Anda di halaman 1dari 8

International Journal of Food Microbiology 75 (2002) 119 126 www.elsevier.

com/locate/ijfoodmicro

Microbiological analysis of seed sprouts in Norway


Lucy J. Robertson a,*, Gro S. Johannessen b, Bjrn K. Gjerde a, Semir Loncarevic b
a

Section of Parasitology, Department of Pharmacology, Microbiology and Food Hygiene, The Norwegian School of Veterinary Science, PO Box 8146 Dep., 0033 Oslo, Norway b Section for Food and Feed Microbiology, National Veterinary Institute, PO Box 8156 Dep., 0033 Oslo, Norway Received 24 June 2001; accepted 5 November 2001

Abstract As part of larger survey of microbial contamination of fruits and vegetables in Norway, four different sprouted seed products were analysed for bacterial and parasitic contaminants (n = 300 for bacterial analyses and n = from 17 to 171 for parasite analyses, depending on parasite). Escherichia coli O157, Salmonella, Listeria monocytogenes, Cyclospora oocysts, Ascaris eggs and other helminth parasites were not detected in any of the sprout samples. Thermotolerant coliform bacteria (TCB) were isolated from approximately 25% of the sprout samples, with the highest percentage of TCB positive samples in alfalfa sprouts. Most TCB were Enterobacter spp. and Klebsiella. E. coli was isolated from 8 of 62 TCB positive mung bean sprout samples. Cryptosporidium oocysts were detected in 8% of the sprout samples and Giardia cysts were detected in 2% of the samples. All the Cryptosporidium positive samples, and most of the Giardia positive samples, were mung bean sprouts. Parasite concentrations in positive samples were low (between 1 and 3 oocysts/cysts per 50 g sprouts). Sprout irrigation water was also analysed for microbial contaminants. E. coli O157 and L. monocytogenes were not detected. TCB were isolated from approximately 40% of the water samples. Salmonella reading was isolated from three samples of spent irrigation water on 3 consecutive days. Cryptosporidium and Giardia were also isolated from spent irrigation water. Additionally, eight samples of unsprouted mung bean seed were analysed for Cryptosporidium oocysts and Giardia cysts. One or both of these parasites were detected in six of the unsprouted seed samples at concentrations of between 1 and 5 oocysts/cysts per 100 g unsprouted seed. Whilst our results support spent irrigation water as the most suitable matrix for testing for bacteria, unsprouted seed is considered the more useful matrix for analysing for parasite contaminants. D 2002 Elsevier Science B.V. All rights reserved.
Keywords: Alfalfa; Ascaris; Bean sprouts; Cryptosporidium; Cyclospora; Escherichia coli O157; Giardia; Helminths; Listeria monocytogenes; Mung; Radish; Parasites; Salmonella; Seed sprouts; Thermotolerant coliform bacteria; Vegetables

1. Introduction Consumption of seed sprouts, which has been ubiquitous for many centuries in oriental culture, has been growing in global popularity over the past 30 years. In

Corresponding author. Tel.: +47-22-96-49-66; fax: +47-2296-49-65. E-mail address: lucy.robertson@veths.no (L.J. Robertson).

the USA, soybean sprouts were heralded as a new and marvellous form of nutrition during the second world war, and since then, increasing interest in eating raw, alternative products has popularised sprout consumption yet further. In a US survey in 1998 1999, of 12,755 people questioned, 976 (7.7%) reported having eaten alfalfa sprouts during the previous 7 days (C.D.C, 1999). Reputedly there are over 450 US-based sprout growers producing over 300,000 tons of sprouts annually to fulfil the consumer demand (Kurtzweil, 1999).

