Escherichia Coli O157 and Salmonella Infections Associated with Sprouts in California

Escherichia coli O157 and Salmonella Infections Associated with Sprouts in California, 1996-1998

Janet C. Mohle-Boetani, MD, MPH; Jeff A. Farrar, DVM, PhD; S. Benson Werner, MD, MPH; Dzovag Minassian, MPH; Ray Bryant; Sharon Abbott; Laurence Slutsker, MD, MPH; Duc J. Vugia, MD, MPH, for the Investigation Team

Background:  In California, from 1996 through 1998, more than 50% of multicounty outbreaks with confirmed food vehicles were related to alfalfa or clover sprouts.

Objective:  To summarize investigations of sprout-associated outbreaks.

Design:  Matched case-control studies.

Setting:  California.

Patients:  Outbreak-associated patients and matched population controls.

Measurements:  Matched odds ratios and 95% CIs; traceback and environmental investigations of sprout and seed growers; and pulsed-field gel electrophoresis of isolates from patients, sprouts, and seeds.

Results:  Five sprout-associated outbreaks of salmonellosis and one outbreak of infection with nonmotile Shiga toxin-producing Escherichia coli O157 occurred. Six hundred patients had culture-confirmed disease, and two died. It is estimated that these outbreaks caused 22 800 cases of gastrointestinal illness or urinary tract infection. In the case-control studies, odds ratios for the association between illness and alfalfa sprout consumption ranged from 5.0 to infinity (all were statistically significant). Three sprout growers were implicated, and each was associated with two outbreaks. Outbreak strains of Salmonella were isolated from sprouts supplied by two sprout growers and from seeds used by the third sprout grower.

Conclusions:  As currently produced, sprouts can be a hazardous food. Seeds can be contaminated before sprouting, and no method can eliminate all pathogens from seeds. Seed and sprout growers should implement measures to decrease contamination. The general public should recognize the risks of eating sprouts, and populations at high risk for complications from salmonellosis or E. coli O157 infection should avoid sprout consumption.

Ann Intern Med. 2001;135:239-247.
 
Sprouts have been used medicinally in China for more than 5000 years and are now a staple of health-conscious people in the United States (1). However, raw sprouts have been the source of salmonellosis outbreaks in the United States, Canada, and Europe (2-8) and of Shiga toxin-producing Escherichia coli O157 infections in the United States and Japan (8-10). The U.S. Food and Drug Administration recently issued consumer advisories for sprouts and now recommends that sprout growers decontaminate seeds and obtain microbiologic testing of in-line or finished sprouts (11).

From 1996 through 1998, we investigated five multicounty outbreaks of salmonellosis and one outbreak ofE. coli O157 infection in California that were associated with alfalfa or clover sprouts. We present the results of these investigations so that clinicians will be aware of the hazards of consuming raw sprouts.

Methods

To identify food vehicles involved with multicounty outbreaks, we conducted a case-control study of each outbreak identified by California's Microbial Diseases Laboratory (MDL). We then conducted traceback investigations of sprouts eaten by patients and seeds used to produce the sprouts, as well as environmental investigations of implicated sprout growers and seed growers. The investigations were conducted to control an outbreak by identifying a food vehicle that could be removed from sale. Because of the public health urgency, the investigations were exempt from review by an institutional review board.

 
Identification of Outbreaks

In California, serotyping of all Salmonella isolates is required by regulation. The MDL serotypes most Salmonella isolates and identified the five salmonellosis outbreaks discussed here by noting an increase in the number of isolates of a particular serotype above that historically expected in a particular time frame and geographic area. The salmonellosis outbreaks were caused by S. enterica serotype Montevideo (outbreak 1, 417 patients, 1996), S. enterica serotype Meleagridis (outbreak 2, 75 patients, 1996), S. enterica serotype Senftenberg (outbreak 3, 60 patients, 1997-1998), S. enterica serotype Havana (outbreak 5, 18 patients, 1998), and S. enterica serotype Cubana (outbreak 6, 22 patients, 1998). The MDL also confirms isolation of E. coli O157 and conducts tests for Shiga toxin. The MDL detected a cluster ofE. coli O157 infections (outbreak 4, 8 patients, 1998) by an increase in the submission of E. coli O157 isolates that werenonmotile (lacking the flagellar antigen) but produced Shiga toxin I and II.

