Case report of Salmonella cross-contamination in a food laboratory

Background: This paper describes a case of Salmonella cross-contamination in a food laboratory. In 2012, chocolate bars shipped from Belgium to the USA were prevented from entering the USA because a Salmonella Rissen strain had been isolated from one of the chocolate bars in a Belgian food laboratory. However, a retrospective study of the Salmonella isolates sent from the laboratory to the Belgian National Reference Laboratory for Salmonella revealed that 7 weeks prior, a Salmonella Rissen strain has been isolated from fish meal in the same food laboratory. The chocolate bars were not expected to be contaminated with Salmonella because the ingredients all tested negative during the production process. Furthermore, because Salmonella Rissen is only rarely isolated from food, it was hypothesized that the two Salmonella Rissen isolates belonged to the same strain and that the second isolation event in this laboratory was caused by cross-contamination. To confirm this hypothesis, both Salmonella Rissen isolates were fingerprinted using different molecular techniques. To evaluate the discriminatory power of the techniques used, 11 other Salmonella Rissen isolates from different origins were included in the comparison. Pulsed-field gel electrophoresis, repetitive element palindromic PCR and three random amplified polymorphic DNA PCR assays were used.

Results: Repetitive element palindromic PCR and random amplified polymorphic DNA PCR assays were insufficiently discriminatory, whereas pulsed-field gel electrophoresis using the combination of two restriction enzymes showed sufficient discrimination to confirm the hypothesis.

Conclusions: Although cross-contamination in food laboratories are rarely reported, cross-contamination can always occur. Laboratories should therefore always be aware of the possibility of cross-contamination, especially when enrichment is used in the microbiological analysis. Furthermore, it is advised that results showing isolates of the same serotype isolated in a short time frame from unrelated food products should be interpreted carefully and should be confirmed with additional strain typing.