Sprouts and Foodborne Disease

Sprouts and foodborne disease

Food Safety Hygiene,  A bulletin for the Australian Food Industry, Food Science Australia, November 2000

The consumption of raw sprouted seeds has led to a large number of outbreaks of foodborne illness in a great number of countries. Alfalfa sprouts have been implicated most often, but other sprouts have also caused illness. Contaminated seed was the likely source of the pathogens in these outbreaks. While there have been no reported outbreaks from sprouts produced in Australia, seeds grown in Australia have been associated with outbreaks in other countries.

The first recorded outbreak of foodborne disease from the consumption of raw, sprouted seeds was in 1973 and this was from soy, mustard and cress grown in home-sprouting packs which were contaminated with Bacillus cereus. In 1988, there were large outbreaks of food poisoning in both the UK and Sweden from eating raw mung bean sprouts. Five different Salmonella serotypes were associated with the outbreaks, and 3 of these serotypes were detected in bags of mung bean seeds which had come from Australia.

In the following year, sprouts of cress contaminated with S. Gold-Coast were implicated in another outbreak in the UK. In the early 1990s, three outbreaks of salmonellosis from sprouts were identified in Finland. One outbreak of over 490 cases, in both Finland and Sweden, was due to S. Bovismorbificans in alfalfa sprouts and these seeds also had been imported from Australia. Several other outbreaks were reported in the 1990s in which salmonellae were implicated and alfalfa sprouts were the usual vehicle.

White radish sprouts used in school lunch programs were associated with a very large outbreak of Escherichia coli O157:H7 infections with more than 6000 cases in Japan in 1996. In the next year also, smaller outbreaks occurred in two Japanese cities from radish sprouts contaminated with E. coli O157:H7. Two outbreaks of E. coli O157 infection have also occurred in the US which have been linked to the consumption of alfalfa sprouts or an alfalfa/clover mixture. E. coli O157 is one member of a group of pathogenic E. coli known as enterohaemorrhagic E. coli or EHEC, for short.

Bacterial growth during sprouting

In the production of sprouts, seeds are soaked in water for a few hours and then held at 20-26°C for some days while being intermittently sprayed with water during germination and growth. These conditions, together with nutrients from the seeds and sprouts, provide good conditions for bacterial growth. The population of bacteria naturally on the seed increases rapidly. During the pre-germination soaking, the population may increase 10-fold or more. After two days of germination, even under hygienic conditions, bacterial numbers on sprouts will be up to millions per gram. The population frequently contains very large numbers of coliforms as well as many thermotolerant coliforms (ie able to grow at 44°C) that are normally associated with plant material. These may mistakenly be regarded as "faecal coliforms". Some (eg Klebsiella oxytoca andEnterobacter cloacae) also produce indole and may be mistaken for E. coli.

If a pathogen is present on the seeds, it too may grow extensively during germination. In the B. cereus outbreak in the 1970s, cells of B. cereus grew from about 100 per gram on the seeds to over a million per gram on the sprouts. Proliferation of Salmonella, E. coli O157 and Listeria monocytogenes has also been demonstrated to occur during germination and growth of sprouts. It is the considerable growth of bacteria during the normal process of sprouting that increases the risk of foodborne disease from sprouts compared to that from other vegetables.

Contamination of sprouts

While contamination with pathogens can occur from the water used in germination and sprouting or from the equipment used in the sprouting process, seed is the usual source. Seed can be contaminated in the field from agricultural water, from improperly composted manure, contaminated soil, and from feral or grazing farm animals. Seed can be contaminated from harvesting equipment, from residues in storage containers and screw-conveyors. Seed can be contaminated at the seed-mill where it is cleaned, graded and packed. Salmonellae have frequently been found in the dust that accumulates during seed cleaning. Good agricultural and handling practices will reduce seed contamination. However, total prevention of contamination is not possible.

Decontamination of seed

Considerable research activity has been directed recently towards seed decontamination. Washing seeds with water reduces bacterial contamination less than 10 fold. Adding 100-200 ppm of active chlorine to the wash water often has little extra effect in reducing contamination. Even with 1000-2000 ppm of active chlorine the reduction in contamination is sometimes only 10-100 fold. With 10,000-20,000 ppm of active chlorine at pH 6.8, reductions of 100 -1000 fold can be obtained. However, when alfalfa seeds, contaminated with around 480 cells of E. coli O157 per gram, were treated for 3 minutes with 20,000 ppm active chlorine, viable E. coli could still be recovered from the treated seed.

Other sanitizers which have been demonstrated to have some bactericidal effect include 4% trisodium phosphate, 1% hydrogen peroxide and various concentrations of alcohol. Heating seeds in hot water is effective in destroying contaminants like Listeria, Salmonella and EHEC. However, a compromise between decontamination and maintenance of seed germination is necessary irrespective of whether chemical or physical treatment is used.