We reviewed the laboratory slips accompanying all isolates, which gave the clinical site of isolation of the organism and the age and sex of patients. We asked local health departments to collect data on symptoms, hospitalization, and outcomes for all patients involved in the outbreaks. To detect patients who were involved in these outbreaks but did not reside in California, health departments of bordering states and states that received implicated alfalfa sprouts and seeds (as far east as Maryland) identified potential infections with the same serotype of Salmonella or nonmotile E. coli O157 occurring at the same time as the California cases. The isolates of potential cases were sent to the MDL for further subtyping.

 
Laboratory Analysis

To identify the "outbreak strain" for each outbreak, we used pulsed-field gel electrophoresis (PFGE) and compared patterns with those of background isolates of the same Salmonella serotype or nonmotile strains of E. coli O157 from previous outbreaks. These background isolates were obtained from MDL (12, 13). The outbreak strain was defined as the predominant strain in an outbreak that had a specific PFGE pattern different from that of the background strains. For outbreaks 3 through 6, we used PFGE to characterize all isolates. Because outbreaks 1 and 2 were so large, we characterized only a sample of isolates from each. We used PFGE to test isolates submitted by other states, thereby identifying infected patients who did not live in California.

We used enrichment culture to test for Salmonella or E. coli O157 in sprouts that were aseptically sprouted in the laboratory from implicated seed lots for all outbreaks and from sprouts collected from the sprouting facility for outbreaks 1 through 4 (14). For outbreaks 1 and 2, we also ground samples of seeds and then tested them for Salmonella using enrichment media (14, 15).

 
Case-Control Studies

To identify a food associated with illness, we conducted matched case-control studies for outbreak 1 and outbreaks 3 through 6. We did not conduct a case-control study for outbreak 2. Although outbreak 2 was caused by a different serotype of Salmonella, it occurred at the same time and in the same geographic region as outbreak 1. In addition, four patients were co-infected with outbreak strains from outbreaks 1 and 2, and we suspected a common food vehicle for both outbreaks.

For each of the case-control studies, case-patients were infected with the outbreak strain, resided in the United States, and developed illness during the outbreak period (Table 1). The matching ratio of case-patients to controls was 1:1 for outbreak 1; 1:2 for outbreaks 3, 4, and 6; and 1:4 for outbreak 5. The matching ratio that was chosen correlates with the number of case-patients in the study; for studies with fewer case-patients, more controls per case-patient was selected. For all outbreaks, potential controls were included only if they had not had a diarrheal illness in the reference month. For outbreak 1, we matched controls to case-patients by age (+5 years). For outbreaks 3 through 6, we matched controls to case-patients by age (+10 years) and sex. For outbreaks 1, 4, and 6, we identified controls strictly by serial-digit telephone dialing, in which 1 is added or subtracted from each case-patient's telephone number. This method ensures that case-patients and controls have the same telephone exchange. For outbreaks 3 and 5, more than 50% of the controls were identified by asking case-patients for the names and telephone numbers of same-sex friends in their age group. If no healthy, same-sex, age group-matched friend controls could be identified, we obtained controls by serial-digit dialing or by contacting clinic physicians who had treated the case-patient.