0168-1605/02/$ - see front matter D 2002 Elsevier Science B.V. All rights reserved. PII: S 0 1 6 8 - 1 6 0 5 ( 0 1 ) 0 0 7 3 8 - 3

120

L.J. Robertson et al. / International Journal of Food Microbiology 75 (2002) 119126

Although consumption of seed sprouts has a healthy image, seed sprouts have been demonstrated to have been the vehicle for transmission in a number of foodborne outbreaks of infection. These outbreaks of infection have included salmonellosis and Escherichia coli O157 infection, and implicated seed sprouts include alfalfa, clover, cress, mung bean, radish and soy. Most of the outbreaks are detailed in a review paper (Taormina et al., 1999), and have occurred in Canada, Denmark, Finland, Japan, Sweden, UK and US. More recently (2000 and 2001), outbreaks of salmonellosis have been linked to mung bean sprouts in California, USA involving 45 cases (ProMED-mail Post, 2000a) and in Edmonton, Canada, involving over 30 cases (Honish and Nguyen, 2001) and an outbreak of Salmonella enteriditis PT4b infection involving 25 cases associated with soya bean sprout consumption in the Netherlands has been reported (ProMED-mail Post, 2000b). Furthermore different unsprouted sesame seed products (tahini and halva) have recently been associated with outbreaks of salmonellosis in Sweden and Australia (ProMED-mail Post, 2001). As part of a larger survey of microbial contamination of fruits and vegetables in Norway (Robertson and Gjerde, 2001a), various sprouted seed product samples were analysed for bacterial and parasitic contaminants. Additionally, water samples used during the sprout production process and unsprouted seeds were analysed. The purpose of this study was to determine the microbiological quality of sprouted seed products on the Norwegian market, to examine the potential public health importance of microbial contamination of seed sprouts, and to investigate the most appropriate approach for analysing this product for bacteria and parasites.

samples to the laboratory. Bacteriological analysis of the samples was conducted either directly on arrival at the laboratory, or following overnight storage in the refrigerator. Parasitological analysis was commenced within 72 h of arrival of samples at the laboratory. Not all samples were analysed for all bacteria and parasites. Numbers of samples of produce analysed for the various bacterial and parasitic contaminants are described in Tables 2 and 3. 2.2. Processing water Water samples associated with the production of mung bean sprouts were analysed. Altogether 46 samples of spent irrigation water were analysed for the presence of bacterial contaminants. Samples were taken weekly, except for 2 weeks when samples were taken daily. Approximately 1-l volumes of water were sampled into sterile bottles, transported to the laboratory and analysed directly on arrival. Three water samples were also analysed for Cryptosporidium oocysts and Giardia cysts. Two 10-l samples were of spent irrigation water, and one 10-l sample was water taken directly from the source, before contact with the sprouting seeds. 2.3. Unsprouted mung bean seeds Samples of unsprouted mung bean seeds were obtained from the bean sprout supplier. Eight 100-g samples were analysed for Cryptosporidium and Giardia. 2.4. Bacteriological analysis of sprout and water samples The sprout and water samples were analysed for the presence of thermotrophic coliform bacteria (TCB), Salmonella, E. coli O157 and Listeria monocytogenes using standard methods as listed in Table 1. 2.5. Analysis of sprout samples for Ascaris and other helminths, Cryptosporidium, Giardia and Cyclospora, and analysis of water and seed samples for Cryptosporidium and Giardia Development of the methods used in this survey for analysing the sprouts and sprout seeds are described in

2. Materials and methods 2.1. Seed sprouts Alfalfa, mung bean, radish sprouts and sprout mix (including sprouts of green peas, adzuki beans, lentils and chick peas) were included in the analyses. All had been grown in Norway by the same producer. The three main distributors of fruit and vegetables in Norway collected the samples in rotation and transported the

L.J. Robertson et al. / International Journal of Food Microbiology 75 (2002) 119126 Table 1 Methods of bacteriological analysis used for samples of sprouted seeds and processing water Microorganism Thermotolerant coliforms Analysis Sample type Method or reference method

121

Enumeration Sprouts

E. coli O157

Isolation

Water Sprouts

Water

Salmonella spp.

