a nutrient in Japan (Bari et al., 1999).  Bari et al. (1999) conducted a study to determine the growth inhibitory effects of calcinated calcium on E. coli O157:H7 during radish sprout production, when the seeds and/or water are contaminated with very small numbers of E. coliO157:H7 (1-1.6 x 10³ CFU/ml) at an early stage in sprout production.  They found that 0.07% (pH 11.2) calcinated calcium completely inhibited the growth of 1 x 10³to 1.6 x 10³ CFU/ml of E. coli O157:H7 in radish seed extract medium and no recovery was observed during incubation times of up to 96 h.  They also tested the effectiveness of calcinated calcium to decontaminate water used during sprout production and which had been  inoculated with 1-1.6 x 10³ CFU/ml of E. coliO157:H7.  No viable E. coli O157:H7 cells were detected in the seed suspension that contained $0.4% of calcinated calcium after an overnight incubation at 25EC.  Weissinger & Beuchat (2000) tested the use of 1.0% calcinated calcium in reducing Salmonella on alfalfa seeds.  They noticed that Salmonella was reduced by 2.88 logs.

Ozone Treatment

Ozone, the triatomic form of oxygen, is produced commercially by passing dry air or oxygen between high voltage electrodes that produce a corona discharge (Turner, 1983).  It is a powerful oxidant that is highly unstable and must be generated at the site of application (Turner, 1983).  Its principal use as an antimicrobial agent has been in the purification of water, though it is ineffective against surface contaminants or bacteria protected by organic material (Turner, 1983).  Naito et al., (1988) and Naito and Shiga (1989) used ozone and ozonated water as a treatment to decrease microbial levels on alfalfa seeds, beans, peas, grain and spices.  This treatment reduced the natural microflora by 1 to 3 logs without adversely affecting the sprouting of the seeds and beans.

            Hot Water Treatment

Hot water treatments have also been explored as a possible means for reducing pathogenic levels on seeds.  Soaking seeds in 54EC water for 5 to 10 min reduced, but did not eliminate S.Stanley populations (Jaquette et al, 1996).  Higher temperatures and longer times caused significant declines in germination (Jaquette et al., 1996).  Jaquette et al., stated that having such a narrow temperature range between treatment efficiency and seed injury makes relying on heat alone difficult on a commercial scale.

            Ethanol

Beuchat (1997) investigated the effectiveness of 80% ethanol in reducing salmonellae populations on alfalfa seeds.  The use of an 80% ethanol soak for 10 min reduced microbial populations by 3 logs (Beuchat, 1997).  In 1999, Taormina and Beuchat found that 30 to 70% ethanol decreased E. coli O157:H7 populations on alfalfa seeds, but also reduced the percent germination.

Irradiation

In preliminary studies, gamma irradiation has been shown,  to be an effective antimicrobial treatment for both seeds and sprouts (NACMCF, 1999).  E. coli O157:H7 and Salmonella were inactivated, without reducing seed viability when treatment with Cs¹³⁷ (5kGy) at a temperature of 5EC was used (NACMCF, 1999).

Other (active oxygen solution, trisodium solution, etc.)

Taormina and Beuchat (1999) investigated the use of both trisodium phosphate and active oxygen solutions as a way to decrease populations of E. coliO157:H7 on seeds.  The use of 4 % trisodium phosphate caused a significant reduction in the levels E. coli O157:H7, on seeds, to undetectable levels.  Treatment with active oxygen also caused a decrease in E. coli O157:H7 populations (Taormina & Beuchat, 1999; Weissinger & Beuchat, 2000).  Calcium hydroxide caused a 2.84 log₁₀ reduction of Salmonella in alfalfa seeds (Weissinger & Beuchat, 2000).  Beuchat et al. (2001) studied the effectiveness of ‘Fit’ a liquid prototype produce wash product, in reducing the level of Salmonella and E. coli O157:H7 on alfalfa seeds.  A 2.3 log₁₀ cfu/g reduction and  a greater than 5.4 log₁₀ cfu/g reduction in Salmonella and E. coli O157:H7 respectively, occurred when the inoculated alfalfa seeds were treated with this liquid prototype produce wash.  Park et al. (2000) investigated the efficacy of allyl isothiocyanate (AIT) in reducing E. coli O157:H7 on alfalfa seeds.  Allyl isothiocyanate is the result of hydrolysis of glucosinolates by myrosinase in cruciferous plants, including mustard and horseradish.  It was demonstrated that when dry alfalfa seeds were exposed to 50 :L of AIT for 24h at 37EC, the E. coli O157:H7 population on the seeds was reduced by 2 logs  (Park et al., 2000).  Park et al. (2000) also noted that the activity of AIT diminished with decreasing temperature.

The use of novel two-step,  organic acid/hypochlorite treatments have also been investigated for their effect on reducing E. coli O157:H7 populations from alfalfa seeds (Lang et al, 2000).  Lang et al. (2000) compared the efficacy of organic acid treatment (lactic acid or acetic acid) followed hypochlorite treatment versus the use of organic acid treatment alone in reducing E. coli O157:H7 in reducing E. coliO157:H7 on alfalfa seeds.  Lang et al. (2000) showed that the use of successive organic acid and hypochlorite treatments were more effective than organic acid treatments alone in reducing E. coli O157:H7 on alfalfa seeds.

The inability of disinfectants to completely eliminate Salmonella and E. coliO157:H7 from seeds is probably due to the nature of the seeds, more than to the efficacy of these chemicals or the durability of the pathogen.  The natural surface of an alfalfa seed contains numerous crevices that could harbour many types of microbial cells (Taormina & Beuchat, 1999).  E. coli O157:H7 may lodge within these cracks and crevices and therefore be protected.   Beuchat (1997) stated that the inaccessibility of Salmonella cells (in crevices and between the cotyledon and testa of seeds) to lethal concentrations of chemicals are thought to be the reason for their lack of effectiveness.

Water Usage during Sprouting

Water is a main component used throughout the spouting process.  The survey conducted by CFIA showed that 30 out of the 38 sprout manufacturing facilities used water from a municipal aqueduct, 4 of which treated and/or tested the quality of the municipal water.  The other 8 sprout manufacturers used well water.  The majority (7/8) of those sprout manufacturers using well water tested and/or treated the water before use, however, the analyses were not performed on a frequent or regular basis.

The water quality can influence the potential for contamination of sprouted seeds/beans with Salmonella and     E. coli O157:H7, since it can be a carrier of both these pathogens (NACMCF, 1998).  Water may also become contaminated directly or indirectly by improperly managed human and/or animal waste.  Human contamination may occur from improperly designed or  malfunctioning septic systems, as well as from sewage treatment facility discharges such as combined sewer and storm sewer overflows.