For each outbreak, we first administered a hypothesis-generating questionnaire to a small number of patients (20 patients for outbreak 1 and 5 to 10 patients for outbreaks 3 through 6). The first questionnaire asked whether the patient had eaten any of 65 food items in the 4 days (outbreak 1) or 7 days (outbreaks 3 through 6) before onset of his or her illness. We used the results of each of these hypothesis-generating questionnaires to develop outbreak-specific questionnaires with approximately 20 food items. For these, we included only food items eaten by at least 30% (outbreak 1) or at least 50% (outbreaks 3 through 6) of the patients who answered the hypothesis-generating questionnaire. We then used the outbreak-specific questionnaires to query all case-patients and their matched controls. We conducted these interviews 1 to 2 months after onset of the case-patient's illness. We asked matched controls about consumption of foods throughout the calendar month of the matched case-patient's illness onset.

The methods used to investigate outbreaks 3 through 6 differed from those used to investigate outbreak 1 because the reference periods used to elicit food histories were different. Because some data suggest that the incubation period for a low dose of Salmonella can be as long as 7 days, we extended the period of inquiry about food exposures from 4 days (outbreak 1) to 1 week before illness onset (outbreaks 3 through 6) (16). With this extended incubation period, we elected to include only foods reported by at least 50% of case-patients in the hypothesis-generating interviews (rather than by >30%). Fewer hypothesis-generating interviews were needed to "pare down" the list of plausible food items.

 
Traceback and Environmental Investigations

We asked case-patients where they had purchased or consumed sprouts. We then visited retailers and reviewed invoices to identify sprout growers and visited implicated sprout growers. Sprout growers produce sprouts from seeds but do not grow the seeds. When sprouts are produced, hydroponic techniques are preferable to soil, and seeds should be disinfected. Sprout growers are considered food manufacturers and must adhere to regulations referred to as Good Manufacturing Practices. To determine where the implicated sprouts had been distributed, we reviewed the sprout sales records from the 2 to 3 weeks preceding the earliest onset date through the date of the most recent onset.

For each sprout grower, we reviewed invoices for seed purchases and records for sprout manufacturing processes to determine possible sources of contamination. We collected samples of implicated seeds from all sprout growers. For outbreaks 1 through 4, we also collected samples of sprouts at serial production points and stool samples from all employees. We inspected all implicated seed distributor plants and reviewed their records of receipt of seed. We then attempted to investigate implicated seed growers and fields.

 
Statistical Analyses

To identify a specific food associated with illness, we used the outbreak-specific questionnaires for each case-control study to calculate matched odds ratios (ORs) and 95% CIs for the association between each food item and illness (Epi-Info, Atlanta, Georgia) (17). For undefined ORs, we calculated a lower confidence limit by using exact methods (18).

 
Role of the Funding Source

The funding source had no role in the collection, analysis, or interpretation of the data or in the decision to submit the manuscript for publication.

Results
 
Outbreaks 1 and 2, Sprout Grower A

Sprout grower A was implicated by epidemiologic data for outbreak 1 and by laboratory data for outbreak 2. Case-patients in outbreak 1 were more likely than controls to have consumed alfalfa sprouts (38% vs. 11%; OR, 5.0 [95% CI, 1.4 to 27.0]). Of patients who ate sprouts, 71% reported eating sprouts from sprout grower A. The outbreak strain of S. Meleagridis (outbreak 2) was isolated from leftover alfalfa sprouts recovered from a patient and from alfalfa, clover, and alfalfa-clover mixed sprouts from sprout grower A.

Sprout grower A used soil and did not disinfect seeds before sprouting. The facility had evidence of rodents. Several months before outbreak 1, sprout grower A began periodically testing finished sprouts for Salmonella using a private laboratory. At the beginning of outbreak 1, this sprout grower received a laboratory report ofSalmonella but took no action. Employee stool cultures were negative forSalmonella.

Sprout grower A used one lot of alfalfa seed from the end of April through 2 July 1996, when production ceased. In California in 1996, 70% of culture-confirmed infections with both S. Montevideo (outbreak 1) and S.Meleagridis (outbreak 2) occurred between 1 May and 15 July, which correlates with the time that sprouts from the implicated seed lot were available for consumption. The alfalfa seeds, which were harvested from a farm in southern California, were fertilized with uncomposted chicken manure and irrigated with nonpotable canal water. At our inspection, approximately 1 year after the implicated seeds were harvested, the alfalfa fields had visible wildlife feces and runoff from livestock feces;S. enterica serotype Schwarzengrund and S. enterica serotype Infantis were isolated from dry chicken manure.