Isolation

Sprouts

L. monocytogenes Isolation

Water Sprouts Water

NMKLa no. 125. Presumptive colonies cultivated on blood-agar prior to confirmation by gas-production in EC-broth, indole production (BBLk DrySlidek Indole, Becton Dickinson and Company, France) and identification by API 20E (bioMerieux, France). Detection limit 10 cfu/g. NSb 4792. Detection limit 1 cfu/100 ml. 25 g sample enriched in buffered peptone water (BPW) (225 ml) at 37 F 1C for 6 hrs prior to immunomagnetic separation (IMS) (according to Dynal) followed by plating on Sorbitol MacConkey agar with cefixime and tellurite (CT-SMAC Oxoid Ltd., England) and Chromagar O157 (Chromagar Microbiology, France). Presumptive positive colonies confirmed by subcultivation on blood agar, tested for indole-production (DrySlide Indole) and agglutination (drySpot E. coli O157, Oxoid Limited, England). NMKL no. 164. 100 ml of water filtered and filter transferred to 225 ml mTSB with Novobiocin (Difco, USA) and incubated at 41.5 F 0.5C for 6 h prior to IMS. After IMS, samples plated onto Sorbitol MacConkey agar (SMAC Oxoid Ltd., England) and CT-SMAC (containing cefixime and tellurite). Confirmation performed as described in the NMKL method above. Pre-enrichment and selective enrichment performed as described in NMKL no. 71. A total of 0.1 ml of selective enrichment broth used for the ELISA test (Salmonella ELISA test, bioline, Denmark). Presumptive positive samples confirmed according to NMKL no. 71. NMKL no. 71. NMKL no. 136. In addition to Oxford and Palcam agar (Oxoid Ltd., England), the samples also plated on blood-agar. NMKL no. 136. Samples plated on Oxford agar only.

a b

NMKL = Nordic Committee on Food Analysis, 1996, 1999a,b,c. NS = Norwegian Standards Association, 1990.

detail elsewhere (Robertson and Gjerde, 2000; Robertson et al., 2000). In brief, 50-g samples of the sprouts, or 100-g samples of unsprouted seeds, were weighed into homogeniser bags containing a central , France and Etranger) and profilter (BagPage, Brevete cessed. Sample processing consisted of four distinct stages: (a) elution into aqueous suspension, (b) concentration of suspension, (c) separation of the parasites in the suspension (for samples analysed for Cryptosporidium and Giardia and/or Cyclospora) using paramagnetic beads, and (d) screening by microscopy. The procedures are described in greater detail elsewhere (Robertson and Gjerde, in press (a)). For all parasites, appropriate calculations were performed to estimate the number of parasites in the whole of the sample from the number detected in the sub-sample. The procedure for analysing water samples for Cryptosporidium and Giardia was based upon Method

1623 (Anonymous, 1999b). The analytical technique can be divided into 5 distinct sections as follows: (a) membrane filtration of sample, (b) elution of material from membrane filter using detergent solution as described elsewhere (Anonymous, 1997), (c) concentration of eluted material by centrifugation, (d) isolation of parasites from concentrated eluted material by IMS, and (e) screening by microscopy. 2.6. Recovery efficiency of techniques used for parasitological analysis Seeding experiments were conducted to assess the recovery efficiency of the techniques used. Details of some of these experiments can be found elsewhere (Robertson and Gjerde, 2001b; Robertson et al., 2000). In brief, 50-g samples of mung bean sprouts, and 100g samples of unsprouted mung beans, were seeded with appropriate parasites of known concentration, the

122

L.J. Robertson et al. / International Journal of Food Microbiology 75 (2002) 119126 Table 3 Results of parasitological analyses of sprouted seed samples Number positive/number analysed. In brackets are % positive samples Cyclospora Helminths Cryptosporidium Giardia (Ascaris) Produce Alfalfa Mung bean Radish Total 0/1 (0) 0/16 (0) 0/0 ( ) 0/17 (0) 0/13 (0) 0/81 (0) 0/2 (0) 0/96 (0) 0/16 (0) 14/149 (9) 0/6 (0) 14/171 (8) 0/16 (0) 3/149 (2) 1/6 (17) 4/171 (2)

