Printer Friendly

An outbreak of salmonella javiana associated with consumption of watermelon.

Foodborne salmonellosis outbreaks associated with fruits or other produce are rare. Less than two percent of more than 340 salmonellosis outbreak reports to the Centers for Disease Control and Prevention (CDC) from 1983 to 1987 implicated fruits or vegetables as a vehicle of transmission (1). Watermelons in particular are an unusual vehicle; in a recent review, the U.S. Food and Drug Administration (FDA) found only three published reports of salmonellosis outbreaks associated with watermelons in the past 40 years (2). The following report describes a recent outbreak in which watermelon was implicated epidemiologically and microbiologically.


On June 17 and 18, 1991, the Oakland County (Michigan) Health Division received reports of salmonellosis in three children. All three attended the same elementary school and had illness onset within the same 24-hour period. Additional cases of gastro-intestinal illnesses among other children, mostly morning kindergarten students, were identified in following days. Most ill persons had participated in either or both of two parties: an indoor picnic for the morning kindergarten classes on June 11, and an in-school party on June 12.


Data collection

An epidemiologic study was conducted to explore associations between food or activity exposures on June 11--12 and illness among the morning kindergarten students, staff and visitors present on those days. A list of foods served was obtained and a questionnaire was designed to assess standard demographic variables, food consumption and activity histories on the dates in question, and illness experience. The questionnaire was administered by telephone.

Case definition

A presumptive case was defined as any attendee of the kindergarten classes on June 11 or 12 reporting at least two of the following symptoms between June 11 and June 17: diarrhea, abdominal cramps, fever, vomiting or headache. A confirmed case was any such person who submitted a stool specimen from which Salmonella was cultured.

Epidemiologic analysis

Cases were compared to non-ill persons with respect to activity and food consumption history. Food-specific attack rates and corresponding relative risks were computed to measure associations between exposure factors and illness. Ninety-five percent confidence intervals of the relative risks and 2x2 chi-square tests were used to assess statistical significance. Persons whose reported illness failed to meet the case definition were excluded from analysis. Epi Info software (version 5.01) was used for data management and statistical analysis.

Laboratory and environmental investigation

Stool specimens were collected from willing ill persons. Bacterial cultures for Salmonella species were performed by standard methods at the Oakland County Health Division laboratory. Controlled plasmid profile and chromosomal DNA restriction analyses were performed, by methods described elsewhere (3), at the Michigan Department of Public Health Molecular Biology Laboratory. An environmental investigation reviewed the sources of foods served and preparation methods.


Epidemiologic analyses

Fifty-one of 56 enrolled morning kindergarten students were interviewed, along with six non-students (three teachers, two parents, one sibling). Eight were excluded from analysis because their illness failed to meet the case definition. Twenty-one cases were identified in the study, an adjusted attack rate of 43 percent; 12 were confirmed.

Food-specific attack rates are presented in Table 1. Watermelon consumed at the picnic on June 11 showed the largest difference in attack rates (62% vs. 15% ill, relative risk = 4.02, 95% confidence limits 1.08 and 15.04). No other food showed a statistically significant association with illness.

In addition to cases included in the epidemiologic study, five more primary cases (persons who ate leftovers of the incriminated food at two subsequent exposure events) were identified: three from a birthday party where the watermelon leftovers were available on the afternoon of June 12; and two from a neighboring household to which leftovers were given (received on June 14, eaten on June 17). Three of these five were laboratory or culture confirmed. Thirteen secondary cases were identified among household contacts of cases (five consumed, eight presumptive).

The epidemic curve is illustrated in Figure 1. Based on 10 cases with a single known exposure time, the mean incubation period was 49 hours (range 32--63 hours). The duration of the outbreak was 24 days, TABULAR DATA OMITTED involving a total of 39 identified cases (26 primary, 13 secondary). A frequency distribution of reported symptoms is given in Table 2.

Laboratory investigation

All positive stool cultures were typed as S. javiana. A sample of the epidemiologically-incriminated watermelon grew S. javiana. Plasmid profile analysis and chromosomal DNA restriction analysis comparing the watermelon isolate with six representative stool culture isolates revealed identical plasmid profiles and identical restriction patterns.

Environmental investigation

A parent purchased the watermelon whole from a local supermarket on June 10. It was brought to the school whole on the morning of June 11 and cut with a clean knife on a clean plastic cutting board, both from the purchaser's home. It was not washed prior to cutting. Watermelon slices were served on a clean tray at room temperature over approximately a three hour period. Some remaining watermelon was sliced and placed in a plastic-ware container, then kept in the school refrigerator. This portion was available the following day at a last-day-of-school party in two of the three morning kindergarten classes; it was then brought to a birthday party the afternoon of June 12. The rest of the melon was brought home by the purchaser, kept in a refrigerator, and given to a neighbor on June 14.

No illnesses were reported in the purchaser's household prior to or at the time the melon was bought. No means of cross-contamination at the time it was cut and served could be identified. No information was available at the supermarket regarding shipping conditions of the melon. The purchaser of the melon did not observe any defect in the melon rind.