Water can also be a useful tool for reducing potential microbial loads through the rinsing/washing of sprouted seeds/beans, but one must be careful because if untreated, it may serve as a source of contamination or cross-contamination.  Therefore, the use of water containing antimicrobial chemicals may help maintain quality of the water and reduce contamination from Salmonella and E. coli O157:H7.

Problems may also arise from the re-use of water during the sprouting process, since this may result in cross-contaminations, as well as the build-up of microbial loads in and on the sprouted seeds/beans.  To minimize Salmonella and E. coli O157:H7 cross-contamination via water during the entire sprouting process, sprout producers should consider changing the water and cleaning/sanitizing water contact surfaces (eg: buckets, rotary drums) as often as necessary.  Sprout manufacturers using wells should have their water tested to make sure that it is of potable quality.

Harvesting of Sprouted seeds/beans

During harvesting of sprouted seeds/beans, contamination can occur from water rinses, workers and harvesting equipment.  Any surfaces that come into contact with the sprouted seeds/beans during harvesting can potentially be source(s) of contamination.  The majority of sprout manufacturers surveyed (30/38) rinsed/washed the sprouted seeds/beans at the end of the process.  Out of 38 sprout manufacturers, 24  let the water drain off the sprouts by gravity, while 14 used a centrifuging machine to remove the excess water. Approximately 11% (4/38) of the sprout manufacturers carried out microbiological testing of their finished product.

During harvesting, to aid in minimizing the exposure to Salmonella and E. coliO157:H7 as well as other pathogens, sprout manufacturers should consider cleaning/disinfecting buckets, tables and any material used in the harvesting between crops.  The general cleanliness of the harvesting equipment will help in reducing the risk of cross-contamination of the seeds/beans and sprouted seeds/beans.  It is also important to keep the harvest storage facility clean, as well as to keep the delay between harvesting and packing as short as possible.  After harvest, almost all sprout manufacturers surveyed (36/38) refrigerated the sprouted seeds/beans in a cold room held at 4EC.

Packing Facility

An improperly maintained packaging facility  may attract rodents and pests, which in turn may expose the sprouted seeds/beans to Salmonella and E. coliO157:H7.  Daily cleaning in and around the packing facility will also reduce the risk of rodent entry into the facility.

Refrigerated Storage of Sprouted Seeds/Beans Prior to Transportation

After packing, storage of the product in a refrigerated area can minimize growth and proliferation of remaining Salmonella and E. coli O157:H7.  Almost all manufacturers (36/38) refrigerated the sprouted seeds/beans in a cold storage room at 4EC.  Sprout manufacturers should not rely solely on the refrigeration of the final product for controlling Salmonella and E. coli O157:H7.  As mentioned in section 4.2.2.2, if Salmonella and/or E. coli O157:H7 are present on/in the seeds, the level of contamination may be very high (up to 10⁷ cfu/g) at the end of the sprouting process.  Refrigeration of the finished product can slow the growth of the above pathogens, but it will not reduce its population.  The potential for foodborne illness would therefore still exist.  In addition, at retail, raw sprouted seeds/beans are usually kept in the vegetable section where it is not always possible to refrigerate them.

Transportation

Microbial cross-contamination from other foods and food sources and contaminated surfaces may occur during the loading, unloading, storage and transportation of sprouted seeds/beans.  The regular cleaning, sanitizing and monitoring of transportation containers and vehicles will help reduce the exposure of sprouted seeds/beans to contamination by Salmonella and E. coli O157:H7.  The use of refrigerated vehicles, such as trucks,  will also aid in minimizing the growth and proliferation of any remaining Salmonella an/or E. coli O157:H7.    Out of 38 sprout manufacturers, 19 used refrigerated trucks to deliver their sprouts to retail, though generally, the transportation vehicles were not sanitized on a regular basis as per a written program.   Separation of sprouted seeds/beans from other foods and non-food sources during transport to the retailer will also help reduce exposure to various contaminants.

4.2.4    Consumption of Sprouted Seeds/Beans in Canada

Data reported in this section came from the Federal-Provincial Nutrition Survey conducted between 1991 and 1994, in Nova Scotia, Quebec, Saskatchewan and Alberta, on 8167 Canadian adults aged 18 to 74.  The following data on all raw sprouted seeds/beans consumed are based on responses from 160 individuals.   This survey may be considered a ‘snapshot’ in the sense that it represents a single day’s intake, (i.e., the fraction of the population who might consume sprouts on a given day).  This survey does not bear any relationship to the fraction of the population who might consume sprouts one or more times over a longer period of time (eg: one year).  Two types of sprouts were used to represent this food commodity: Foodcode 2090, Alfalfa seeds, sprouts or seed raw and Foodcode 2121, Beans, Mung, mature seeds, sprouted or seed raw.  It is important to keep in mind that the results here do not make a distinction between consumption at or away from home.

Ignoring gender, an estimated 2.5% (2.1, 2.8) of Canadian adults aged 18 to 74 years, consume raw sprouts of any kind, 1% (0.8, 1.2) consume raw alfalfa sprouts and 1.3% (1.1, 1.6) of Canadian adults consume raw mung bean sprouts on any given day.  The elderly (75 years and above) and  the young (less than 18 years) were not covered in this nutrition surveys.  Therefore, it is not known whether the consumption frequency of the above mentioned susceptible group resembles or differs from the consumption frequency mentioned in this section .  There is also some uncertainty when extrapolating from the consumption characteristic of sprouts in the early 1990’s to the present.  Again, it is important to note that these estimates represent only a fraction of the Canadian population in the provinces of Nova Scotia, Quebec, Saskatchewan and Alberta who consumed sprouts on a given day selected at random.

The consumption patterns of different ethnic groups was not covered in this nutrition survey.  No further information regarding industry production, production disappearance at retail is known at this time.

4.2.5    Conclusion on Exposure Assessment

Exposure of the seeds/beans and sprouted seeds/beans to Salmonellaand/or E. coli O157:H7 can occur at various stages during seed production  and sprout manufacture (section 4.2.1.1 & 4.2.1.2).  Once present on the seeds, these pathogens can survive for extended periods of time (section 4.2.2.1).  If present on the seeds/beans during the sprouting process, Salmonella and E. coli O157:H7 can proliferate to very high levels that can cause foodborne illness (section 4.2.2.2).  Presently, little is known regarding the processes that seed producers and sprout manufacturers are taking to minimize exposure and proliferation of Salmonella andE. coli O157:H7 on/in seeds/beans and sprouted seeds/beans.  Seed producers may not necessarily know whether the seeds they produce are destined for sprouting, as well as not being aware of the potential health risks involved (section 4.2.3.1).  Data from CFIA’s 1999 survey of the 38 sprout manufacturing facilities, indicated that a large percentage of sprout manufacturers across Canada required improvements in their manufacturing practices, and that only a few are following good manufacturing practices in production of sprouted seeds/beans (section 4.2.3.2).