described method followed and the recovery efficiency calculated. Although relatively high recovery efficiencies have been found using these methods from other vegetables (Robertson and Gjerde, 2000; Robertson et al., 2000) recovery efficiencies from bean sprouts were more variable and lower, being between approximately 25% and 35% for Cryptosporidium, 4% and 42% for Giardia, 38% and 50% for Ascaris and 4% for Cyclospora (Robertson and Gjerde, 2000; Robertson et al., 2000). Possible approaches for improving parasite recovery efficiency from bean sprouts are described elsewhere (Robertson and Gjerde, 2001c). Recovery efficiency from mung bean seeds was between 45% and 50% for Cryptosporidium (n = 2) and 65% and 70% for Giardia (n = 2). Recovery efficiency from water samples was approximately 43% for Cryptosporidium and 67% for Giardia (Robertson and Gjerde, 2001b).

No sample was positive simultaneously for Cryptosporidium and Giardia.

3. Analysis of data Data was compiled in a spreadsheet (Microsoft Excel) and analysed as appropriate using descriptive statistics, and by construction of contingency tables to enable Chi-squared analysis for association.

4. Results 4.1. Microbiological analysis of seed sprouts E. coli O157, Salmonella, and L. monocytogenes were not detected in any of the samples. Thermotoler-

Table 2 Results of bacteriological analyses of seed sprouts Number (%) samples positive TCBa Produce Alfalfa (n = 27) Mung bean (n = 259) Radish (n = 6) Sprout mix (n = 8) Total (n = 300) 8 62 0 2 72 (30) (24) (0) (25) (24)

Escherichia coli O157, Salmonella and Listeria monocytogenes were not detected in any of the samples. a Thermotrophic coliform bacteria.

ant coliform bacteria (TCB) were isolated from approximately 25% of the samples of sprouts in varying numbers (between 0.2 102 and 1.4 107 cfu/g) (Table 2). Enterobacter spp. and Klebsiella spp. were isolated from most of the TCB positive samples. The strains were typically identified as E. cloacae, E. sakazakii and K. pneumoniae ssp. pneumoniae. E. coli was isolated from eight of the 62 TCB positive mung bean sprout samples. Although a greater percentage of alfalfa sprouts was TCB positive than other sprout types, there was no statistically significant association between alfalfa sprouts and TCB contamination compared to the other sprout types. Neither Cyclospora oocysts nor Ascaris eggs or other helminth transmission stages were detected on any of the samples. Eighteen (10.5%) of the samples examined for Cryptosporidium oocysts and Giardia cysts were found to be positive (Table 3). No samples were found to be simultaneously positive for both parasites. Concentrations of Cryptosporidium and Giardia detected were generally low (Table 4). Although all Cryptosporidium positive samples and most Giardia positive samples were mung bean sprout samples (Table 3), no statistically significant association between mung bean sprouts and parasite contamination was detected as compared to the other sprout types ( p = 0.33). Only one of the samples that was positive for Cryptosporidium or Giardia was positive for bacterial contaminants; in this sample of mung bean sprouts, 1 Cryptosporidium oocyst/50 g and 1.0 103 cfu/g E. coli were detected.

L.J. Robertson et al. / International Journal of Food Microbiology 75 (2002) 119126 Table 4 Concentrations of Cryptosporidium oocysts and Giardia cysts detected in sprouted seeds Number of samples Cryptosporidium positive samples Mung bean sprouts 6 6 2 Giardia positive samples Mung bean sprouts 1 1 1 Radish sprouts 1 (Oo)cysts per 100 g sprouts 2 4 6

123

ing seeds. However, parasites were detected in the samples of spent irrigation water; a single Cryptosporidium oocyst was detected in one sample and a single Giardia cyst was detected in the other sample. 4.3. Parasitological analysis of unsprouted mung bean seed Of the eight 100-g portions of mung bean seeds analysed, parasites were not detected in two of them. The remaining six all had Cryptosporidium oocysts detected in them (range = 1 5 oocysts per 100 g portion, mean = 2.3 oocysts, median = 2 oocysts), and of these, Giardia cysts were detected in three portions (one Giardia cyst in each sample).

