Prevalence of Foodborne Pathogens in Retailed Foods in Thailand

Abstract

The consumption of foodborne pathogens contaminated in food is one of the major causes of diarrheal diseases in Thailand. The objective of this study was to evaluate the prevalence and types of contaminating bacteria in retailed foodstuffs in Thailand. Food from four categories (137 samples total), including meat (51 samples), vegetables (38 samples), fish or seafood (37 samples), and fermented food (11 samples), was purchased randomly from seven different open-markets and seven supermarkets in Thailand from August 2010 to March 2011. Seven types of major foodborne pathogens were identified using conventional culture methods. Approximately 80% of meat samples tested was contaminated with Salmonella spp. In contrast, the Salmonella spp. contamination rate of vegetable (5%) or fermented food (9%) samples was comparatively low. Six strains of Cronobacter sakazakii and two strains of Yersinia enterocolitica were also isolated. A substantially higher rate of contamination by Bacillus cereus was observed in fermented food (82%) than in samples of meat (2%) and fish or seafood (5%). Seven Listeria spp. isolates were obtained from meat and fish or seafood samples. Approximately 39% of samples tested were found to be contaminated with Staphylococcus spp. (54 isolates). The rate of bacterial contamination of meat did not depend on the type of market. However, the contamination rate of Staphylococcus spp. in vegetables was higher in open markets than in supermarkets, and the contamination rate of Salmonella spp. and Staphylococcus spp. in fish or seafood samples purchased in open markets was likewise higher than in those purchased in supermarkets. Therefore, improvement of hygienic practices throughout the food chain may be required to reduce the risk of food poisoning.

Introduction

P roduction and processing of foods is now making an important contribution to development of countries. In the case of Thailand, the government has clearly emphasized food security and safety by promoting “safe food products” for both domestic and international markets. In addition, the Thai government has established a policy to develop Thailand as the “Kitchen of the World.” It may be needless to say that conducting hygienic food practices in food chain increases the quality of not only exported foods but also domestic foods. However, various kinds of foodborne pathogens have been found in Thai food, and they may be a major cause of diarrheal diseases in this country (Boonmar et al., 2003; Indrawattana et al., 2011; Minami et al., 2010; Padungtod and Kanene, 2006; Suthienkul et al., 1990; Vindigni et al., 2007; Vuddhakul et al., 2000).

There are two types of retail markets in Thailand: supermarkets and open markets. Supermarkets are indoors, often air conditioned, typically offer controlled-temperature environments, and have displays of prepackaged products under refrigeration. In contrast, open markets are traditional open-air markets where foodstuffs are sold by individual vendors or farmers and are usually displayed unwrapped and at ambient temperature. Same categories of foods (such as vegetables, raw meats, fishes) tend to be sold by each vender in the open markets. These markets naturally have multiple sources of potential contamination (rodents, insects, sewage, and water). The state of the prevalence of foodborne pathogens in retailed foodstuffs sold in these markets is not clear.

The objective of this study, therefore, was to evaluate the contamination rate and types of foodborne pathogens including Escherichia (E). coli, Salmonella spp., Cronobacter (C). sakazakii, Yersinia spp., Listeria spp., Staphylococcus spp., and Bacillus (B.) cereus in four categories of retailed food (meat, vegetables, fish or seafood, and fermented food) in two kinds of Thai markets.

Methods

Collection of Thai food samples

Raw food samples (137 total) from four categories including meat (51 samples include of 23 pork samples, 17 chicken samples, 11 beef samples), leafy vegetables (38 samples), fish or seafood (37 samples) and fermented meat/fish food (11 samples) were purchased randomly from geographically different seven open-markets and seven supermarkets in the Bangkok and Pathum Thani areas from August 2010 to March 2011. All samples were transported to the laboratory at Rajamangala University of Technology Thanyaburi in an insulated box with ice to maintain the temperature at 4–6°C.