 
Outbreaks 3 and 4, Sprout Grower B

Sprout grower B was implicated through epidemiologic and laboratory data for outbreak 3 and through epidemiologic data for outbreak 4. Outbreak 3 persisted for 11 months (from September 1997 through July 1998), with two peaks in the number of cases. The first peak occurred from October to November 1997. In the case-control study, 63% of case-patients and 21% of controls consumed alfalfa sprouts (OR, 6.4 [CI, 1.3 to 31.8]). All patients who reported eating "alfalfa" sprouts ate sprouts from sprout grower B. The only type of "alfalfa" sprouts that sprout grower B produced were alfalfa-clover mixed sprouts. We reported our results to sprout grower B in December 1997 and recommended that alfalfa and clover seed lots used from September through December 1997 be discontinued. Cultures of sprouts and seeds that we collected in December 1997 did not yield Salmonella.

From January 1998 through May 1998, one to three cases of S. Senftenberg infection were reported per month. A peak of 20 cases occurred in June 1998 and July 1998. Through interviews, we found that patients who received a diagnosis of infection in June 1998 and patients who were infected during the first part of the outbreak had similar age, sex, geographic distribution, and sprout consumption histories. Sprouts from sprout grower B were recalled on 5 July 1998, and no case-patients had infection onset after 30 July 1998. The outbreak strain of S. Senftenberg was isolated from alfalfa-clover mixed sprouts collected on 5 July 1998 from a sprouting drum in sprout grower B's facility. Outbreak 3 was prolonged, probably because sprout grower B continued to use the seed lots that he was advised not to use after the first part of the outbreak was investigated.

In outbreak 4 (infection with nonmotile E. coli O157), 100% of case-patients and 0% of controls ate alfalfa sprouts (OR, undefined [CI, 1.2 to infinity]). Sprout grower B was one of several sources of sprouts for 25% of case-patients and was the sole source of sprouts for 75% of case-patients. As in outbreak 3, these patients actually ate alfalfa-clover mixed sprouts. Sprout grower B disinfected seeds irregularly. We found numerous violations of Good Manufacturing Practices within the facility. Stool cultures from all employees of sprout grower B tested negative for Salmonella and E. coli O157.

Several related lots of clover seed but no specific lot (or related lots) of alfalfa seed could explain outbreak 3. The source of the clover seeds for outbreaks 3 and 4 was a distributor who obtained seeds from farms in Oregon. To explore a possible association between the outbreak strain of Salmonella and strains found in agricultural products, we used PFGE to further characterize seven background strains of S. Senftenberg isolated from 1990 to 1995 from animal feed in Washington and Oregon. We found that three of the seven isolates from animal feed matched our outbreak strains of S. Senftenberg.

 
Outbreaks 5 and 6, Sprout Grower C

Sprout grower C was implicated through epidemiologic and laboratory data for outbreaks 5 and 6. In outbreak 5 (S. Havana), 67% of case-patients and 29% of controls ate alfalfa sprouts (OR, 10.0 [CI, 1.2 to 83.1]). Among case-patients who recalled eating sprouts, sprout grower C was one of several sources of sprouts for 25% and was the sole source of sprouts for 63%.

In outbreak 6 (S. Cubana), 78% of case-patients and 11% of controls ate alfalfa sprouts (OR, 6.0 [CI, 1.2 to 40.4]). Sprout grower C was one of several sources of sprouts for 80% of case-patients who recalled eating them. All case-patients in outbreak 6 who resided outside California, except those who resided in Maryland, either lived in or had recently visited areas where restaurants and stores had accounts with sprout grower C. The outbreak strains of both S.Havana and S.Cubana were isolated from the same lot of alfalfa seeds used by sprout grower C and from a sprout grower in Maryland during the outbreak period (May-September 1998); 50% of the case-patients from Maryland reported eating alfalfa sprouts.