As indicated in section 4.2.4, in the early 1990’s, an estimated 2.5% (2.1, 2.8) of a representative Canadian adult population (ages 18 to 74 years) consumed raw sprouted seeds/beans on a randomly chosen day.  This single day consumption frequency is only appropriate when considering a single episode exposure to sprouted seeds/beans and cannot be used to extrapolate when estimating the annual consumption of sprouted seeds/beans in Canada.  There are still some unknowns regarding the likelihood of ingestion of Salmonella and E. coli O157:H7 via sprouted seed/beans because there is no data to determine the initial levels and numbers of Salmonella and E. coli O157:H7 frequency of exposure of seeds/beans and sprouted seeds/beans to Salmonella and E. coli O157:H7.

4.3       HAZARD CHARACTERIZATION

Based on the data obtained from the Salmonella and E. coli O157:H7 outbreaks, this section provides a qualitative description of the severity and duration of adverse effects that have resulted from the ingestion of Salmonella and E. coliO157:H7 via sprouted seeds/beans.  See Appendix A, for a more detailed description of the number of cases, age distribution and severity of illness pertaining to each outbreak.

4.3.1    Infectious dose

Salmonella spp.

The minimum number of Salmonella that will cause human illness from the consumption of sprouted seeds/beans is unknown.   In general, the infectious dose of Salmonella spp. is dependent on age and health of the host and strain differences, however, as low as 1-20 cells is known to have caused illness in humans (D`Aoust, 1999).  Aabo & Baggesen (1997) determined, through a Most Probable Number (MPN) analysis on a portion of the incriminated seed lot that was involved in the S. Newport outbreak in Denmark, that given an intake of 25-100g of alfalfa sprouts per consumer, the infectious dose would be in the range of 5 to 460 cfu/g.

E. coli O157:H7

The minimum number of E. coli O157:H7 that will cause human illness from the consumption of sprouted seeds and beans is unknown.  The infectious dose in other foodborne outbreaks is also unknown, but is considered to be low (10 to 1000) (Doyle and Padhye, 1989; USFDA, 1992b; Farber 1989).

The low infectious dose for both Salmonella and E. coli O157:H7 suggest that growth in foods is not necessary for illness to occur.  In the case of sprouted seeds/beans, if the seeds/beans (initial product) have been exposed and thereby contaminated with Salmonella and/or E.coli O157:H7, the finished product (sprouted seeds/beans) may have high levels, up to 10⁷ cfu/g, of these pathogens (section 4.2.2.2).

4.3.2    Symptoms and Pathogenicity:  Severity of illness

Salmonella spp.

            General Description

Human infection with non-typhoid salmonellae commonly results in enterocolitis (D’Aoust, 1997). Illness is caused by the penetration and passage ofSalmonella organisms from the gut lumen into the epithelium of the small intestine, resulting in inflammation of the small intestine.  The severity of nontyphoid Salmonella infection (known as salmonellosis) varies with the number ingested, and the susceptibility of the individuals.   The less severe cases of salmonellosis may go unreported and therefore undocumented (CDC, 1998).

Acute symptoms of salmonellosis include nausea, vomiting, abdominal cramps, and non-bloody diarrhea (USFDA, 1992a). The incubation period is between 8 to 72h after ingestion of the pathogen (D’Aoust, 1989; D’Aoust, 1997).  The symptoms may persist for 3-4 days and usually resolve within 7 days.  The symptoms may be prolonged depending on host factors, ingested dose and/or strain characteristics.  Approximately 50% and 10-20% of individuals will continue to excrete Salmonella 2 – 4 weeks and 4-8 weeks, respectively, after remission.

Potential chronic sequelae including aseptic reactive arthritis, Reiter’s syndrome and ankylosing spondylitis may follow 3-4 weeks or longer after the onset of acute symptoms (D’Aoust, 1997b; USFDA, 1992a).  In severe cases of Salmonellainfection, hospitalization is required (CDC, 1998).

Related to Outbreaks Linked to Sprouted Seeds/Beans

Data obtained from outbreak of salmonellosis linked to the consumption of sprouted seeds/beans, revealed that the majority of symptoms reported included gastrointestinal infection and diarrhea (see Appendix A).  A few cases required hospitalisation and one death, an elderly patient, has been reported due to illness caused by the consumption of alfalfa sprouts contaminated with either S.Montevideo or S. Meleagridis (NACMCF, 1999; Johnson, 1999).

E. coli O157:H7

General Description

E. coli O157:H7 causes  foodborne infection.  Three principal syndromes have been linked to the consumption of food commodities contaminated with E. coli O157:H7.  The first syndrome, hemorrhagic colitis (HC) is characterized by a sudden onset of severe abdominal cramps, followed within 24 h by diarrhea that is initially watery and most often becomes grossly bloody (Riley et al., 1983; Doyle & Padhye, 1989; Willshaw, 2000).  Occasionally, vomiting may occur.  The illness is self-limiting and may last from 2 to 9 days, with a median of 4 days (Doyle & Padhye, 1989).  A few individuals may only experience watery diarrhea (USFDA, 1992b).  The second syndrome, hemolytic uremic syndrome (HUS), is a leading cause of acute renal failure and hemolytic anemia in children (Farber 1989; Karmali et al., 1983).  It results from blood clots plugging the convoluted tubules in the kidney, which in turn can cause an accumulation of waste products in the blood (Doyle & Cliver, 1990).  Individuals with HUS may permanently lose kidney function, require dialysis and blood transfusions, and may also develop central nervous system diseases characterized by frequent seizures and prolonged coma.  In the worst cases, death may occur (Doyle & Padhye, 1989).  Thrombotic thrombocytopenic purpura (TTP), the last syndrome, is similar to HUS, though central nervous system involvement is the dominant feature (Ramsey and Neil, 1986).  Symptoms of TTP include, fever, a decrease in blood platelets, and purpura, purplish or brownish discolouration of the skin, caused by hemorrhage into the tissues (USFDA, 1992b; Willshaw, 2000). Though infrequent, brain damage may occur (Doyle & Cliver 1990; Doyle & Padhye, 1989; USFDA, 1992b).  Illness due to E. coli O157:H7 infection or intoxication may be fatal for the young, the elderly and immunocompromised.

