Impact of sample size on the Limit of Detection LOD<sub>50</sub> to detect Campylobacter in foods

Jacobs-Reitsma, W.F.; W.M., Van Overbeek; Hazeleger, W.C.; Jongenburger, I.


The European Commission Mandate M/381 asked the European Committee for Standardization (CEN) to provide full validation of the revised versions of ISO 10272-1/-2 by inter-laboratory studies and to establish the performance characteristics of these ISO methods for detection and enumeration of Campylobacter in foods. After fi nalising the study on Campylobacter detection, the elaborated data were inserted in ISO 10272- 1. Results are expressed as the limit of detection at 50% (LOD50): the concentration in CFU/sample for which the probability of detection is 50%. No legislative requirements within the EU exist for testing on presence of
Campylobacter in a specifi c portion size, in contrast to e.g. the requirement for absence of Salmonella in 25 gram (EC 2073/2005). Therefore, for practical and economic reasons, the size of the test portions in the validation study on detection of Campylobacter was decided to be 10 gram. As test portions of 25 gram are also commonly used in detection methods, our laboratory compared the LOD50 for Campylobacter detection in 10 gram and 25 gram
test portions of foods. Matrices as used in the inter-laboratory study were tested, using both samples sizes with identical inoculum at 2 diff erent levels (and a blank as control) and the LOD50 was calculated for each sample size. Preliminary results indicate better Campylobacter detection in 10 gram portions of minced meat and frozen spinach, in contrast to better detection in 25 gram portions of raw milk. Experiments are ongoing to elucidate
the possible eff ect of sample size on the LOD50.