2 4 6 2

4.2. Microbiological analysis of processing water E. coli O157 and L. monocytogenes were not detected in any of the 46 samples of spent irrigation water. TCB were isolated from approximately 40% of the water samples (Table 5). The numbers of TCB varied from 2 cfu/100 ml to 6.0 105 cfu/100 ml. E. coli was isolated from three, Enterobacter spp. from 12 and Klebsiella spp. were isolated from eight of the water samples. The Enterobacter spp. identified were mostly E. cloacae and the klebsiellas were identified as K. pneumoniae ssp. pneumoniae. Salmonella reading was detected in three samples of spent irrigation water from the same batch of sprouts on 3 consecutive days. Mung bean sprouts from the batch from which this contaminated water had been sampled were unfortunately not analysed as a part of this project. Parasites were not detected in the single water source sample which had not been in contact with the sprout5. Discussion The occurrence of TCB in 24% of the sprout samples and 41% of the samples of spent irrigation water could be a warning that pathogens might be present. However, E. coli was only isolated from eight TCB positive sprout samples and three TCB positive spent irrigation water samples. The method used for the detection of TCB may not be an optimal method for this type of product. There was a high level of background flora present which may have limited the detection of indicator bacteria such as E. coli, which may have been present at low levels. Most of the TCB strains isolated belonged to either Enterobacter spp. or Klebsiella spp. These species are normally present in the environment, and are opportunistic pathogens for humans not usually considered to be of food hygiene

Table 5 Results of microbiological analyses of processing water used in the production of mung bean sprouts Number positive Spent irrigation water (1-l volumes) n = 46 Spent irrigation water (10-l volumes) n = 2 Source water (before contact with sprouting seeds; 10-l volumes) n = 1 0 0

TCB E. coli O157 Salmonella L. monocytogenes Cryptosporidium Giardia


a

19 0 3 0 a

1 1

Not examined.

124

L.J. Robertson et al. / International Journal of Food Microbiology 75 (2002) 119126

importance. This suggests that E. coli, rather than TCB, should be used as indicator organism when examining the hygienic state of sprouts and other vegetables. Whilst the only parasites detected in the sprouts were Cryptosporidium oocysts and Giardia cysts, their occurrence was indicative of faecal contamination. However, only one of the samples was simultaneously positive for Cryptosporidium, Giardia and E.coli; this may be a reflection of important differences between bacteria and these parasites, particularly the potential for amplification and dissemination of bacteria, as opposed to the protozoan parasites, through a production lot of seed sprouts. Despite the detection of the human pathogens, Cryptosporidium, Giardia and Salmonella reading, in the sprouts and water samples, no known cases of infection were associated with contaminated seed sprouts throughout the survey period. For both parasites and Salmonella, it is possible that if infection occurred, it was not always identified or reported or not associated with specific product consumption. Additionally, parasitic infection may not have occurred because the level of contamination was below the human infectious dose or because the parasites were non-viable or non-infectious to humans. No attempt was made to assess viability, infectivity or strain/ isolate of parasite during the survey. The contamination of the sprouts and the spent irrigation water with bacteria and parasites is probably due to the use of contaminated seeds (Mahon et al., 1997; Anonymous, 1999a). It has been suggested that contaminated seed has been the source of most, if not all, sprout associated disease outbreaks, and seed is apparently frequently grown, milled and stored in conditions where contamination can occur readily (Anonymous, 1999a). Contamination with pathogenic bacteria may be low, but conditions and processes during sprouting are ideal for amplifying numbers of Salmonella and E. coli and also for spreading the infective agent throughout the entire production lot. However, for pathogens such as Cryptosporidium and Giardia, which do not amplify outside their hosts, the risks of adverse public health consequences are considered similar for sprouts to those for other products, or, indeed, less due to the extensive washing during sprout production (Anonymous, 1999a). However, our survey indicates that sprouts may be contaminated with Cryptosporidium oocysts and Giardia cysts, and

