Validation according to ISO 16140:2003 of a commercial real-time PCR-based method for detecting Campylobacter jejuni, C. coli, and C. lari in foods.

Campylobacteriosis was the most frequently reported zoonosis in the European Union (EU) in 2010, with Campylobacter jejuni, Campylobacter coli, and Campylobacter lari as the most frequently reported species in foodborne outbreaks (FBOs). Relatively sensitive to environmental factors, these species may be present in low numbers. In line with EU policy for food control and FBO detection and in view of the need to reduce response time, we validated an alternative molecular method according to ISO 16140:2003 which establishes the general principle and technical protocol for the validation of alternative methods in the microbiological analysis of food. We used a qualitative real-time PCR commercial kit for the detection of C. jejuni, C. coli, and C. lari in two food categories "fruit and vegetable-based products" and "dairy products". The validation protocol comprises two phases: the first is a method comparison study of the alternative method against the reference method, and the second is an interlaboratory study of each of the two methods. In the first step, ISO 16140:2003 validation examines the following parameters: limit of detection (LOD); relative accuracy, relative specificity and sensitivity; relative detection level (RDL); and inclusivity and exclusivity. Except for LOD, inclusivity and exclusivity, the other steps were performed against the reference method (ISO 10272:2006). The LOD of the real-time PCR method was set at 4CFU/25g or mL for both food categories. Relative accuracy (98.33%), specificity (96.77%), and sensitivity (100%) were recorded for the food category "fruit and vegetable-based products" and 93.3%, 88.24%, 100%, respectively, for "dairy products". The RDL according to Fisher's exact test was p=1 for both food categories, for each level, and each food/strain combination. The interlaboratory study results showed correct identification of all 24 blind samples with both methods by all the participating laboratories. The results show that this commercial kit is suitable for the rapid detection of C. jejuni, C. coli, and C. lari on fruit, vegetables and dairy products and may aid in official controls. In conclusion, the use of alternative methods is recommended for the rapid identification of positive samples and the identification of the possible bacterial source in a FBO within 48 h.