Related to Outbreaks Linked to Sprouted Seeds/Beans

In several of the E. coli O157:H7 outbreaks linked to sprouted seeds/beans, most patients, some of which required hospitalization, had developed bloody diarrhea.  In total, 6 deaths, all in Japan, have resulted from the consumption of white radish sprouts contaminated with E. coli O157:H7 (LCDC, 1997; Watanabe, 1999; Gutierrez, 1997).

4.3.3    Susceptibility Factors

            Salmonella spp.

Persons of all ages are susceptible to illness when exposed to Salmonella, but symptoms are most severe in the elderly, infants and immunocompromised (USFDA, 1992a).  AIDS patients suffer more frequently (20-fold more) and from more recurrent episodes of salmonellosis, as compared to the general population (USFDA, 1992a).  The fatality rate of most forms of salmonellosis is less than 1% (USFDA, 1992a).  S. Enteritidis has a mortality rate of 3.6% in hospital/nursing homes, with the elderly being most at risk (USFDA, 1992a).

In the case of Salmonella outbreaks for which an age distribution was reported, the median age was found to be between 30 and 40 years (see Appendix A).  For the outbreaks reporting a gender distribution, the ratio was approximately 1:1, with the female population having a slightly higher number of cases (see Appendix A).

E. coli O157:H7

All people are believed to be susceptible to the HC syndrome, though the young appear to be particularly so..  Up to 16% of HC victims may develop HUS (Karmali et al., 1983; USDA, 1992b; Cassin et al., 1998), which in turn can lead to kidney failure (Farber 1989).  TTP can have a mortality rate of as high as 50% in the elderly (USFDA, 1992b).

Most outbreaks of E. coli O157:H7 occurred in Japan (5 outbreaks).  Of those cases in North America, for which an age distribution was mentioned, the median age was found to be between 31and 35 years while for the gender distribution, slightly greater than 50% of the cases were female (see Appendix A).

Scientific literature does not make mention of the difference between males and females in their susceptibility to illness due to salmonellosis or E.coli O157:H7 infection.  Therefore, the age and gender distribution, may likely represent the demographic distribution of the population that eat sprouts.

4.3.4    Diagnosis and Treatment

            Salmonella spp.

Salmonellosis is diagnosed through serological identification of cultures isolated from the stools of patients infected with Salmonella.  Human infection resulting from Salmonella is usually self-limiting.  Supportive therapy (i.e.; fluid and electrolyte replacement) for uncomplicated cases may be the only treatment required for a successful recovery.  The use of antibiotics in such cases is unnecessary,. and may even be detrimental, because it tends to prolong the carrier state and the intermittent excretion of Salmonella.  The latter is due to the fact that the native gut flora, which normally competes with salmonellae, is repressed (D’Aoust, 1997).

E. coli O157:H7

HC is diagnosed by isolating E. coli O157:H7 from diarrheal stool samples obtained from patients infected with E. coli O157:H7.  Another method of diagnosing HC from diarrheal stool samples is by directly testing for the presence of verotoxin (VT).  Confirmation can be obtained by isolation of E. coli O157:H7 from the incriminated food(s).

Infection with E.coli O157:H7 is usually self-limiting (Willshaw, 2000).  The duration of symptoms do not appear to be reduced by treatments involving antimicrobial therapy or antimotility agents (Dundas & Todd, 2000; Willshaw, 2000).  The use of such treatments may increase the risk of progression from HC to HUS or TTP (Dundas, 2000; Willshaw, 2000).  Patients with HC should be given plenty of fluids to prevent dehydration.  In the case of HUS, blood transfusions or kidney dialysis may be required.

4.3.5    Conclusions on Hazard Characterization

Though the incidence of sprouted seeds/beans contaminated with Salmonella and/or E. coli O157:H7 is unknown, the general population, especially the elderly, the young and the immunocompromised can become ill after the consumption of contaminated sprouted seeds/beans.  The elderly, the young and the immunocompromised are most susceptible to further complications that may lead to death.  The probability of death due to the consumption of contaminated sprouted seeds/beans is unknown.

The data obtained from the Salmonella and E.coli O157:H7 outbreaks linked to sprouted seeds/beans showed that the median age was 30-40years for those cases reporting an age distribution, while the gender distribution was approximately 1:1 with the number of females slightly higher.

4.4       RISK CHARACTERIZATION

This section will integrate the Hazard Identification, Hazard Characterization and Exposure Assessment sections in order to provide qualitative estimates of the likelihood and severity of the adverse effects which could occur in the Canadian  population due to the consumption of contaminated sprouted seeds/beans.  Where appropriate, it will include a description of the uncertainties associated with these estimates.

4.4.1    Introduction

Based on the available epidemiological investigations it is suggested that seeds are the most likely source of Salmonella and/or E. coli O157:H7 contamination in the sprouting industry (Puohiniemi et al., 1997; CDC, 1997; Mahon et al., 1997; Itoh et al., 1998).   E. coli O157:H7 is commonly found in soil and water contaminated with faecal matter from domestic or wild animals.  Sources of salmonellae include contaminated water, soil, insects, animal feces, etc.  As mentioned in section 4.2.1.1, seeds are raw agricultural commodities, and therefore may be exposed to high levels of these pathogens from the field via contaminated agricultural water, improperly managed/treated animal manure, contact with wild animals, etc.  Exposure of seeds and beans to Salmonella and E. coli O157:H7 may also occur during harvesting, cleaning and transportation.  Cross-contamination from other contaminated seed/bean lots may also expose these raw agricultural commodities to both Salmonella and E. coli O157:H7.  In the sprouted seeds or beans, these pathogens can reach very high levels due to the conditions used for sprouting since they are optimal for their growth and proliferation (section 4.2.2.2).  Even if the initial Salmonella and E. coli O157:H7 contamination levels on or in the seed/beans are  low, the environmental conditions and nutrients present during sprout production can support their growth to potentially high levels.

As shown in section 4.2.3.1 and 4.2.3.2,  there are currently no, or very few, inherent step(s) used in the production of seeds/beans and sprouted seeds/beans to reduce or eliminate Salmonella and E. coli O157:H7.  Salmonella and E. coli O157:H7 are also capable of survival and/or proliferation in sprouted seeds/beans at the retail level.  Therefore, the simple rinsing of the sprouted seeds/beans prior to consumption is not an adequate treatment for reducing or removing these pathogens.