Escherichia coli

A 25-g sample was homogenized with 225 mL of sterile mEC enrichment broth (Eiken Co., Ltd., Tokyo, Japan), and the mixture was incubated at 42°C for 20 h. Following incubation, one loopful of broth was streaked onto Rainbow Agar O157 (Biolog Inc., Hayward, CA), Chrom Agar O26-O157 (SEL; Kanto Chemical Co. Inc., Tokyo, Japan), XM-EHEC agar (Nissui Co., Ltd., Tokyo, Japan), and CT-MacConkey Sorbitol agar (Nissui Co., Ltd.) and incubated at 35°C for 20 h. Typical E. coli serotype O26, O157, or O111 suspected colonies was picked and confirmed by their biochemical characteristics in TSI/LIM tubes (Nissui Co., Ltd.) and API 20E (BioMérieux, Marcy l'Etoile, France) diagnostic kits. The possibility of production of Shiga-like toxin (stx) 1 and Shiga-like toxin (stx) 2 was assayed by NH Immunochromato VT1/2 (Cosmo Bio Co., Carlsbad, CA) and O-157 VT1/2 PCR Typing Set Plus (Takara Biotechnology Co., Ltd., Osaka, Japan).

Salmonella spp.

A 25-g sample was homogenized with 225 mL of sterile EEM enrichment broth (Nissui Co., Ltd.) and incubated at 35°C for 20 h. After incubation, 0.5- and 1.0-mL portions of the EEM-enriched sample were transferred to 10 mL of Rappaport-Vassiliadis (RV) broth and Hajna tetrathionate (TT) broth (Eiken Co., Ltd., Tokyo, Japan), respectively. The RV broth was incubated at 42°C for 20 h, and the TT broth was incubated at 35°C for 20 h. Following incubation, both cultures were streaked on DHL agar (Nissui Co., Ltd.), bismuth sulfide agar (Difco Lab, Detroit, MI), and MLCB agar (Nissui Co., Ltd.) and incubated at 35°C for 20 h. Suspected colonies were picked and subjected to biochemical testing with TSI/LIM tubes and with the Salmonella LA “Seiken” latex-agglutination test kit (Denka Seiken Co., Ltd., Tokyo, Japan). Positive isolates were confirmed by API 20E diagnostic kits.

Yersinia spp.

A 25-g sample was homogenized in 225 mL of phosphate-buffered saline (PBS, pH 7.0) and incubated at 10°C for 7 days. After enrichment, a loopful of culture was streaked onto Yersinia selective agar (Oxoid Ltd., Basingstoke, UK) with Yersinia selective supplement SR0109 (Oxoid Ltd.) and incubated at 30°C for 20 h. Following incubation, colonies were isolated and confirmed using an API 20E (BioMérieux) diagnostic kit. Positive Y. enterocolitica strains were isolated and confirmed by 16S ribosomal DNA sequencing.

Cronobacter sakazakii

Samples were homogenized in Mossel-Bouillon enrichment medium (Merck, Darmstadt, Germany) and followed by incubation at 35°C for 20 h. Pre-enrichment cultures were used for isolation of C. sakazakii on Chromocult Enterobacter Sakazakii agar (Merck). Colonies were subjected to further characterization using an API 20E kit and 16S ribosomal DNA sequencing for C. sakazakii.

Listeria spp.

A 25-g food sample was homogenized in 225 mL of Listeria enrichment broth (Oxoid Ltd.) with Listeria selective enrichment supplement SR0141E (Oxoid Ltd.) and incubated for 48 h at 30°C. After incubation, a loopful of culture was streaked onto Listeria selective agar (Oxoid Ltd.) supplemented with SR0140E (Oxoid Ltd.) and incubated at 30°C for 48 h. Suspected colonies were restreaked on TSA agar with 0.6% yeast extract, and their identity was confirmed using an API Listeria (BioMérieux) diagnostic kit. Confirmation tests were performed with an NH Immunochromato Listeria kit (Cosmo Bio Co.) or by 16S ribosomal DNA sequencing.

Bacillus spp.

A 25-g sample was homogenized in 225 mL of buffered peptone water (Merck) and incubated at 30°C for 20 h. This pre-enrichment culture was streaked on NGKG agar (Nissui Co., Ltd.) with 20% egg yolk and incubated at 30°C for 20 h. Suspected colonies were isolated, identified by microscopic examination, and confirmed using an API 50CH with API CHB (BioMérieux) diagnostic kit. Production of diarrheal enterotoxin was determined with a CRET-RPLA Kit (Denka Seiken Co., Ltd.). A PCR Detection kit (Takara Bio Inc., Tokyo, Japan) was used to amplify the cereulide synthetic enzyme (CRS) gene.

Staphylococcus spp.

One loopful of pre-enrichment culture used for the isolation of Bacillus spp. was streaked on mannitol salt agar (Nissui Co., Ltd.) with 20% egg yolk and incubated at 30°C for 20 h. Suspected colonies were isolated and identified by microscopic examination and confirmed using an API Staph (BioMérieux) diagnostic kit and 16S ribosomal DNA sequencing. Enterotoxin-producing S. aureus strains were isolated and confirmed using a PS Latex (Eiken Chemical Co., Ltd.) immuno-latex agglutination test kit.