Sprout grower C disinfected seeds with a sodium hypochlorite solution (200 to 2000 parts per million) and did not use soil. This sprout grower had minor violations of Good Manufacturing Practices and no evidence of factors that could contribute to contamination or cross-contamination. The implicated seeds were supplied by a seed mill from southern California that had also supplied seeds used by sprout grower A. The seeds supplied to sprout grower C were grown in southern California, but mixing of seed lots at the mill precluded identification of a single seed grower.

In total, 600 laboratory-confirmed illnesses occurred during these six outbreaks (Table 1). Affected persons were predominantly adults and women, and all outbreaks were multistate, primarily involving western states. Table 2 summarizes the clinical information. Most patients affected by the salmonellosis outbreaks had diarrhea, but in two outbreaks more than 30% of outbreak strains were isolated from urine. Up to 25% of affected patients were hospitalized during each outbreak, and two patients died of sepsis. Table 3 summarizes the epidemiologic investigations, and Table 4 summarizes the results of the traceback and laboratory investigations. In four outbreaks, the outbreak strains were isolated from sprouts or seeds. Outbreak Salmonella strains were isolated from sprouts grown by sprout growers A and B and from seeds used by sprout grower C.

Discussion

From 1996 through 1998, five multistate outbreaks of salmonellosis and one outbreak of E. coli O157 infection were associated with alfalfa or alfalfa-clover mixed sprouts in which the implicated sprout grower and most of the case-patients were located in California. These outbreaks represented more than 50% of the outbreaks that occurred during this time in California and involved food vehicles confirmed by epidemiologic or laboratory data. One outbreak was particularly large, affecting more than 400 culture-confirmed patients, and two outbreaks resulted in deaths from sepsis.

Our environmental investigations revealed problems with the way that sprouts are currently produced. At sprout grower A's facility, we found the outbreak strain of S.Meleagridis in both alfalfa and clover sprouts that were not mixed with each other, suggesting cross-contamination. For sprout grower B, we found the outbreak strain of S. Senftenberg in animal feed from Oregon and Washington supporting our hypothesis that the source of contamination for outbreak 3 was clover seed from Oregon. Finally, for sprout grower C, two outbreak-associated strains of Salmonella (S.Havana and S.Cubana) were found in seed lots used by this sprout grower.

Alfalfa and clover seeds are grown under the same field conditions as other agricultural products (such as forage crops for animal feed) and can be contaminated by unpasteurized fertilizers, nonpotable water, or grazing livestock. After harvest, seeds are stored in warehouses in open containers where they can be exposed to rodent feces. Equipment used to remove debris from seeds is not routinely disinfected. Although sprout growers should disinfect seeds before sprouting to reduce the load of pathogens, there is currently no reliable method of eliminating all pathogens. Since sprouts are produced under incubator-type conditions with high moisture and warm temperatures, even low numbers of Salmonella and E. coli that survive disinfection can multiply rapidly to high levels. Most consumers and retailers do not cook sprouts, and since bacteria on the seed surface can become internalized during sprouting, washing sprouts is probably an ineffective way to eliminate pathogens.

The six outbreaks discussed in this paper represent 75% of the sprout-associated outbreaks identified in the United States during the same period, although only 10% of the approximately 450 sprout growers in the United States operate in California. Possible reasons for the overrepresentation include a greater consumption of sprouts in California (19); better detection of outbreaks because of a large catchment area and required submission of Salmonella isolates; or differences in sprout-growing practices, such as seed disinfection.

In these outbreaks, we found that demographic features were similar to those seen in other sprout-associated outbreaks of salmonellosis predominantly involving adult women (2-5). In two outbreaks, more than 30% of patients had cystitis but not diarrhea; low-dose contamination could cause asymptomatic colonization of the gastrointestinal tract, which could then cause an ascending infection of the urethra. This high proportion of urinary tract infections was not noted in previous sprout-associated outbreaks. Patients with bacteruria may have been excluded from other investigations. Alternatively, some serotypes of Salmonella may be able to enter the urinary tract more easily than others (20, 21).