4.4.2    Likelihood of Illnesses

            The likelihood of illness of the Canadian population due to the consumption of contaminated sprouted seeds/beans is primarily dependent on the patterns of consumption of this food commodity in Canada.  In Canada, alfalfa as well as other sprouted seeds/beans, such as radish, mustard, onion and bean have become popular and are available both as grocery items and in foodservice operations.  For example, sprouted seeds/beans may often be eaten raw in sandwiches and salads.  Bean sprouts, in particular, are thought to be a prominent feature of Asian-Canadian cuisine.  Sprouted seeds may pose a higher risk than sprouted beans, since sprouted beans may be cooked briefly, such as in stir-fry vegetable dishes, before they are eaten.

The single day frequency and consumption characteristics of sprouted seeds/beans, reported in section 4.2.4, are only appropriate for considering a single episode exposure to sprouted seeds/beans.  The survey revealed that in the early 1990’s, an estimated 2.5% (2.1, 2.8) of healthy adults (from Nova Scotia, Alberta, Quebec and Saskatchewan) aged 18 to 74 years, excluding the young and the elderly, consumed raw sprouted seeds/beans of any kind on a randomly chosen day.  The potential for consumers to fall ill from a single serving of sprouted seeds/beans contaminated with Salmonella and/or E.coli O157:H7 cannot be determined from this data.  It is also important to note that there are some uncertainties involved when extrapolating from four provinces to all of Canada, as well as from extrapolating from the early 1990’s to the present.

4.4.3    Canadian Outbreaks

Salmonella outbreaks

In late 1995 and early 1996, the province of British Columbia and the state of Oregon experienced an outbreak due to S. Newport that was associated with the consumption of alfalfa sprouts.  Approximately 133 cases were culture-confirmed (Van Beneden et al., 1999).  Subsequent to this outbreak, the province of Quebec reported 60 epidemiologically-linked cases of S. Newport infections due to contaminated alfalfa sprouts (Foodborne Outbreaks in Canada Linked to Produce, In Press).  Contaminated alfalfa sprouts were responsible for one further Canadian outbreak in the fall of 1997 (Buck et al., 1998).  In this episode, 78 confirmed cases of S. Meleagridis infections were reported in three provinces (Alberta, Ontario, Saskatchewan).  During August and September of 1999, 61 cases of S. Java in Alberta, Saskatchewan, British Columbia, and Manitoba were linked to the consumption of alfalfa sprouts.  Between April to June 2000 in Alberta and Saskatchewan, epidemiological investigations indicated that at least 10 cases of S.enteriditis PT11 were associated with the consumption of mung bean sprouts.  In February 2001, contaminated bean sprouts were responsible for 33 confirmed cases of S.Enteritidis.  No deaths due to contaminated sprouts have been reported in Canada.

E. coli O157:H7 outbreak

In Canada, there has been no reported cases of E. coli O157:H7 due to the consumption of sprouted seeds/beans, however, outbreaks have occurred in the United States and other parts of the world.  See Appendix A for specific outbreaks.

4.4.4    Product Identification

Large-scale control measures are necessary for the identification of seed, as well as sprouted seed/bean shipments potentially involved in outbreaks, to ensure an effective traceback to the implicated seed producers.  So far the identification system makes it very difficult to link illnesses or outbreaks to specific seed lots.  Better information, in the form of documentation and written records, on the source of seeds/beans, sprouting process and operational controls, may help define more accurately the potential vehicles of transmission, as well as facilitate traceback and recall.

4.4.5    Methodology

4.4.5.1  Detection of Salmonella

Experience in the laboratory has shown that the probability of recovery of salmonellae from dry seeds, using the standard culture method, MFHPB-20 (HPB, 1999), is very low.  This likelihood of recovery can be improved if the seeds are first ground for 2 min at high speed in a blender, or sprouted under sterile conditions. The protocol for germinating alfalfa and clover seeds such as radish, clover, mustard etc., prior to analysis should be done in parallel with the analyses of non-germinated seeds.  The procedure for germinating seeds is described in MFHPB-20A (Amendment to MFHPB-20 for the Analysis of Seeds Used to Manufacture Sprouts).  For more details on the standard culture methods of salmonellae, see the Compendium of Analytical Methods, Volume 2 ( MFHPB-20).

4.4.5.2  Detection of E. coli

The MFHPB-19 (HPB, 1999) method is applicable to the enumeration of coliforms, fecal coliforms and E. coli in foods, including seeds used to manufacture sprouts.  This method has been shown to produce satisfactory results with naturally-contaminated foods for the detection of coliforms, fecal coliforms and E. coli.  See the Compendium of Analytical Methods Volume 2, MFHPB-19, for more details on the standard culture methods of E. coli.  Procedures for detecting E.coli O157, including E. coli O157:H7 from foods are outlined in the Compendium of Analytical Methods, Volulme 3, MFLP-80, -81, 86, 87, 89, 90, 91 and MFLP-93 (HPB, 1999).

4.4.6    Conclusions on Risk Characterization

In sprouted seeds/beans, the microbial hazards of immediate concern have been identified as Salmonella and E. coli O157:H7.  When seeds/beans contaminated with Salmonella and E. coli O157:H7 are used to manufacture sprouted seeds/beans, the conditions prevailing during the process are favourable for growth and proliferation of these pathogens (section 4.2.2.2).  Populations exceeding 10⁷ CFU of E. coli O157:H7 per gram of sprouts can be found.  The consumption of raw sprouts manufactured from seeds/beans contaminated with Salmonella and/or E. coli O157:H7, may expose consumers to a high dose that will cause illness.  Since 1996, 7 reported deaths, Salmonella (1) and E.coli O157:H7 (6),  have been attributed to contaminated sprouted seeds/beans.  Susceptible groups in the population such as young children, the elderly and the immunocompromised are considered to be at a higher risk of acquiring infection and are more likely to develop complications.

In North America, as in other parts of the world, sprouted seeds/beans may often be consumed raw.  Should the seeds/beans or sprouted seed/beans be exposed to Salmonella and/or E. coli O157:H7 anywhere during the production chain, rinsing will not be successful in removing these pathogens.  Therefore, there is little that can be done at the consumer level to reduce the risk of infection linked to consumption of raw sprouted seeds/beans.  The present system of sprout manufacturing cannot guarantee the absence of Salmonella, E. coli O157:H7 or any other pathogen.  Thus, there is a strong need to address control measures that will focus on seed quality, sprouting process, sanitation and growing practices, final product testing and storage/transportation of the final product.

5.         REFERENCES

Aabo, S. and Baggesen, D.L. Growth of Salmonella Newport in Naturally contaminated alfalfa sprouts and estimation of infectious dose in a Danish Salmonella Newport outbreak due to alfalfa sprouts    In: Salmonella and Salmonellosis (Proceedings), 587-589, Ploufragan, France, May 1997.