Statistical analysis

One-way analysis of variance (ANOVA) was used to evaluate differences in the number of positive samples per food group. A statistically significant deviation of the contamination rate of each of the tested pathogens among the four categories of tested food samples was detected by chi square test. Differences between the two types of market were evaluated using Fisher's exact probability test. Microsoft EXCEL 2007 (Microsoft, Redmond, WA) was used for the data analysis.

Results

A total of 137 samples, including meat, vegetables, fish or seafood, and fermented food, were obtained from different open markets and supermarkets.

Forty-two of 137 (31%) food samples were contaminated with several kinds of three tested serotypes (O26, O157, and O111) of E. coli: 57% meat samples (29 of 51 meat samples), 22% fish or sea food samples (eight of 37 fish or sea food samples), 11% vegetables (four of 38 vegetables), and 9% fermented pork or fish samples (one of 11 fermented pork or fish samples). A significant (p≤0.05) difference in the E. coli contamination rate between open market and supermarket was observed in the case of fish or seafood: open market 76% and supermarket 45%. Similarly, in the case of vegetable samples the following was found: open market 44% and supermarket 15%. In 156 of E. coli O26, O157, and O111 suspected strains isolated from all the tested samples, no strain harbored stx1/2 genes nor did any produce Shiga-like toxins.

A comparably higher rate of contamination was observed for Salmonella spp. (at 80% in the meat samples). However, only 5% of the vegetable samples were contaminated by Salmonella spp. (Table 1). Two vegetable samples were found to be contaminated with Salmonella spp.; one was a morning glory from the supermarket and the other was kale from an open market. The prevalence of Salmonella spp. in meat samples from open markets was found to be 83% (30 of 36 meat samples, including of 82% of chicken, 78% of beef, and 88% of pork samples) and from supermarkets was found to be 67% (10 of 15 meat samples, including 50% of chicken, 50% of beef, and 86% of pork samples). No significant difference in prevalence of Salmonella spp. in meat was observed between open-market and supermarket samples (p>0.05) (Table 2). However, a significantly (p<0.05) higher rate of contamination by Salmonella spp. was observed in fish or seafood from open markets (65%) compared to supermarkets (10%; Table 2).

Table 1.

Contamination Rate of Pathogens in 137 Retail Foods in Thailand

		Gram-negative pathogens
		Escherichia coli		Salmonella spp.		Yersinia spp.		Cronobacter sakazakii
		Contaminated samples		Contaminated samples		Contaminated samples		Contaminated samples
Categories of foods	Total Numbers of tested samples	Number	(%)	Number	(%)	Number	(%)	Number	(%)
Meat	51	42	(82)	41	(80)	1	(2)	1	(2)
Fish or seafood	37	22	(59)	13	(35)	2	(5)	1	(3)
Vegetable	38	11	(29)	2	(5)	0	(0)	4	(11)
Fermented pork or fish	11	2	(18)	1	(9)	0	(0)	0	(0)
Total	137	77^*	(56)	57^*	(42)	3	(2)	6	(4)

		Gram-positive pathogens
		Listeria spp.		Bacillus cereus		Staphylococcus spp.
		Contaminated samples		Contaminated samples		Contaminated samples
Categories of foods	Total Numbers of tested samples	Number	(%)	Number	(%)	Number	(%)
Meat	51	6	(12)	1	(2)	29	(57)
Fish or seafood	37	1	(3)	2	(5)	8	(22)
Vegetable	38	0	(0)	13	(34)	16	(42)
Fermented pork or fish	11	0	(0)	9	(82)	1	(9)
Total	137	7	(5)	27^*	(20)	54^*	(39)

Total value shows the existence of significant deviation of the rate among four categories of foods.

Contamination rates of seven pathogens into four categories of foods were shown.

Table 2.

