Search

OBJECTIVE: Human Salmonella infections associated with dry pet food have not been previously reported. We investigated such an outbreak of Salmonella Schwarzengrund and primarily affecting young children. PATIENTS AND METHODS: Two multistate case-control studies were conducted to determine the source and mode of infections among case-patients with the outbreak strain. Study 1 evaluated household exposures to animals and pet foods, and study 2 examined risk factors for transmission among infant case-patients. Environmental investigations were conducted. RESULTS: Seventy-nine case-patients in 21 states were identified; 48% were children aged 2 years or younger. Case-households were significantly more likely than control households to report dog contact (matched odds ratio [mOR]: 3.6) and to have recently purchased manufacturer X brands of dry pet food (mOR: 6.9). Illness among infant case-patients was significantly associated with feeding pets in the kitchen (OR: 4.4). The outbreak strain was isolated from opened bags of dry dog food produced at plant X, fecal specimens from dogs that ate manufacturer X dry dog food, and an environmental sample and unopened bags of dog and cat foods from plant X. More than 23 000 tons of pet foods were recalled. After additional outbreak-linked illnesses were identified during 2008, the company recalled 105 brands of dry pet food and permanently closed plant X. CONCLUSIONS: Dry dog and cat foods manufactured at plant X were linked to human illness for a 3-year period. This outbreak highlights the importance of proper handling and storage of pet foods in the home to prevent human illness, especially among young children.

BACKGROUND: In May 2008, PulseNet detected a multistate outbreak of Salmonella enterica serotype Saintpaul infections. Initial investigations identified an epidemiologic association between illness and consumption of raw tomatoes, yet cases continued. In mid-June, we investigated two clusters of outbreak strain infections in Texas among patrons of Restaurant A and two establishments of Restaurant Chain B to determine the outbreak's source. METHODOLOGY/PRINCIPAL FINDINGS: We conducted independent case-control studies of Restaurant A and B patrons. Patients were matched to well controls by meal date. We conducted restaurant environmental investigations and traced the origin of implicated products. Forty-seven case-patients and 40 controls were enrolled in the Restaurant A study. Thirty case-patients and 31 controls were enrolled in the Restaurant Chain B study. In both studies, illness was independently associated with only one menu item, fresh salsa (Restaurant A: matched odds ratio [mOR], 37; 95% confidence interval [CI], 7.2-386; Restaurant B: mOR, 13; 95% CI 1.3-infinity). The only ingredient in common between the two salsas was raw jalapeño peppers. Cultures of jalapeño peppers collected from an importer that supplied Restaurant Chain B and serrano peppers and irrigation water from a Mexican farm that supplied that importer with jalapeño and serrano peppers grew the outbreak strain. CONCLUSIONS/SIGNIFICANCE: Jalapeño peppers, contaminated before arrival at the restaurants and served in uncooked fresh salsas, were the source of these infections. Our investigations, critical in understanding the broader multistate outbreak, exemplify an effective approach to investigating large foodborne outbreaks. Additional measures are needed to reduce produce contamination.

Background. Infection due to Salmonella species causes an estimated 1.4 million illnesses and 400 deaths annually in the United States. Orange juice is a known vehicle of salmonellosis, for which regulatory controls have recently been implemented. We investigated a nationwide outbreak of Salmonella infection to determine the magnitude of the outbreak and to identify risk factors for infection. Methods. We identified cases through national laboratory-based surveillance. In a case-control study, we defined a case as infection with Salmonella serotype Typhimurium that demonstrated the outbreak pulsed-field gel electrophoresis pattern in a person with illness onset from 1 May through 31 July 2005; control subjects were identified through random digit dialing. Results. We identified 152 cases in 23 states. Detailed information was available for 95 cases. The median age of patients was 23 years; 46 (48%) of the 95 patients were female. For 38 patients and 53 age-group matched control subjects in 5 states, illness was associated with consuming orange juice (90% vs. 43%; odds ratio, 22.2; 95% confidence interval, 3.5-927.5). In a conditional logistic regression model, illness was associated with consuming unpasteurized orange juice from company X (53% vs. 0%; odds ratio, 38.0; 95% confidence interval, 6.5 infinity). The US Food and Drug Administration found that company X was noncompliant with the juice Hazard Analysis and Critical Control Point regulation and isolated Salmonella serotype Saintpaul from company X's orange juice. Conclusions. Unpasteurized orange juice from company X was the vehicle of a widespread outbreak of salmonellosis. Although the route of contamination is unknown, noncompliance with the juice Hazard Analysis and Critical Control Point regulation likely contributed to this outbreak. Pasteurization or other reliable treatment of orange juice could prevent similar outbreaks.