Abeyta, C., Jr., and Wacol, M.M.  1988.  Potential sources of Aeromonas hydrophila.  J. Food Protect.  49: 643-646.

Andrews, W.H., et al. 1979. Bacteriological survey of sixty health foods.  Appl. Environ. Microbiol. 37:559-566.

Andrews W.H., et al. 1982.  Microbial hazards associated with bean sprouting.  J. Assoc. Off. Anal. Chem. 65:241-248.

Bagley 1985.  Habitat association of Klebsiella species.  Infect. Control 6:52-58.

Bari M.L., 1999.  Inhibition of growth of Escherichia coli O157:H7 in fresh radish (Raphanus sativus L.) sprout production by calcinated calcium.  J Food Prot. 62: 128-132.

Beuchat, L.R.  1997.  Comparison of chemical treatments to kill Salmonella on alfalfa seeds destined for sprout production.  Int.  J.  Food  Microbiology. 34:329-333.

Beuchat L.R. et al. 1990.  Presence and public health implications of Listeria monocytogenes on vegetables.  In: Foodborne Listeriosis, Topics in Microbiology.  ed. Miller, A.J., Smith, J.L., and Somkuti, G.A.  Amersterdam: Elsvier.  pp. 175-181.

Beuchat L.R. et al.  2001.  Comparison of chlorine and a prototype produce wash product for effectiveness in killing Salmonella and Escherichia coli O157:H7 on alfalfa seeds.  Journal of Food Protection. 64:152-158.

Brown C and R.J., Seidler et al., 1973.  Potential pathogens in the environment: Klebsiella pneumoniae, a taxonomic and ecological enigma.  Appl Microbiol.  25:900-904.

Buck et al. 1998.  Would you like a little Salmonella with your sandwich?  In: Program and abstracts of the 47^th Annual Epidemic Intelligence Service Conference, International Night Atlanta, CA

Campden Chorleywood Food Research Association., U.K.  Guidelines for the hygienic manufacture, distribution and retail sale of sprouted seeds with particular reference to mung beans.  Technical Manual No.25. September 1989.

Cassin MH, et al. 1998.  Quantitative risk assessment for Escherichia coli O157:H7 in ground beef hamburgers.  Intl J. Food Microbiology. 4 1:21-44.

Centers of Disease Control and Prevention (CDC).  1997.  FoodNet population survey atlas of exposures.  7/96-6/97.

Centers for Disease Control and Prevention. (CDC) Salmonellosis. 1998.

Communicable Disease Surveillance and Response (CSR).  1996.  Enterohaemorrhagic Escherichia coli in Japan – Update.  In: Disease outbreaks reported 28 & 29 August 1996.

D’Aoust, J-Y.  1989.  Salmonella.  In: Foodborne Bacterial Pathogens.  M.P. Doyle (ed), Marcel Dekker Inc., pp: 341-351.

D’Aoust, J-Y. 1997a.  Dynamics of Salmonella spp. in the global food chain.  In:Progress in Microbial Ecology.  M.T. Martins et al. (eds.)  Brazilian Society for Microbiology, Brazil.  pp. 461-466.

D’Aoust, J-Y. 1997b.  Salmonella species.  In: Food Microbiology: Fundamentals and Frontiers.  M.P. Doyle, L.R. Beuchat and T.J. Montville (eds.) American Society for Microbiology, Washington, D.C.  pp. 138 – 159.

D’Aoust, J-Y.  1999.  Salmonella.  In: The Microbiological Safety and Quality of Food, Volume II.  B.M. Lund et al. (ed),  Aspen Publishers, Inc.  Chapter 25.

D’Aoust, J-Y.  2000.  Foodborne salmonellosis: current international concerns.  In: Proceedings of the Second NSF International Conference on Food Safety: Preventing Foodborne Illness Through Science and Education, October 11-13 2000 Savannah, Georgia, USA

Delaquis, P.J. et al.  1999.  Disinfection of mung bean seed with gaseous acetic acid.   J. Food Prot.  62: 953-957.

Doyle M.P and D.O. Cliver.  1990.  Escherichia coli.  In: Foodborne Diseases. D.O. Cliver (ed),  Academic Press Inc.  pp: 213-215.

Doyle M.P. and V.V. Padhye.  1989.  Escherichia coli. In: Foodborne Bacterial Pathogens.  Ed: Marcel Dekker Inc.  pp: 262-270.

Dundas, S. & W.T., Todd.  2000.  Clinical presentation, complications and treatment of infection with verocytoxin-producing Escherichia coli. challenges for the clinician.  J. Appl. Microbiol.  88: Suppl:24S-30S.

Dychdala, G.R. 1983.  Chlorine and chlorine compounds.  In: Disinfection, Sterilization and Preservation.  Seymour S. Block (ed.), 3^rd edition.  pp: 157-182.

Farber, J.M.  1989.  Foodborne pathogenic microorganisms: characteristic of the organisms and their associated diseases I: Bacteria.   J. Can. Inst. Food Sci. Technol. 22:311-321.

Farber, J.M.  and P.I., Peterkin.  1991.  Listeria monocytogenes, a food-borne pathogen.   Microbiol Rev 55:476-511.

FSNET.  August 30, 1996.  Enterohaemorrahgic E. coli infection – Japan.

FSNET.  February 22, 2001.  Sprout grower shut down by Salmonella.

Food Safety and Security.  February 2001.  Bean sprouts linked to Dutch Salmonellaoutbreak. [copyright 2001 PJ Barnes & Associates

Griffin and Tauxe.  1991.   The epidemiology of infections caused by Escherichia coliO157:H7, other enterohemorrhagic E. coli, and the associated hemolytic uremic syndrome.   Epidemiol Rev.  13:60-98.

Glynn et al., 1998.  When health food isn’t so healthy- An outbreak of salmonellaserotypes Anatum and Infantis infections associated with eating contaminated sprouts, Kansas and Missouri, 1997.  In: Program and Abstracts of the 47^th Annual Epidemic Intelligence Service Conference.  Atlanta, GA.

Gutierrez, E.  1997.  Japan prepares as O157 strikes again.  Lancet.  349:1156.

Hara-Kudo, Y.et al.,  1997. Potential hazard of radish sprouts as a vehicle of Escherichia coli O157:H7. J. Food Prot. 60: 1125-1127.

Highsmith and Jarvis.  1985.  Klebsiella pneumoniae: selected virulence factors that contribute to pathogenicity.  Infect. Control. (2):75-77.