the results from analyses of pre-rinse and spent irrigation water and also of unsprouted mung bean seeds indicates that seeds are almost certainly the source of contamination. Regular bacteriological analyses of the water source used by the sprout production facility mdal, pers. comm.) also (Gro Gjedebo, KNT Gla indicate that the seeds, rather than the water, are likely to have been the source of bacterial contaminants. The possible routes for seed contamination are described in full elsewhere (Anonymous, 1999a), but the primary reason appears to be that the seeds are treated as a raw agricultural product rather than a food product and may carry microorganisms from their original environment. Indeed, most seed grown is used for agricultural purposes rather than sprouting, and the decision is often not made until post-harvesting. The use of manure as a fertiliser, faecally contaminated water for irrigation, birds, insects, animals, unclean harvesting equipment, storage equipment and transportation vessels are all factors that should be considered when attempting to identify possible contamination routes for agricultural products (Beuchat and Ryu, 1997). Cracks and cavities in the seeds where the presumptive pathogenic bacteria can survive, and eventually create biofilms (Fett, 2000) and where parasitic protozoa may adhere, may also be of importance. It is difficult to remove such contamination, as the forces holding the pathogens to the seeds are strong, and would require vigorous cleaning to effect removal. Research has been conducted on approaches for disinfection of seeds (Delaquis et al., 1999; Taormina and Beuchat, 1999; Rajkowski and Thayer, 2000; Weissinger and Beuchat, 2000). However, the optimal disinfection procedure is not obvious, as the method must not destroy the seeds capacity to sprout, and the potentially harmful effects of discharge of large volumes of disinfectant must also be considered. Furthermore, whereas some disinfection regimes may eliminate or diminish bacterial survival, it is unlikely to be effective for Cryptosporidium and Giardia, which are notoriously robust and can withstand a range of disinfectants. As well as supplying information on pathogen occurrence, this study provides useful insights on approaches to analysing sprouted seeds for pathogens. The results indicate that pathogen testing for sprouted seeds is probably not best conducted on the seed

L.J. Robertson et al. / International Journal of Food Microbiology 75 (2002) 119126

125

Table 6 Comparison of factors affecting the most appropriate analytical regime for bacterial pathogens and indicators and protozoan pathogens in seed sprouts Bacterial pathogens and indicators (e.g. E. coli, Salmonella) Concentration in unsprouted seeds Concentration in seed sprouts Low; only surviving, unable to proliferate or grow High; sprouting provides excellent conditions (temperature, humidity, nutrition) for resuscitation and proliferation of bacteria Protozoan pathogens (e.g. Giardia cysts, Cryptosporidium oocysts) Low; unable to proliferate or grow Lower; unable to proliferate or grow 1 kg sprout seed may yield between 5 to 8 kg sprouts. Losses in irrigation water. Also lower recovery efficiency of analytical methods Lower; unable to proliferate or grow Large volumes of water generally used for irrigation, particularly of mung bean sprouts Probably uneven Probably less uneven than in unsprouted seeds due to mixing Probably less uneven than in unsprouted seeds due to mixing and water running through whole production batch Unsprouted seeds

Concentration in spent irrigation water

Distribution in unsprouted seeds Distribution in seed sprouts Distribution in spent irrigation water In conclusion: preferred matrix for detection

High (although may not be as high as in sprouts); sprouting provides excellent conditions (temperature, humidity, nutrition) for resuscitation and proliferation of bacteria Probably uneven Probably less uneven than in unsprouted seeds due to mixing, and amplification Probably less uneven than in unsprouted seeds due to mixing, amplification and water running through whole production batch Spent irrigation water

