Multiplex polymerase chain reaction and testing of complex food matrices: A preliminary study on E. coli and Salmonella spp.

Foodborne diseases are caused by various pathogens and generally present with similar symptoms, mainly digestive disorders. Adopting a syndromic approach is therefore important when investigating foodborne disease outbreaks. This involves using multiplex PCR-based methods to test stool and food samples. In this study, the BioFire® FilmArray® Gastrointestinal Panel (BF-GIP) was evaluated as an alternative food microbiology method, in compliance with the ISO 16140-2 standard. We tested 205 naturally and artificially-contaminated mixed salad samples for Escherichia (E.) coli and Salmonella spp., comparing the results with BF-GIP and culture-based reference methods. For Salmonella spp., BF-GIP sensitivity was 100 % and the rate of false positives was 2.4 %, suggesting a good specificity. For E. coli, BF-GIP sensitivity was 92.9 %, with no significant difference from the reference method sensitivity. Shortcomings were highlighted for the accurate identification of E. coli strains artificially inoculated in food samples, with error percentages of 14 % and 9 % for the inclusivity and exclusivity studies respectively. BF-GIP is a suitable alternative method for the screening of Salmonella spp. and E. coli in complex food matrices, notably when investigating foodborne disease outbreaks. However, due to a lack of specificity, the determination of the E. coli strain(s) needs to be confirmed by a reference method.