The detection of case-patients for the E. coli O157 cluster was limited because these isolates are not required to be submitted to the MDL. Therefore, the magnitude of this outbreak has probably been underestimated. The epidemiologic studies are limited because interviews were conducted 1 to 2 months after the reference period, and recall is not perfect. In addition, for the two large outbreaks (outbreaks 1 and 2), we could not interview all case-patients because of resource limitations. Recall of sprouts may be particularly difficult since they are often used as a condiment for salads and sandwiches. However, our finding of outbreak strains in either sprouts or seeds for all three implicated sprout growers mitigates these limitations.

In general, only a small fraction of persons with salmonellosis (2.6%) receive laboratory confirmation because they do not report to their physicians or because stools are not cultured (22). We estimate that the 600 culture-confirmed infections discussed in this study represent approximately 22 800 persons who had gastrointestinal illnesses or urinary tract infections related to these outbreaks.

Although production of pathogen-free sprout seeds is not currently feasible, seed growers and seed mills could reduce the risk for contamination of seeds intended for human consumption by 1) identifying seed crops intended for human consumption before planting; 2) using good agricultural practices for seed production, including synthetic fertilizers and potable water; and 3) insisting that seed mills maintain lot integrity throughout the seed-cleaning process by cleaning and disinfecting seed milling equipment before cleaning seeds intended for sprouting. Sprout growers can decrease the risk for contamination by 1) following a Hazard Analysis and Critical Control Points approach for sprout production, which should include disinfection of seeds with a calcium hypochlorite solution (20 000 parts per million) or an equivalent process; 2) performing microbiological tests on the irrigation water or finished product, as recommended in the current U.S. Food and Drug Administration guidelines; and 3) refrigerating sprouts (at < 23 °C or < 41 °F) from harvest to consumption (11, 23).

Government agencies should continue to inspect sprout growers, investigate outbreaks, implement such preventive measures as warning labels on sprout packages, and prohibit the serving of raw sprouts in facilities that serve populations at high risk for complications from Salmonella or E. coli O157 infections. Research should continue on decontamination methods for seeds and sprouts, including irradiation, ozone treatment, and the use of competitive microflora.

Although sprouts are often touted as a health food, they are potentially hazardous and can cause significant morbidity and even death. Fewer adults in California consume alfalfa sprouts than other foods known to cause outbreaks of foodborne illness (19). However, from 1996 to 1998, sprouts caused more multicounty outbreaks than any other vehicle. Since reliable methods for disinfecting seeds or sprouts are not currently available, clinicians and the general public should recognize that as currently produced, raw sprouts are an inherently dangerous food, especially for people at risk for high morbidity from salmonellosis or E. coli O157 infection, such as young children, immunocompromised persons, and elderly persons. Raw sprouts should not be served in facilities for high-risk persons, such as child care centers, preschools, kindergartens, assisted living facilities, nursing homes, long-term care facilities, and hospitals. Consumers who want to decrease their risk for foodborne illness should avoid eating raw sprouts (24).

Appendix

The members of the Investigation Team were Howard Backer, MD, MPH, Greg Inami, and Michael Gutierrez, California Department of Health Services; Eric Mouzin, MD, MPH, Centers for Disease Control and Prevention; and Roshan Reporter, MD, MPH, Los Angeles County Department of Health Services.

Author and Article Information

From California Department of Health Services, Berkeley, and Los Angeles County Department of Health Services, Los Angeles, California; and Centers for Disease Control and Prevention, Atlanta, Georgia.

For members of the Investigation Team, see Appendix.