Health Canada – Bureau of Microbial Hazards (BMH).  2001.  Personal Communication

Health Canada – Population and Public Health Branch (formerly LCDC). 2000.  Infectious Diseases News Brief, April 28 2000: Salmonella Enteritidis: California.

Health Protection Branch (HPB).  1999.  Compendium of analytical methods. Volume II.

Health Protection Branch (HPB).  1999.  Compendium of analytical methods. Volume III.

Inami, G.B. and S.E., Moler,  1999.  Detection and isolation of Salmonella from naturally contaminated alfalfa seeds following an outbreak investigation. J. Food Prot.  62: 662-664.

Ingram DT, Kantor MA, Meng J. Survival of E. coli O157:H7 during sprouting of inoculated alfalfa seeds. Int Assoc Milk Food Environ Sanit Ann Mtg, Prog and Abst Book, August16-19, 1998, Nashville, TN. p. 36.

International Sprouts Growers Association (ISGA).  Sanitary guidelines for the growing & packing for sale of fresh sprouted beans & seeds.  Copy revised 20 November 1996.

Itoh, Y., et al., 1998.  Enterohemorrhagic Escherichia coli O157:H7 present in radish sprouts. Appl. Environ. Microbiol. 64:1532-1535.

Janda, J.M., and P.S., Duffey.  1988.  Mesophilic aeromonads in human disease: current taxonomy, laboratory identification and infectious disease spectrum.  Rev. Infect. Dis. 10:980-97.

Jaquette et al.  1996.  Efficacy of chlorine and heat treatment in killing SalmonellaStanley inoculated onto alfalfa seeds and growth and survival of the pathogen during sprouting and storage.  Appl. Environ. Microbiol. 62: 2212-2215.

Jernghlinchan, J and K, Saitanu.  1993. The occurrence of Salmonellae in bean sprouts in Thailand.  Southeast Asian J. Trop. Med. Public Health, 24:114-118.

Joce, R. et al. 1990.  A national outbreak of Salmonella Gold-Coast.  Commun Dis Rep CDR Rev 4:3-4.

Johnson, E.A.  1990.  Bacillus cereus Food Poisoning.  In: Cliver D.O (ed). Foodborne Diseases.  Academic Press, Inc.  San Diego, CA.  pp.128-134.

Johnson, K.  1999.  Outbreaks of Salmonella and E. coli O157 infections from Sprouts – US 1995-99.  In: Sprout Summit: Best practices & recommendations for the production of safer sprouts from seeds.  November 15-16, 1999.  Chicago, Illinois.

Karmali MA. 1989.  Infection by verocytotoxin-producing Escherichia coli. Clin. Microbiol. Rev.  2:15-38.

Kapperud, G.  1991.  Yersinia enterocolitica in food hygiene. Int. J.  Food Microbiol. 12: 53-65.

Kontiainen, S., et al.  1996.  High frequency of septicaemia in an outbreak of Salmonella Stanley infections.  Scand. J. Infect. Dis. 28:207

Krovacek, K., et al. 1992. Prevalence and characterization of Aeromonas spp. isolated from foods in Uppsala, Sweden.  Food Microbiol.  9:29-36.

Krovacek, K., et al. 1995. Isolation and virulence profiles of Aeromonas hydrophilaimplicated in an outbreak of food poisoning in Sweden. Microbiol. Immunol.  39:655-661

Laboratory Centre for Disease Control – Health Canada (LCDC).  1997.  Enterohemorrhagic Escherichia coli Infection.  In: Canadian Communicable Disease Report.  Volume 23-17, September 01, 1997.

Laboratory Centre for Disease Control (LCDC), Health Canada. 1999. Material data safety sheet – infectious substances: Aeromonas hydrophila.

Laboratory Centre for Disease Control (LCDC), Health Canada  1997.  Material Data Safety Sheet – Infectious Substances: Bacillus cereus.

Lang M.M. et al.  2000.  Efficacy of novel organic acid and hypochlorite treatments for eliminating Escherichia coli O157H7 from alfalfa seeds prior to sprouting.  Journal of Food Microbiology. 58: 73-82

Mahon B.E. et al. 1997.  An international outbreak of Salmonella infection caused by alfalfa sprouts grown from contaminated seeds.  J. infect. Dis.  175: 876-82.

Ministry of Health and Welfare of Japan.  1997.  National institute of infectious diseases and infectious disease control division.  Verocytotoxin-producing Escherichia coli (enterohemorrhagic E. coli) infection, Japan, 1996-June 1997.  Infectious Agents Surveillance Report 1997; 18:153-154

Montgomerie J.Z.  et al.  1970.  Klebsiella in fecal fora of renal transplant patients.  Lancet. 2:787-792.

Morbidity and Mortality Weekly Report.  August 15 1997.  Outbreaks of Escherichia coli O157:H7 infection associated with eating alfalfa sprouts – Michigan and Virginia, June-July 1997.46: 741-744.

Naito S. et al., 1988.  Studies on the utilization of ozone in food preservation. V. Changes in microflora of ozone-treated cereals, grain, peas, beans, and spices during storage.  J. Jap. Soc. Food Sci.  Technol.  35: 69-77 (Abstract in English)

Naito S. & Shiga. 1989.  Studies on Utilization of Ozone in Food Preservation. IX.  Effects of Ozone Treatment on Elongation of Hypocotyls and Microbial Counts of Bean Sprouts.  J. Jap. Soc. Food Sci. Technol.  36:181-188 (Abstract in English).

National Advisory Committee on Microbiological Criteria for Foods (NACMCF).  1998.  Microbiological safety evaluations and recommendations on fresh produce.  Food Control 10:321-347.

National Advisory Committee on Microbiological Criteria for Foods (NACMCF). 1999.  Microbiological safety evaluations and recommendations on sprouted seeds and beans.

O’Mahony, M.J.  et al.  1990.  An outbreak of Salmonella Saint-Paul infection associated with bean sprouts.  Epidemiol. Infect. 104: 229-235.

Park, C.E. and Sanders G. W.  1990.  Source of Klebsiella pneumonia in alfalfa and mung bean sprouts and attempts to reduce its occurrence.  Can. Inst. Food Sci. Technol. J. 23:189-192.

Park C.M. et al.  2000.  Efficacy of allyl isothiocyanate in killing enterohemorrhagic Escherichia coli O157:H7 on alfalfa seeds.  International Journal of Food Microbiology. 56:13-20

Patterson, J.E. and Woodburn, M. J.  1980.  Klebsiella and other bacteria on alfalfa and bean sprouts at the retail level.  J. of Food Sci. 45:492-495.

