On 16 August 2007, the Danish regional food authority (Fødevareregion Øst) and the Statens Serum Institut (SSI) became aware of an outbreak of Shigella sonnei infections. The first cases to be reported were employees of two companies. They had eaten a variety of vegetables, including raw baby corn and sugar snaps in their workplace canteens. Preliminary interviews with further cases indicated that the probable source was imported baby corn or sugar snaps that had been distributed at the beginning of August. The suspected foods were distributed by one wholesaler to greengrocers, catering firms, restaurants and shops throughout the country. Due to the strong suspicion about these food vehicles, the Veterinary and Food Administration issued a recall of baby corn and sugar snaps on 17 August. Furthermore, the SSI undertook investigations to determine the extent of the outbreak and its source.
A cohort study was undertaken in one of the many workplaces to have been affected by the outbreak to test the primary hypothesis that infection was associated with eating baby corn in the work canteen. Of 103 web-based questionnaire responses returned by 21 August from people who had eaten in the canteen, 24 reported gastrointestinal symptoms consistent with Shigella infection. In the questionnaire, one set of questions focused on which days the employees had eaten in the canteen and the others on which food items had been eaten in the canteen on the 6 and 7 August, the two days when the suspected baby corn was known to have been served. There was a higher, although non-significant, relative risk of illness among people who had eaten in the canteen on 6, 7 or 8 August. The relative risk for gastrointestinal symptoms among people who had eaten baby corn was 4.6 (95% CI: 2.0 -10.9) on 6 August and 4.0 (95% CI: 1.7-9.6) on 7 August. The attack rate in those employees who ate baby corn on 6 or 7 August was 64%.
Concurrent with the cohort study, notified cases from different parts of Denmark were interviewed and asked about food intake prior to onset of symptoms. The results of these interviews were consistent with baby corn being the source of infection. Moreover, a large number of workplaces were found to be affected by the outbreak and their canteens had served baby corn from the suspected imported batches in their salad bars. Taken together, the available epidemiological and food trace-back evidence strongly supported the finding that baby corn imported from Thailand was the source of the outbreak.
Between 6 and 24 August 2007, the SSI received notifications of 122 S. sonnei isolates. This is almost triple the number of isolates confirmed for the whole of 2006 (46 isolates). A case in this outbreak was defined as any case of S. sonnei infection acquired in Denmark after 1 August 2007 excluding those who had travelled to an endemic area in the three days before onset of symptoms or those that could be explained by an alternative exposure.
To date, 120 cases have met the agreed case definition (2 of the 122 notified ones were excluded as travel-related). Cases were reported through the laboratory surveillance system from the whole of Denmark, but most cases (97/120, 81%) were reported from Zealand. The median age was 38 years (range 1-92 years) and 90 cases (75%) were female. Information on symptom onset dates, which ranged from 6 August until 17 August, was available for 55 cases. A quarter of these cases (13/55) were known to have been admitted to hospital. To date, in-depth interviews have been performed for 35 cases. Of these, all reported diarrhoea, with half (17/35) experiencing bloody diarrhoea and 91% (32/35) reporting stomach cramps.
Antibiotic resistance typing on 11 samples taken from cases has revealed that isolates were resistant to tetracycline, ampicillin, sulfonamides, cephalothin, and streptomycin, but susceptible to nalidixic acid, ciprofloxacin, chloramphenicol, mecillinam, and gentamicin. Typing of isolates by Pulse Field Gel Electrophoresis is ongoing. Microbiological examination of the suspected batches of imported baby corn has detected high levels of Escherichia coli, indicating faecal contamination. Additionally Salmonella (serotypes as yet undetermined) have been found in two batches. Shigella have so far not been detected, but analyses are still ongoing.
An Early Warning Response System report was issued on 18 August. The available information suggests that the outbreak is confined to Denmark but we encourage other countries to be aware of potential clusters of S. sonnei cases. A small number of cases have been reported from Sweden but those contacted so far all appear to have been infected in Denmark. However, it is known that a small part of a batch of baby corn imported into Denmark was sold on to Sweden. This is currently undergoing microbiological examination to determine whether it was contaminated. Furthermore, among the cases in the outbreak there were individuals who acquired the infection on the ferry between Oslo and Copenhagen and it was found that baby corn from an incriminated batch was served on this ferry.

The reported concurrent outbreaks of Shigella sonnei infections in Denmark and Australia have been found to be linked to a common baby corn packing house in Thailand via trace-back of the distribution chain as estimated before. Distribution records indicated that three additional countries received affected product from the implicated Thai packing house during the period of potential contamination. These countries were notified through the World Health Organization's International Food Safety Authorities Network (INFOSAN). Associated cases of S. sonnei have not been reported in these three countries.

Denmark
In Denmark, 218 cases of laboratory-confirmed S. sonnei infection were reported to the Statens Serum Institut, Copenhagen between 1 August and 30 September 2007. During investigations, 12 cases reported recent foreign travel, two had an alternative exposure and three were deemed to be secondary cases and were therefore excluded, leaving 201 primary domestic cases.

Australia
In Australia, a total of 12 laboratory-confirmed cases were reported, all of whom acquired their infection in the state of Queensland. This included two residents of the state of Victoria and one from New Zealand. The onset of illness among all cases was between 9 and 27 August 2007.  

Food analysis
Microbiological examination of the suspected batches of imported baby corn in Denmark detected various serotypes of Salmonella enterica and high levels of Escherichia coli, indicating faecal contamination. Shigella spp. were not detected, suggesting that contamination levels were low. Australia was unable to test the implicated batch of corn, but did detect >100 colony forming units of E. coli per gram in 25 gram samples from other batches from the same importer.

Conclusion
Baby corn was implicated in the Danish outbreak following case reports of baby corn consumption and baby corn being found to be significantly associated with illness in a cohort study. The possibility that the outbreaks might be linked was raised by the Australian public health authorities following the publication of the Danish outbreak strain's antibiotic resistance pattern. This was further reinforced when isolates from Denmark and Australia were found to display indistinguishable PFGE patterns. As well as the report in Eurosurveillance, international awareness of the outbreak and the outbreak strain was raised via numerous international networks, such as the Early Warning Response System (EWRS) and Rapid Alert System for Food and Feed (RASSF) of the European Union, Enternet, Pulsenet Europe, Pulsenet US, Pulsenet Asia, and INFOSAN. This outbreak highlights the importance of timely international communications in helping to identify when a contaminated food enters international trade and results in human illness.

References
http://www.eurosurveillance.org/ew/2007/070830.asp
http://www.eurosurveillance.org/ew/2007/071004.asp

Back