Initially, when seed treatments with 100 ppm active chlorine failed to give large reductions in bacterial contaminants, it was thought that this might be due to rapid destruction of chlorine by the large organic load associated with seed. However, with solutions containing 2,000-20,000 ppm chlorine most of the active chlorine remains and bacteria still survive. Some bacterial cells are held in crevices on the seed and between the testae and cotyledons. These cells appear to be protected from chemical disinfectants. Surfactants (eg Tween 80) have been added with active chlorine, or used prior to the chlorine treatment, in an attempt to give better access of disinfectant chemicals to the bacterial cells. These approaches have not been very successful.

In recent studies with mung beans, treatment with gaseous acetic acid for 24 hours at 22°C resulted in destruction of very large numbers of cells of Salmonella spp., L. monocytogenes and E. coli O157 with only a small reduction in seed germination. Further studies are needed to assess the benefits of this treatment for other seed types, particularly as some of these are thought to be more difficult to disinfect. A variety of chemical and physical disinfection processes are being studied although none has yet proven ideal.

Disinfection during germination and sprouting

Present chemical disinfectant treatments of seed can fail to inactivate all bacterial pathogens that may be initially present. Surviving pathogens can then grow to more than one million per gram during sprout production. In an effort to overcome this problem, intermittent spraying of the germinating seed and of the growing sprouts with a variety of chemical solutions has been examined. Unfortunately, if pathogens have survived seed disinfection, no chemical treatment has been found which will consistently and significantly reduce the number of pathogens on the sprouts by more than that given by a simple water wash. In spite of this, small amounts of chlorine (2-5 ppm) in the water used during germination and sprouting are useful for destroying pathogens that may be in the water supply. E. coli O157:H7 has been shown to contaminate the edible parts of radish sprouts when the hydroponic water in which the roots were immersed was contaminated.

The failure, by intermittent spraying of disinfectant solutions, to prevent the growth of pathogens during germination and sprout growth may partly be due to clumps of bacterial cells being within clusters of sprouts, or in the root mat, where they are protected from the chemical. The formation of a biofilm on sprouts may also protect pathogens from the antimicrobial activity of chemicals.

Decreasing the risk of foodborne disease

Several factors are currently important in the risk of foodborne disease from the consumption of raw sprouts make it difficult to design a system that removes the risk of foodborne illness from the consumption of raw sprouts. They are:

  • protection of a percentage of pathogens on seed from chemical disinfectants;
  • extensive bacterial growth during normal germination and sprouting;
  • protection of resulting pathogens on sprouts from chemical disinfection and
  • sprouts are frequently consumed raw.

In spite of the problems of achieving large reductions in contaminants, seed disinfection has been shown to be important in reducing the incidence of sprout-associated illness. In studies of recent outbreaks in the US, foodborne disease was associated with sprouts made by manufacturers who did not consistently disinfect seed or used disinfectants at relatively low and ineffective levels. On the other hand, manufacturers who consistently used seed disinfectant treatments, such as 2,000-20,000 ppm calcium hypochlorite, were not implicated in foodborne disease, even when seed from the same contaminated batch was used. However, US regulatory bodies suggest that intervention strategies need to be able to achieve at least a 100,000 fold reduction in pathogens such as Salmonella or E. coli O157 to provide a sufficient margin of safety. So far no process has been shown to give this level of pathogen reduction for a variety of sprouts. It has been recommended that seeds for sprouting be given one or more approved treatments (eg calcium hypochlorite 20,000 ppm) for the reduction of pathogens in seeds or sprouts.

Testing sprouts for pathogens would be more likely to detect contaminated batches than testing the seed; however, the short shelf-life of sprouts makes a "test and hold" program probably unrealistic. Disinfectants used during sprouting may destroy pathogens free in the irrigation water but leave most of them viable on the sprouts where they can harbour in biofilms. Weighing these advantages and disadvantages, the US FDA guidelines recommend the sampling of irrigation water, where sprouts are grown hydroponically, as early as 48 hours after the start of sprouting. Two one litre samples of water are collected as opposed to thirty-two 50 gram samples of sprouts.

The US approach is to test for pathogens only eg Salmonella spp. and E. coli O157:H7. To test only for the E. coli O157 serotype is limited given the increasing recognition of pathogens of non-O157 serotypes. Though Listeria has been shown to grow during the production of some sprouts, they have not been reported in human infections. Further work is needed to assess the risk from this bacterium. In the draft ANZFA Food Standards Code, there is a zero tolerance requirement forSalmonella spp. There is currently insufficient information to consider alternative indicators eg E. coli.

Some countries have initiated consumer awareness programs. In the US, as a result of a series of outbreaks, the FDA issued in 1998 and 1999 advice "to make all persons aware of the risks associated with the consumption of raw sprouts (e.g. alfalfa, clover and radish)", "that persons who wish to reduce the risk of foodborne illness should not eat or consume raw sprouts", and "that advice against consumption is particularly important for persons at high risk of developing serious illness due to foodborne disease (i.e. children, the elderly, and persons with weakened immune systems)".

Contact: Dr Trish Desmarchelier, Telephone 07 3214 2000