Pezzino, G. et al.  1998.  A multi-state outbreak of Salmonella serotypes Infantitis and Anatum – Kansas and Missouri, 1997.  Kansas Medicine.  98:10-12.

Piernas V., and J.P. Guiraud.  1997.  Disinfection of rice seeds prior to sprouting J.  Food Sci.  62:611-615.

Ponka A.,  et al.  1995. Salmonella in alfalfa sprouts.  Lancet 345:462-463

Portnoy B.L., et al., 1976.  An outbreak of Bacillus cereus food poisoning resulting from contaminated vegetable sprouts.  Am J. Epidemiol.  103:589-594.

Prokopowich, D. and G. Blank. 1991. Microbiological evaluation of vegetable sprouts and seeds. J. Food Prot. 54:560-562.

Puohiniemi, R.T., et al., 1997.  Molecular epidemiology of two international sprout-borne Salmonella outbreaks. J. Clin. Mocrobiol. 35:2487-2491.

Raevuori, M., et al.  1976. An outbreak of Bacillus cereus food poisoning in Finland associated with boiled rice. J. Hyg. Camb. 76:319.

Rafii F., et al., 1995.  Survival of Shigella flexneri on vegetables and detection by polymerase chain reaction J. Food Prot.  58:727-737.

Ramsey, P.G. and M.A. Neil.  1986.  Epidemiologic Notes and Reports Thrombotic Thrombocytopenic Purpura Associated with Escherichia coli O157: H7 — Washington.  Morbid. Mortal. Weekly Rep. 35:549-551.

Renwick S.A. et al. 1993.  Evidence of direct transmission of Escherichia coli O157:H7 infection between calves and a human. J. Infect Dis.168:792-93.

Riley et al. 1983.  Hemorrhagic colitis associated with a rare Escherichia coli serotype.  N. Engl. J. Med. 308: 681-685.

Sabota, J.M.  et al.  1998.  A new variant of food poisoning: Enteroinvasive Klebsiella pneumoniae and Escherichia coli sepsis from a contaminated hamburger.  Am J Gastroenterol.  93:118-119.

Schlech, W.F. 3d. et al.  1983.  Epidemic listeriosis–evidence for transmission by food.  N Engl J Med 308:203-206.

class=Section5>

Shooter R.A. et al., 1971.  Isolation of Escherichia coli, Pseudomonas aeruginosa, and Klebsiella from food in hospitals, canteens, and schools.   Lancet. 21:390-392.

Skrowronek, F., et al. 1998. Hygienic status and biogenic amine content of mung bean sprouts.   Z. Lebensm Unters Forsch. A. 207:97-100.

Splittstoesser, D.F., Queale, D.T and Andaloro, B.W.  1982.  The microbiology of vegetable sprouts during commercial production.  J. of Food Safety. 5:79-86.

SproutNet.  February 26 2001a.  Outbreak of S. Enteritidis PT4b infection in the Netherlands

SproutNet.  February 26 2001b.  Additional Salmonella cases sprouting up.

Stelma, Gerard N Jr., 1989. Aeromonas hydrophila In: Foodborne bacterial pathogens. Doyle, M. P. pp. 3-70.

Taormina P.J. and L.R. Beuchat.  1999.  Comparison of chemical treatments to eliminate enterohemorrhagic Escherichia coli O157:H7 on alfalfa seeds.  J.  Food Prot.  62:318-324.

Taormina P.J. and L.R. Beuchat.  1999.  Behaviour of enterohemorrhagic Escherichia coli O157:H7 on alfalfa sprouts during the sprouting process as influenced by treatments with various chemicals.

J.  Food Prot. 62:850-856.

Taormina P.J. et al., 1999.  Infections associated with eating seed sprouts: an international concern.  Emerg. Infect. Dis. 5:626-634.

Turner, F.J.  1983.  Hydrogen peroxide and other oxidant disinfectants.  In:Disinfection, sterilization and preservation.  Seymour S. Block (ed.), 3^rd edition.  pp: 240-250.

U.S. Food and Drug Administration Center for Food Safety and Applied Nutrition (USFDA-CFSAN). 1991.  Fodborne pathogenic microorganisms and natural toxins handbook: Aeromonas hydrophila..

U.S. Food and Drug Administration Center for Food Safety and Applied Nutrition (USFDA-CFSAN) 192a. Fodborne pathogenic microorganisms and natural toxins handbook: Salmonella spp.

U.S. Food and Drug Administration Center for Food Safety and Applied Nutrition (USFDA-CFSAN). 1992b. Fodborne pathogenic microorganisms and natural toxins handbook: Escherichia coli 0157:H7.

U.S. Food and Drug Administration Center for Food Safety and Applied Nutrition (USFDA-CFSAN). 1992c. Fodborne pathogenic microorganisms and natural toxins handbook: Listeria monocytogenes.

U.S. Food and Drug Administration Center for Food Safety and Applied Nutrition (USFDA-CFSAN). 1992d. Fodborne pathogenic microorganisms and natural toxins handbook: Bacillus cereus and other Bacillus spp.

Van Beneden C.A.et al., 1999.  Multinational outbreak of Salmonella entericaserotype Newport infections due to contaminated alfalfa sprouts.  JAMA 281:158-162.

Watanabe, Y. et al.  1999.  Factory outbreak of Escherichia coli O157:H7 infection in Japan.  Emerg. Infect. Dis. 5:424-428.

Wegener, H.C., et al.  1997.  An outbreak of human salmonellosis in Denmark caused by alfalfa sprouts.  In: Salmonella and Salmonellosis (Proceedings), 587-589, Ploufragan, France, May 1997.

Weissinger W. R. and L.R. Beuchat.  2000.  Comparison of aqueous chemical treatments to eliminate Salmonella on alfalfa seeds.  Journal of Food Protection. 63:1475-1482.

Willshaw, G.A., et al.  2000.  Escherichia coli.  In:  The microbiological safety and quality of food, Volume II  Aspen Publishers Inc.  pp: 1136-1177.

Wisconsin Department of Health and Family Services – Reference Center. 1999.  State lifts advisory on sprouts following abatement of Salmonella outbreak.

APPENDIX A:       Reported Outbreaks Associated with Sprouts, 1973-2001

APPENDIX B:      Treatment of Seeds/Beans/Sprouts to Reduce Total Microbial Load

(NB: Appendix B represents a summary of the research in the field of treatments of Seeds/Beans/Sprouts, for the reduction of foodborne pathogens, that have been published)

^* Radish seed were soaked in E. coli O157:H7-inoculated water containing different concentrations of calcinated calcium and was incubated overnight (O/N) at 25EC.