sprout produce itself, but on either the spent irrigation water, or, for parasites, on the unsprouted seeds (Table 6). In this survey, the detection rate for TCB was higher from spent irrigation water than in the actual sprouts. This may be because proportionally larger portions of the sprouts were tested when the spent irrigation water was analysed, as the water runs through the whole batch of sprouts before the water is collected. Testing of spent irrigation water for bacterial contaminants is supported by the recently published data of Stewart et al. (2001). As Cryptosporidium and Giardia do not proliferate outside their hosts, there is no numerical amplification advantage of testing sprouts and/or spent irrigation water following germination procedures. Indeed, cyst/ oocyst concentrations per unit of matrix (sprout or water) will be reduced, and recovery efficiencies in sprouts are also adversely affected. The disadvantage of seed testing is that if contamination is localised (for example a single bag, contaminated in a single corner), then it would be unlikely to be detected in seed testing, but may be more likely to be detected if spent irrigation water is analysed. In USA, most sprout growers participate in a voluntary programme of testing spent irrigation water from sprout production for bacterial pathogens (Ano-

nymous, 1999c), and our study supports this regime. However, sprout producers do not routinely test for Cryptosporidium and/or Giardia, and, indeed, to our knowledge, our study is the first to suggest that sprouted seeds may pose a risk of infection with these protozoan parasites.

6. Conclusion This Norwegian survey provides information on the occurrence of pathogens in sprouted seeds, and also indicates the utility of different approaches for analysis of this product for different pathogens. Our data indicate the requirement for further research in this area so that the risk of infection with these pathogens via consumption of sprouted seed may be more fully understood, and the importance and most pertinent approach for analysis assessed.

Acknowledgements This work was supported by a grant from the Norwegian Food Control Authority. We acknowledge mdal, and the ready cothe assistance of KNT Gla

126

L.J. Robertson et al. / International Journal of Food Microbiology 75 (2002) 119126 outbreak of Salmonella infections caused by alfalfa sprouts grown from contaminated seeds. J. Infect. Dis. 175, 876 882. NMKL, 1996. Thermotolerant coliform bacteria. Enumeration in foods. Method 125, 3rd edn. Nordic Committee on Food Analysis, Oslo, Norway. NMKL, 1999a. Salmonella. Detection in foods. Method 71, 5th edn. Nordic Committee on Food Analysis, Oslo, Norway. NMKL, 1999b. Listeria monocytogenes. Detection in foods. Method 136, 2nd edn. Nordic Committee on Food Analysis, Oslo, Norway. NMKL, 1999c. Escherichia coli O157. Detection in food and feeding stuffs. Method 164, 1st edn. Nordic Committee on Food Analysis, Oslo, Norway. NS, 1990. Water analysis. Thermotolerant coliform bacteria and presumptive E. coli. Membrane filter method. Method 4792, 1st edn. Norwegian Standards Association, Oslo, Norway. ProMED-mail Post, 2000a. Salmonella, bean sproutsUSA (California): recall. 20000423.0608. ProMED-mail Post, 2000b. Salmonellosis, bean sprouts?Netherlands. 20001222.2258. ProMED-mail Post, 2001. S.typhimurium DT104Australia, Sweden: recall. 20010730.1494. Rajkowski, K.T., Thayer, D.W., 2000. Reduction of Salmonella spp. and strains of Escherichia coli O157 by gamma radiation of inoculated sprouts. J. Food Prot. 63, 871 875. Robertson, L.J., Gjerde, B., 2000. Isolation and enumeration of Giardia cysts, Cryptosporidium oocysts, and Ascaris eggs from fruits and vegetables. J. Food Prot. 63 (6), 775 778. Robertson, L.J., Gjerde, B., Campbell, A.T., 2000. Isolation of Cyclospora oocysts from fruits and vegetables using lectin-coated paramagnetic beads. J. Food Prot. 63 (10), 1410 1414. Robertson, L.J., Gjerde, B., 2001a. Occurrence of parasites on fruits and vegetables in Norway. J. Food Prot. 64 (11), 1793 1798. Robertson, L.J., Gjerde, B., 2001b. Occurrence of Cryptosporidium oocysts and Giardia cysts in raw waters in Norway. Scand. J. Pub. Health 29, 200 207. Robertson, L.J., Gjerde, B., 2001c. Factors affecting recovery efficiency in isolation of Cryptosporidium oocysts and Giardia cysts from vegetables for standard method development. J. Food Prot. 64 (11), 1799 1805. Stewart, D.S., Reineke, K.F., Ulaszek, J.M., Tortorello, M.L., 2001. Growth of Salmonella during sprouting of alfalfa seeds associated with salmonellosis outbreaks. J. Food Prot. 64 (5), 618 622. Taormina, P.J., Beuchat, L.R., 1999. Comparison of chemical treatments to eliminate enterohemmorhagic Escherichia coli O157: H7 on alfalfa seeds. J. Food Prot. 62 (4), 318 324. Taormina, P.J., Beuchat, L.R., Slutsker, L., 1999. Infections associated with eating seed sprouts: an international concern. Emerging Infect. Dis. 5 (5), 626 634. Weissinger, W.R., Beuchat, L.R., 2000. Comparison of aqueous chemical treatments to eliminate Salmonella on alfalfa seeds. J. Food Prot. 63 (11), 1475 1482.

