Outbreak of Escherichia coli O157:H7 infections associated with nonintact blade-tenderized frozen steaks sold by door-to-door vendors.

Steaks have not been recognized as an important vehicle of Escherichia coli O157:H7 infection. During 11 to 27 June 2003, the Minnesota Department of Health (MDH) identified four O157 infection cases with the same pulsed-field gel electrophoresis (PFGE) subtype. All four case patients consumed brand A vacuum packed frozen steaks sold by door-to-door vendors. The steaks were blade tenderized and injected with marinade (i.e., nonintact). Information from single case patients in Michigan and Kansas identified through PulseNet confirmed the outbreak. The MDH issued a press release on 27 June to warn consumers, prompting a nationwide recall of 739,000 lb (335,506 kg) of frozen beef products. The outbreak resulted in six culture-confirmed cases (including one with hemolytic uremic syndrome) and two probable cases in Minnesota and single confirmed cases in four other states. The outbreak PFGE subtype of O157 was isolated from unopened brand A bacon-wrapped fillets from five affected Minnesota households. A fillet from one affected household was partially cooked in the laboratory, and the same O157 subtype was isolated from the uncooked interior. The tenderizing and injection processes likely transferred O157 from the surface to the interior of the steaks. These processing methods create new challenges for prevention of O157 infection. Food regulatory officials should reevaluate safety issues presented by nonintact steak products, such as microbiologic hazards of processing methods, possible labeling to distinguish intact from nonintact steaks, and education of the public and commercial food establishments on the increased risk associated with undercooked nonintact steaks. Information on single cases of O157 infection in individual states identified through PulseNet can be critical in solving multistate outbreaks in a timely manner.

Occurrence of Escherichia coli, noroviruses, and F-specific coliphages in fresh market-ready produce.

Forty samples of fresh produce collected from retail food establishments were examined to determine the occurrence of Escherichia coli, F-specific coliphages, and noroviruses. An additional six samples were collected from a restaurant undergoing investigation for a norovirus outbreak. Nineteen (48%) of the retail samples and all outbreak samples were preprocessed (cut, shredded, chopped, or peeled) at or before the point of purchase. Reverse transcription-PCR, with the use of primers JV 12 and JV 13, failed to detect norovirus RNA in any of the samples. All six outbreak samples and 13 (33%) retail samples were positive for F-specific coliphages (odds ratio undefined, P = 0.003). Processed retail samples appeared more likely to contain F-specific coliphages than unprocessed samples (odds ratio 3.8; 95% confidence interval 0.8 to 20.0). Only two (5.0%) retail samples were positive for E. coli; outbreak samples were not tested for E. coli. The results of this preliminary survey suggest that F-specific coliphages could be useful conservative indicators of fecal contamination of produce and its associated virological risks. Large-scale surveys should be conducted to confirm these findings.