Comparison of the Contamination Rate of Foods Sold in Open Markets and Supermarkets

		Contaminated samples sold in
		Open markets		Supermarkets		Fisher's Direct
Categories of Foods	Pathogens	Numbers	(%)	Numbers	(%)	P-value
Meat	Escherichia coli	30/36	(83)	12/15	(80)	0.76
	Salmonella spp.	30/36	(83)	10/15	(67)	0.95
	Yersinia spp.	1/36	(3)	0/15	(0)	0.71
	Cronobacter sakazakii	1/36	(3)	0/15	(0)	0.71
	Listeria spp.	4/36	(11)	2/15	(13)	0.80
	Bacillus cereus	1/36	(3)	0/15	(0)	0.71
	Staphylococcous spp.	20/36	(56)	9/15	(60)	0.51
Fish or seafood	Escherichia coli	13/17	(76)	9/20	(45)	0.05
	Salmonella spp.	11/17	(65)	2/20	(10)	0.00
	Yersinia spp.	1/17	(6)	1/20	(5)	0.71
	Cronobacter sakazakii	1/17	(6)	0/20	(0)	0.46
	Listeria spp.	1/17	(6)	0/20	(0)	0.46
	Bacillus cereus	1/17	(6)	1/20	(5)	0.71
	Staphylococcous spp.	7/17	(41)	1/20	(5)	0.01
Vegetable	Escherichia coli	8/18	(44)	3/20	(15)	0.05
	Salmonella spp.	1/18	(6)	1/20	(5)	0.73
	Yersinia spp.	0/18	(0)	0/20	(0)	1.00
	Cronobacter sakazakii	3/18	(17)	1/20	(5)	0.26
	Listeria spp.	0/18	(0)	0/20	(0)	1.00
	Bacillus cereus	7/18	(39)	6/20	(30)	0.41
	Staphylococcous spp.	12/18	(67)	5/20	(25)	0.01

The contamination rate of each 7 pathogens into 3 categories of foods sampes was analized. The samples included in the category of “fermented pork or fish” did not compared because we purchased all of them from open market. A stastically significant difference of the contamination rates of Salmonella spp. or Staphylococcus spp. in fish or seafoods and Staphylococcus spp. in vegetable was observed.

In 137 collected samples, two Y. enterocolitica strains were isolated from one beef sample from an open market (2% of 51 meat samples) and from one shrimp sample from a supermarket (3% of 37 fish or seafood samples). In addition, six C. sakazakii strains were isolated from one meat sample from an open market (2% of 51 meat samples), from one fish sample from an open market (3% of 37 fish or seafood samples), and from four vegetable samples (11% of 38 vegetable samples), including three vegetable samples from open markets and one vegetable sample from a supermarket.

The highest prevalence of Staphylococcus spp. was found in meat samples (57%), while the lowest prevalence was found in fermented food samples (9%). Staphylococcus spp. was present at a greater frequency in vegetables and fish or seafood from open markets (67% and 41%, respectively) than from supermarkets (25% and 5%, respectively; Table 2). A higher rate of contamination by B. cereus was found in fermented food (82%) compared to meat, and fish or seafood (2% and 5%, respectively) (Table 1). There were 27 B. cereus strains isolated, and of those, 18 were found to produce diarrheal enterotoxin, but we were not able to detect the cereulide synthetic enzyme gene in any of the strains (Table 3).

Table 3.

Production of Diarrheal Enterotoxins and Presence of the Cereulide Synthetic Enzyme (CRS) Gene in Isolated B. cereus

Categories of foods	Number of tested sample	Number of isolated B. cereus	Number of enterotoxin producing strains
Meat	51	1	1
Fish or seafood	38	13	10
Vegetable	37	4	1
Fermented pork or fish	11	9	6
Total	137	27	18

A 67% of isolated B. cereus strains produced enterotoxin but no strain harbored CRS gene.

Seven Listeria spp. (including four isolates of L. innocua, one isolate of L. welshimeri, and two isolates of L. monocytogenes) were identified using 16S ribosomal DNA sequencing. Two L. monocytogenes strains were isolated from chicken and ground pork samples from open markets. L. monocytogenes was found at 4% prevalence in 51 meat samples, including 6% of 36 meat samples from open markets.