operation of the seed sprout supplier. We are grateful to Andrew Campbell, Microbiology R&D. Dynal AS, OSLO, for provision of the WGA-Dynabeads for isolation of Cyclospora oocysts.

References
Anonymous, 1997. Method 1622: Cryptosporidium in water by filtration/IMS/FA. DRAFT. U.S. Environmental Protection Agency, Office of Water, Washington, DC 20460, EPA-821-R97-023. Anonymous, 1999a. Microbiological safety evaluations and recommendations on sprouted seeds. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition. National Advisory Committee on Microbiological Criteria for Food. Available at: http://vm.cfsan.fda.gov/~ mow/sprouts2.html. Anonymous, 1999b. Method 1623: Cryptosporidium and Giardia in water by filtration/IMS/FA. U.S. Environmental Protection Agency, Office of Water, Washington, DC 20460, EPA-821-R99-006. Available at: http://www.epa.gov/nerlcwww/. Anonymous, 1999c. Guidance for industry: Reducing microbial food safety hazards for sprouted seeds and guidance for industry: Sampling and microbial testing of spent irrigation water during sprout production. U.S. Food and Drug Administration, Center for Food Safety and Applied Nutrition. Available at: http://vm.cfsan.fda.gov/~ dms/sprougd2.html. Beuchat, L.R., Ryu, J.-H., 1997. Produce handling and processing practices. Emerging Infect. Dis. 3 (4), 459 465. C.D.C., 1999. Centers for Disease Control and Prevention. Foodborne Diseases Active Surveillance Network (FoodNet): Population Survey Atlas of Exposures: 1998 1999. Atlanta: Centers for Disease Control and Prevention. Available at: http:// www.cdc.gov/foodnet/SURVEYS/Popsurv.htm. Delaquis, P.J., Sholberg, P.L., Stanich, K., 1999. Disinfection of mung bean seed with gaseous acetic acid. J. Food Prot. 62, 953 957. Fett, W.F., 2000. Naturally occurring biofilms on alfalfa and other types of sprout. J. Food Prot. 63, 625 632. Honish, L., Nguyen, Q., 2001. Outbreak of Salmonella enteriditis Phage type 913 gastroenteritis associated with mung bean sprouts Edmonton 2001. Canada Communicable Disease Report 27. Available at: http://www.hc-sc.gc.ca/hpb/lcdc/publicat/ ccdr/98vol24/index.html. Kurtzweil, P., 1999. Questions keep sprouting about sprouts. FDA Consum. Mag. 33 (1), 18 22 Available at: http://vm.cfsan. fda.gov/e dms/fdsprouts.html and http://www.fda.gov/fdac/ 199toc.html. Mahon, B.E., Po nka , A., Hall, W.N., Komatsu, K., Dietrich, S.F., Siitonen, A., Cage, G., Hayes, P.S., Lambert-Fair, M.A., Bean, N.H., Griffin, P.M., Slutsker, L., 1997. An international