The source of contamination of the watermelon is unknown. Possibilities include contamination in the field, in transport, at the retail site, or at the time it was prepared for serving. FDA survey studies of imported watermelons and cantaloupes have found evidence of Salmonella contamination on the rind (2, 4), presumably originating in the field; this, or contamination in transport, is the most likely source in this outbreak, as no evidence was found of contamination at the retail site or after purchase. No other concurrent outbreaks of S. javiana were reported in Michigan or neighboring states (personal communication with Michigan Dept. of Public Health and U.S. FDA), suggesting contamination may have been an isolated event involving this melon alone.

Few watermelon-associated salmonellosis outbreaks have been reported; this appears to be the largest. In only one other instance was epidemiologic evidence of the vehicle corroborated by microbiological studies of clinical specimens and the implicated watermelon. Experiments done in conjunction with that outbreak, involving 17 cases of S. miami (in Massachusetts in 1953), established watermelon was capable of supporting Salmonella growth (5). In a 1950 Minnesota outbreak, Salmonella infections of an unspecified serotype in six persons were associated with consumption of watermelon purchased from a roadside stand; cultures of the melon grew S. bareilly. An outbreak of S. oranienburg in 1979 in Illinois affecting 18 persons was linked to watermelon sections purchased at a single supermarket (6). Other recent salmonellosis outbreak investigations involving fruit or vegetables have implicated cantaloupes (4, 7) and tomatoes (8) as vehicles of transmission.

Regardless of the contamination source, extended time at room temperature probably facilitated bacterial proliferation. According to the CDC, improper storage or holding temperature is the leading cause of foodborne illness outbreaks in the U.S. (1). Although such outbreaks are infrequent, physicians and outbreak investigators should consider fruits as potential sources of salmonellosis. This outbreak lends timely support to recent FDA recommendations which advise washing melon skins before serving, using sanitized utensils and surfaces for slicing, and maintaining proper temperatures for cut melons before, during and after service (2).
Table 2. Frequency distribution of symptoms in cases(*)
Symptom Number Percent
Diarrhea 37 100
Abdominal cramps 34 92
Nausea 28 76
Fever 28 76
Fatigue 28 76
Headache 24 65
Vomiting 22 59
Chills 15 41
Muscle aches 16 43
Joint aches 15 41
Dizziness 7 19
* Data available for 37 of 39 cases


1. CDC (1990), Foodborne disease outbreaks, 5-year summary, 1983-1987. In: CDC surveillance summaries, MWWR 1990;39(no. SS-1):15-57.

2. FDA memorandum July 17, 1991, "Update on cantaloupe (melon/interpretation)."

3. Martin, R., F. Downes, S. Dietrich and B. Robinson (1991), Molecular epidemiology of a multi-state outbreak of Salmonella poona. Presented at North Central Regional Epidemiology Conference; October 16-18, 1991, Chicago, IL.

4. CDC (1991), Multi-state outbreak of Salmonella poona infections -- States and Canada, 1991. MWWR; 40:549-52.

5. Gaylor, G.E., R.A. MacCready, J.P. Reardon and B.F. McKernan (1955), An outbreak of salmonellosis traced to watermelon. Public Health Reports 70:311-313.

6. CDC (1979), Salmonella oranienburg gastroenteritis associated with consumption of pre-cut watermelons -- Illinois. MWWR; 28:522-23.

7. Ries, A.A., S. Zaza, C. Langkop, R.A. Tauxe and P.A. Blake (1990), A multi-state outbreak of Salmonella chester linked to imported cantaloupe (Abstract). In: American Society for Microbiology, Program and abstracts of the 30th Interscience Conference on Antimicrobial Agents and Chemotherapy. Washington, D.C. American Society for Microbiology.

8. Wood, R.C., C. Hedberg, K. White, K. MacDonald and M. Osterholm (1991), A multi-state outbreak of Salmonella javiana infections associated with raw tomatoes (Abstract). In: CDC, Epidemic Intelligence Service 40th Annual Conference. Atlanta, U.S. Department of Health and Human Services, Public Health Service.
COPYRIGHT 1993 National Environmental Health Association
No portion of this article can be reproduced without the express written permission from the copyright holder.
Copyright 1993, Gale Group. All rights reserved. Gale Group is a Thomson Corporation Company.

Article Details
Printer friendly Cite/link Email Feedback
Author:Blostein, Joel
Publication:Journal of Environmental Health
Date:Jul 1, 1993
Previous Article:Temporary public eating places: food safety knowledge, attitudes and practices.
Next Article:Food for thought.

Related Articles
Molecular typing of multidrug-resistant Salmonella Blockley outbreak isolates from Greece.
Competitive Exclusion of Salmonella Enteritidis by Salmonella Gallinarum in Poultry.
A multistate outbreak of Salmonella enterica serotype Baildon associated with domestic raw tomatoes. (Dispatches).
Outbreak of Salmonella serotype Anatum infection associated with unpasteurized orange juice.
Salmonella enterica serotype Enteritidis phage type 4b outbreak associated with bean sprouts. (Dispatches).
First incursion of Salmonella enterica serotype Typhimurium DT160 into New Zealand. (Dispatches).
Molecular epidemiologic surveillance of salmonellosis in Arkansas.
Hospitalization and antimicrobial resistance in Salmonella outbreaks, 1984-2002.
Salmonella Agona outbreak from contaminated aniseed, Germany.
Web-based surveillance and global Salmonella distribution, 2000-2002.

Terms of use | Copyright © 2016 Farlex, Inc. | Feedback | For webmasters