Copyright 2002 American Medical Association. All Rights Reserved. Applicable FARS/DFARS Restrictions Apply to Government Use.2002American Medical AssociationThis is an open-access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
Three multistate outbreaks of Salmonella serotype Poona infections associated with eating cantaloupe imported from Mexico occurred in the spring of consecutive years during 2000-2002. In each outbreak, the isolates had indistinguishable pulsed-field gel electrophoresis (PFGE) patterns; the PFGE patterns observed in the 2000 and 2002 outbreaks were indistinguishable, but the pattern from 2001 was unique among them. Outbreaks were identified first by the California Department of Health Services (2000 and 2001) and the Washington State Department of Health (2002) and involved residents of 12 states and Canada. This report describes the investigations, which led ultimately to an import alert on cantaloupes from Mexico. To limit the potential for cantaloupe contamination, the Food and Drug Administration (FDA) continues to work with the Mexican government on a food-safety program for the production, packing, and shipping of fresh cantaloupes.
A total of 47 confirmed cases of S. Poona infections with indistinguishable PFGE patterns were identified from California (26), Washington (10), Nevada (five), New Mexico (three), Oregon (two), and Colorado (one), with illness onset occurring during April 14–June 2. The median age of ill persons was 7 years (range: 1-95 years); 28 (60%) patients were aged <10 years, and nine (19%) were aged >60 years. Twenty-four (51%) patients were male and nine (19%) were hospitalized.
A matched case-control study was conducted; 20 case-patients were matched by age category to 37 community controls. A case was defined as laboratory-confirmed infection with S. Poona of the outbreak PFGE pattern in a person with illness onset during April-June. By multivariable modeling, illness was associated only with eating cantaloupe (matched odds ratio [MOR] = 6.7; 95% confidence interval [CI] = 1.3-34.0), with 16 (80%) case-patients versus seven (19%) controls reporting eating cantaloupe. Cantaloupe was purchased either pre-cut or whole.
In April, an initial cluster of S. Poona was identified in California. Isolates had a rare biochemical trait, the inability to produce hydrogen sulfide (H2S), and PFGE patterns that were indistinguishable. A total of 50 cases of H2S-negative S. Poona infections were identified in residents of California (28), Washington (eight), Nevada (seven), Arizona (six), and Oregon (one). Demographic and illness-history data from the 28 California patients indicated that illness onset occurred during April 6–May 28. The age distribution was bimodal; the 19 children had a median age of 3 years (range: 1-5 years) and the nine adults had a median age of 80 years (range: 39-91 years). Fifteen (54%) patients were female. Ten (36%) patients were bacteremic; one infant girl had S. Poona isolated from a urine specimen. Nine (33%) patients were hospitalized, and two patients (a man aged 78 years and a woman aged 91 years) died with Salmonella septicemia.
A matched case-control study was conducted; 11 case-patients from California (seven), Nevada (two), Arizona (one), and Washington (one) were matched by age category to 19 community controls. Case-patients had laboratory-confirmed infections of the outbreak strain of H2S-negative S. Poona and illness onset during the first 2 weeks of April. Illness was associated only with eating cantaloupe (MOR = 7.4; 95% CI = 1.0-178.0). Eight (80%) case-patients and six (33%) controls recalled eating cantaloupe. Cantaloupe was purchased either pre-cut or whole.
A total of 58 cases with S. Poona isolates with indistinguishable PFGE patterns were identified in California (21), Washington (nine), Oregon (five), British Columbia (four), Colorado (three), Nevada (three), Manitoba (two), Missouri (two), Ontario (two), Saskatchewan (two), Texas (two), Arkansas (one), Minnesota (one), and Vermont (one). Illness onset occurred during March 30–May 31; the median age of patients was 6 years (range: 4 months–91 years); 32 (55%) were aged <10 years, and 11 (19%) were aged >60 years. A total of 31 (55%) were female. Ten patients were hospitalized.
A matched case-control study was conducted; 27 case-patients were matched by age category to 54 community controls. A case was defined as S. Poona infection with the outbreak PFGE pattern in a person aged ≥2 years with illness onset during March 15–May 3. The only exposure significantly associated with illness was eating cantaloupe; 20 (74%) case-patients recalled eating cantaloupe compared with 11 (20%) controls (MOR = 15.5; 95% CI = 3.3-125.0). Case-patients (50%) were more likely than controls (13%) to eat cantaloupe purchased whole (MOR = 5.8; 95% CI = 1.6-23.3) or to eat cantaloupe in a fruit salad or as a garnish (28% versus 5%) (MOR = 6.5; 95% CI = 1.2-63.0). No other factors were significantly associated with illness.
FDA, in conjunction with state and provincial food regulatory agencies, conducted traceback investigations of cantaloupe purchased by patients in all three outbreaks. In each instance, point-of-sale sources of cantaloupe were traced back to shippers and then to farms in Mexico. In response to the 2000 and 2001 outbreaks, FDA conducted on-farm investigations in Mexico and concluded that measures were not in place to minimize microbial contamination in the growing, harvesting, packaging, and cooling of cantaloupe. Possible sources of contamination include irrigation of fields with water contaminated with sewage, processing (cleaning and cooling) produce with Salmonella-contaminated water, poor hygienic practices of workers who harvest and process the cantaloupe, pests in packing facilities, and inadequate cleaning and sanitizing of equipment that comes in contact with cantaloupe. In association with the 2001 outbreak, FDA detained product imported by the shipper on May 31, and the shipper voluntarily recalled its imported Mexican cantaloupe. The shipper and the implicated farm in Mexico remain on detention. In association with the 2002 outbreak, the importer voluntarily recalled the implicated Mexican cantaloupe, and FDA placed the implicated farms on detention. On October 28, 2002, FDA issued an import alert on cantaloupe from Mexico that detains all products offered for entry at all U.S. ports.