Discussion

Recently, Minami et al. (2010) found Salmonella spp. at a 30% prevalence (13 of 44 meat samples, including 48% of chicken samples, but not in any beef or pork samples) in open market meat samples and at a 25% prevalence (12 of 49 meat samples, including 57% of chicken, 24% of beef, and 12% of pork samples) in supermarket meat samples. Similar to this study, there were no statistically significant differences in the prevalence of Salmonella spp. in meat samples between the two types of market (p>0.05). In contrast, Vindigni et al. (2007) found that 93% of meat samples from open market (74 of 80 samples, including 85% of chicken, 100% of beef, and 93% of pork samples) were contaminated with Salmonella spp., while 57% of meat samples from supermarkets (40 of 70 samples, including 35% of chicken, 63% of beef, and 74% of pork samples) were contaminated. Results from these three studies suggest that improvement of hygienic conditions is required at retail outlets especially in open markets. Although we could not observe a significant difference in the prevalence of foodborne pathogens in meat between open markets (87%) and supermarkets (73%), open markets are more susceptible to cross-contamination than supermarkets due to constant exposure to environmental factors such as dust, rodents, and insects. However, if the products displayed at the open markets are received directly from the slaughter house and are sold out at the end of the selling day, the growth of the contaminating pathogens might be limited. Ultimately, efforts must be made by vendors to improve the levels of hygienic conditions to reach that of supermarkets at least.

The prevalence of Salmonella spp. (65%) and Staphyloccus spp. (41%) in fish or seafood from open markets was higher than that in samples from supermarkets (10% and 5%, respectively; Table 2). We observed a statistically significant (p<0.05) difference in the prevalence of foodborne pathogens in fish or seafood between open markets (65%) and supermarkets (10%). One reason for this difference in prevalence may be due to differences in food hygiene control (in the process of food cutting, for example) used in each type of market. In addition, the higher percentage of Staphylococcus spp. in open market samples could be a result of contamination of foods samples by unsuitable handling of foods.

This study revealed that the rate of contamination by Staphylococcus spp. in vegetables from open markets was higher than that in vegetables from supermarkets. No significant (p>0.05) difference in the prevalence of foodborne pathogens in vegetables between open markets (39%) and supermarkets (30%) was observed. The vegetables sold in supermarkets in Thailand are produced by companies on high-quality farms with good packaging and/or wrapping and distribution practices. These companies usually maintain cool temperatures throughout the distribution process and during display at the supermarket prior to purchase by the consumer. In contrast, vegetables sold in open markets are produced by private farms that have poor management of distribution practices and are sold by different vendors. Unwrapped vegetables are often placed directly on benches at ambient temperature until consumption. A higher prevalence of Staphylococcus spp. in open markets occurred due to human contamination of foods samples. Two vegetable samples, one from a supermarket and one from an open market, were contaminated with Salmonella spp. The prevalence of Salmonella in fresh vegetables and fruit is generally considered to be lower than in meat (Baird-Parker, 1991). Salmonella outbreaks are often caused by the consumption of meats; however, a foodborne salmonellosis outbreak caused by consumption of vegetables was reported in Thailand (Ministry of Public Health, 2007). Recently, the incidence of Salmonella in raw vegetables was reported to be 35% in Selangor, Malaysia (Salleh et al., 2003). The potential hazard of pathogenic bacteria present in vegetables should not be underestimated, particularly in those vegetables eaten raw or lightly cooked.

Listeria monocytogenes is the most important human pathogen within the genus Listeria. Among the other listerial species, L. ivanovii, L. seelegeri, and L. welshimeri are rarely pathogenic for humans, while L. grayi, L. innocua, and L. murrayi are considered to be non-pathogenic (Grahan and Collins, 1991). In this study, L. monocytogenes was found at 4% prevalence in 51 meat samples from open markets. In a previous report, L. monocytogenes was found at 8% prevalence in 297 meat samples, including 12% of 140 meat samples from open markets and 4% of 157 meat samples from supermarkets (Minami et al., 2010). Recently, Indrawattanaet al., (2011) reported a 15.4% prevalence of L. monocytogenes in 104 meat samples, supermarkets in Bangkok. Together, these results showed differences in the prevalence of L. monocytogenes in retailed meat samples from both types of market, for reasons including the environmental conditions of the collection area, the properties of the meat samples, the number of samples, and the time of collection.

Rice is arguably the most important foodstuff associated with B. cereus food poisoning (Adams and Moss, 2000). Because of the particular cultivation conditions in rice paddies, where B. cereus constitutes approximately 10% of the soil microflora (Varnam and Evans, 1991), raw rice is contaminated to varying degrees by B. cereus (Sarrias et al., 2002). Many outbreaks of B. cereus food poisoning have been associated with the bulk preparation of rice in advance of consumption or with storage at room temperature for an extended period of time (Adums and Moss, 2000; Mossel et al., 1991). In this study, a high prevalence of B. cereus was found in fermented food samples, and this prevalence might be due to contamination of sticky rice, which is an ingredient of traditional fermented pork and fish in Thailand. Thus, a critical step in the fermentation process is to reduce B. cereus contamination in cooked rice. However, no Y. enterocolitica, C. sakazakii, or Listeria spp. were isolated from fermented food, and Salmonella spp. and Staphylococcus spp. were detected at a low frequency. It is likely that lactic acid bacteria (LAB) will suppress the growth of Listeria spp. and other pathogenic bacteria because LAB can produce a variety of organic acids and bacteriocins. However, two Y. enterocolitica strains were isolated from beef and shrimp samples. Recently, Boonmar et al. (2003) found 2.4% chicken meat samples from retail markets in Thailand were contaminated with non-pathogenic Y. enterocolitica.