Acknowledgments: The authors thank Carmela Groves, Barbara Brynan, Ken Komatsu, Pamela D. Young, and Ellen Kenney for providing information on patients; Mark Miller, Dennis Tan, Rodney Knight, Sabrina Gilliam, and Stephanie Litwak for conducting patient interviews; Liga Kilman and Eleanor Lehnkering for laboratory support; and Barbara Sanderson for providing information on the history of sprout consumption.

Grant Support: By general public funds through the California Department of Health Services, a state agency.

Requests for Single Reprints: Janet Mohle-Boetani, MD, MPH, Disease Investigations and Surveillance Branch, Division of Communicable Diseases Control, 2151 Berkeley Way, Room 708, Berkeley, CA 94704;

Current Author Addresses: Drs. Mohle-Boetani, Werner, and Vugia, Ms. Minassian, Mr. Bryant, and Ms. Abbott: Disease Investigations and Surveillance Branch, Division of Communicable Diseases Control, 2151 Berkeley Way, Room 708, Berkeley, CA 94704.

Dr. Farrar: California Department of Health Services, Food and Drug Branch, 601 North 7th Street, Sacramento, CA 94234.

Dr. Slutsker: Centers for Disease Control and Prevention, 1600 Clifton Road, Mailstop F-22, Atlanta, GA 30333.

Author Contributions: Conception and design: J.C. Mohle-Boetani, J.A. Farrar, S.B. Werner, D. Minassian, D.J. Vugia.

Analysis and interpretation of the data: J.C. Mohle-Boetani, J.A. Farrar, S.B. Werner, D. Minassian, R. Bryant, D.J. Vugia.

Drafting of the article: J.C. Mohle-Boetani, J.A. Farrar, D.J. Vugia.

Critical revision of the article for important intellectual content:

J.C. Mohle-Boetani, J.A. Farrar, S.B. Werner, D. Minassian, L. Slutsker, D.J. Vugia.

Final approval of the article: J.C. Mohle-Boetani, S.B. Werner, D. Minassian, D.J. Vugia.

Provision of study materials or patients: D. Minassian, R. Bryant.

Statistical expertise: D. Minassian.

Administrative, technical, or logistic support: S.B. Werner, D. Minassian, R. Bryant, S. Abbott, L. Slutsker, D.J. Vugia.

Collection and assembly of data: J.C. Mohle-Boetani, J.A. Farrar, D. Minassian, S. Abbott.

Keywords: Escherichia coli O157; Alfalfa sprouts; Food contamination; Salmonella; Disease outbreaks

References
 
1. Kurtzweil P. Questions keep sprouting about sprouts. FDA Consum. 1999;33:18-22. [PMID: 10030142]
 
2. Mahon BE, Pönkä A, Hall WN, Komatsu K, Dietrich SE, Siitonen A, et al. An international outbreak of Salmonella infections caused by alfalfa sprouts grown from contaminated seeds. J Infect Dis. 1997;175:876-82. [PMID: 9086144]
 
3. Glynn MK, Patrick S, Wuhib T, Neiman J, Flahart R, Boxrund D, et al. An outbreak of Salmonella serotypes Anatum and Infantis infections associated with eating contaminated sprouts [Abstract]. 36th Annual Meeting of the Infectious Diseases Society of America (IDSA), November 1998, Denver, Colorado. Abstract no. 539.
 
4. Van Beneden CA, Keene WE, Strang RA, Werker DH, King AS, Mahon B, et al. Multinational outbreak of Salmonella enterica serotype Newport infections due to contaminated alfalfa sprouts. JAMA. 1999;281:158-62. [PMID: 9917119]
 
5. Buck P, Grimsrud K, Waters J, Cardinal R, Talbot J, Anand C, et al. Would you like a little Salmonella with your sandwich? [Abstract] In: Program and Abstracts of the 47th Annual Epidemic Intelligence Service Conference, International Night. Atlanta, GA; 1998.
 