SM Anderson, MPH, Arizona Dept of Health Svcs. L Verchick, MS, Clark County Health Department, Las Vegas; R Sowadsky, MSPH, Nevada State Health Div. B Sun, DVM, R Civen, MD, JC Mohle-Boetani, MD, SB Werner, MD, M Starr, DVM, S Abbott, M Gutierrez, M Palumbo, PhD, J Farrar, PhD, California Dept of Health Svcs. P Shillam, Colorado Dept of Health. E Umland, MD, M Tanuz, M Sewell, DrPH, J Cato, New Mexico Dept of Health. W Keene, PhD, Oregon Dept of Human Svcs. M Goldoft, MD, J Hofmann, MD, J Kobayashi, MD, P Waller, MS, Washington State Dept of Health. Center for Food Safety and Applied Nutrition and the Office of Regulatory Affairs, Food and Drug Administration. C Braden, MD, Div of Bacterial and Mycotic Diseases, National Center for Infectious Diseases; G Djomand, MD, M Reller, MD, W Chege, MD, EIS officers, CDC.
Salmonella infections have been linked to melons at least since 1990 when Salmonella serotype Chester traced to cantaloupe caused 245 illnesses in 30 states.1 The cantaloupe were imported from either Mexico or Guatemala. In 1991, an outbreak of cantaloupe-associated S. Poona infections caused 400 illnesses in 23 states.2 Illness was associated with eating pre-cut cantaloupe in fruit salads or from salad bars. Although industry sources identified the lower Rio Grande Valley in Texas as the probable source of the implicated cantaloupe, some might have come from Mexico. In response to this outbreak, FDA conducted a microbiologic survey that isolated a variety of Salmonella serotypes from approximately 1% of sampled imported cantaloupe and watermelon.2 In 1997, an outbreak of Salmonella serotype Saphra infections affected 25 persons in California. Illness was associated with cantaloupe imported from Mexico.3 After the 2000 and 2001 S. Poona outbreaks, FDA conducted farm investigations in Mexico, issued press releases to warn consumers, placed implicated farms on detention, and conducted sampling surveys of imported cantaloupe. The 1999 and 2000 FDA surveys of imported produce indicated that 5% of cantaloupe sampled (eight of 151) was contaminated with Salmonella.4 A 2001 survey of imported produce indicates that of 29 cantaloupes from Mexico tested, none yielded Salmonella, Shigella, or Escherichia coli O157:H7 (FDA, unpublished data, 2001). The interpretation of the 2001 survey is limited by of the small sample size.
S. Poona is a relatively rare serotype that is responsible for 1% of human Salmonella isolates reported in the United States in 2001; however, of the six cantaloupe-associated Salmonella outbreaks, four were attributed to infections with S. Poona. Typically, human infection with S. Poona is associated with reptile exposure.5,6 The three outbreaks attributed to S. Poona-contaminated cantaloupe traced to Mexican farms suggest the possibility of a unique natural reservoir in the Mexican farm environment, possibly from reptiles such as iguanas drawn to feed on melon crops that enter the packing sheds and contaminate the equipment. Subsequently, water used in the washing and cooling process might spread the contamination.
FDA provides information about the decontamination of melons to the retail industry, food-service establishments, and commercial processors of pre-cut melon.7,8 The use of sodium hypochlorite or other permitted antimicrobials in combination with brushing is recommended. The potential for microbial contamination also might be reduced by using only good-quality fruit that is free from open wounds or defects that might allow bacteria to contaminate the interior of the fruit.9 Additional research is needed to determine the effectiveness of consumer produce-washing practices. Consumers should be sure that fresh-cut melons are refrigerated or surrounded by ice; leftover cut melons should be discarded if left at room temperature for >2 hours. Additional information for consumers is available at http://www.fda.gov/bbs/topics/answers/2002/ans01167.html.
On October 28, 2002, in response to the three outbreaks during 2000-2002 and analytical results from the sampling of imported Mexican cantaloupe, FDA issued an import alert that detains all cantaloupe from Mexico offered for entry at all U.S. ports. FDA will continue to work with the Mexican government on a food-safety program for the production, packing, and shipping of fresh cantaloupe. The Mexican government is developing a certification program based on sound agricultural and manufacturing practices that would allow FDA to identify farms that have adopted and implemented such a food-safety program.
This report is based in part on assistance and data contributed by J Anderberg, Food Safety Program, Washington State Dept of Health. S Stenzel, K Smith, Minnesota Dept of Health. B Labus, P Rowley, Clark County Health District, Las Vegas, Nevada. S Schoenfeld, Vermont Dept of Health. L Gaul, Texas Dept of Health. S Isaacs, A Ellis, Health Canada, Ottawa; M Fyfe, British Columbia Center for Disease Control, Vancouver; H Bangura, Saskatchewan Health, Regina, Canada. J Varma, J Painter, Div of Bacterial and Mycotic Diseases, National Center for Infectious Diseases, CDC.
Multistate Outbreaks of Salmonella Serotype Poona Infections Associated With Eating Cantaloupe From Mexico—United States and Canada, 2000-2002. JAMA. 2002;288(23):2967-2969. doi:10.1001/jama.288.23.2967-JWR1218-7-1