In this study, we observed substantial microbial contamination of retail foods from both open markets and supermarkets in four food categories. Improvement of hygienic conditions could reduce the risk of food poisoning or spoilage of foods purchased in Thai markets. This study documents the prevalence of microbial contamination in retail food samples in Thailand and provides a foundation for future studies. To provide information on the risk of eating Thai food, we will continue to study the level of contamination in Thai food in the future.

Footnotes

Acknowledgments

This work was supported by a grant from the Ministry of Agriculture, the Forestry and Fisheries of Japan (research project for ensuring food safety from farm to table DI-7301).

Disclosure Statement

No competing financial interests exist.

References

Adams

, Moss

. Bacillus cereus and other Bacillus species. Food Microbiology. Bodmin, UK: MPG Books, 2000; 187–192.

Baird-Parker

. Foodborne salmonellosis. Foodborne Illness—A Lancet Review. London: Edward Arnold, 1991.

Boonmar

, Bangtrakulnonth

, Kaneko

. The detection of Yersinia enterocolitica in chicken and swine in Thailand. J ISSAAS, 2003; 9:55–59.

Grahan

CGM

, Collins

. Listeriosis: Biology and implications for the food industry. Trends Food Sci Technol, 1991 April 89–93.

Indrawattana

, Nibaddhasobon

, Sookrung

, Chongsa-Nguan

, Tungtrongchitr

, Makino

, Tungyong

, Chaicumpa

. Prevalence of Listeria monocytogenes in raw meats marketed in Bangkok and characterization of the isolates by phenotypic and molecular methods. J Health Popul Nutr, 2011; 29:26–38.

Minami

, Chaicupa

, Chongsa-Nguan

, Samosornsuk

, Monden

, Takeshi

, Makino

, Kawamoto

. Prevalence of foodborne pathogens in open markets and supermarkets in Thailand. Food Control, 2010; 21:221–226.

Ministry of Public Health. Investigating Salmonellosis outbreak in school, Amphur Wang Nua: Lampang in November 2007. Wkly Epidemiol Surveill Rep, 2007; 39:185–359.

Mossel

, Corry

, Struijk

, Baird

. Food poisoning caused by Bacillus cereus and allied Bacilli, essentials of the microbiology of foods. A Textbook for Advanced Studies. New York: Wiley and Sons, 1991; 150–151.

Padungtod

, Kanene

. Salmonella in food animals and humans in north Thailand. Int J food Microbiol, 2006; 108:346–354.

10.

Salleh

, Rusul

, Hassan

, Reezal

, Isa

, Nishibuchi

, Radu

. Incidence of Salmonella spp. in raw vegetables in Selamgor, Malaysia. Food Control, 2003; 4:475–479.

11.

Sarrias

, Valero

, Salmeron

. Enumeration, isolation and characterisation of Bacillus cereus strains from Spanish raw rice. Food Microbiol, 2002; 19:589–595.

12.

Suthienkul

, Brown

, Seriwatana

, Tienthongdee

, Sastravaha

, Echeverria

. Shiga-like toxin–producing Escherichia coli in retail meats and cattle in Thailand. Appl Environ Microbiol, 1990; 56:1135–1139.

13.

Varnam

, Evans

. Bacillus. Foodborne Pathogens: An Illustrated Text. Varnam

, Evans

. London: Wolfe Publishing, 1991; 267–288.

14.

Vindigni

, Srijan

, Wongstitwilairoong

, Marcus

, Meek

, Riley

, Mason

. Prevalence of foodborne microoganisms in retail foods in Thailand. Foodborne Pathog Dis, 2007; 4:208–215.

15.

Vuddhakul

, Patararungrong

, Pungrasamee

, Jitsurong

, Morigaki

, Asai

, Nishibuchi

. Isolation and characterization of Escherichia coli O157 from retail beef and bovine feces in Thailand. FEMS Microbiol Lett, 2000; 182:343–347.