6. Pönkä A, Andersson Y, Siitonen A, de Jong B, Jahkola M, Haikala O, et al.Salmonella in alfalfa sprouts [Letter]. Lancet. 1995;345:462-3. [PMID: 7853988]
 
7. Puohiniemi R, Heiskanen T, Siitonen A. Molecular epidemiology of two international sprout-borne Salmonella outbreaks. J Clin Microbiol. 1997;35:2487-91. [PMID: 9316894]
 
8. Taormina PJ, Beuchat LR, Slutsker L. Infections associated with eating seed sprouts: an international concern. Emerg Infect Dis. 1999;5:626-34. [PMID: 10511518]
 
9. Outbreaks of Escherichia coli O157:H7 infection associated with eating alfalfa sprouts-Michigan and Virginia, June-July, 1997. MMWR Morb Mortal Wkly Rep. 1997;46:741-4.
 
10. Michino H, Araki K, Minami S, Takaya S, Sakai N, Miyazaki M, et al. Massive outbreak of Escherichia coli O157:H7 infection in schoolchildren in Sakai City, Japan, associated with consumption of white radish sprouts. Am J Epidemiol. 1999;150:787-96. [PMID: 10522649]  
 
11. FDA issues guidance to enhance safety of sprouts. U.S. Food and Drug Administration. Press release. 25 October 1999. Available at
 
12. Tenover FC, Arbeit RD, Goering RV, Mickelsen PA, Murray BE, Persing DH, et al. Interpreting chromosomal DNA restriction patterns produced by pulsed-field gel electrophoresis: criteria for bacterial strain typing. J Clin Microbiol. 1995;33:2233-9. [PMID: 7494007]  
 
13. Swaminathan B, Matar GM. Molecular typing methods. In: Persing DH, Smith TF, Tenover FC, White TJ, eds. Diagnostic Molecular Microbiology: Principles and Practices. Washington, DC: American Society of Microbiology; 1993.
 
14. Inami GB, Moler SE. Detection and isolation of Salmonella from naturally contaminated alfalfa seeds following an outbreak investigation. J Food Prot. 1999;62:662-4. [PMID: 10382657]  
 
15. Andrews WH, June GA, Sherrod PS, Hammack TS, Amaguana RM. Salmonella. In: Bacteriological Analytical Manual. Federal Food and Drug Administration. Gaithersburg, MD: Association of Official Analytical Chemists International; 1995:5.01-5.20.
 
16. O'Mahony M, Barnes H, Stanwell-Smith R, Dickens T, Jephcott A. An outbreak of Salmonella heidelberg infection associated with a long incubation period. J Public Health Med. 1990;12:19-21. [PMID: 2390305]  
17. Epi Info. Version 6.04b. Atlanta, GA: Centers for Disease Control and Prevention; 1995.
 
18. Martin D, Austin H. An efficient program for computing conditional maximum likelihood estimates and exact confidence limits for a common odds ratio. Epidemiology. 1991;2:359-62. [PMID: 1742385]
 
19. Foodborne Diseases Active Surveillance Network (FoodNet). Population Survey Atlas of Exposures: 1998-1999. Centers for Disease Control and Prevention. Atlanta, GA: Centers for Disease Control and Prevention; 1999:6.
 
20. Abbott SL, Portoni BA, Janda JM. Urinary tract infections associated with nontyphoidal Salmonella serogroups. J Clin Microbiol. 1999;37:4177-8. [PMID: 10565958]
 
21. Backer HD, Mohle-Boetani JC, Werner SB, Abbott SL, Farrar J, Vugia DJ. High incidence of extra-intestinal infections in a SalmonellaHavana outbreak associated with alfalfa sprouts. Public Health Rep. 2000;115:339-45. [PMID: 11059427]  
 
22. Mead PS, Slutsker L, Dietz V, McCaig LF, Bresee JS, Shapiro C, et al. Food-related illness and death in the United States. Emerg Infect Dis. 1999;5:607-25. [PMID: 10511517]
 
23. FDA Model Food Code. U.S. Food and Drug Administration. 1999.

24. Consumers advised of the risks associated with raw sprouts. U.S. Food and Drug Administration. Press release. 9 July 1999.