Validation of test portion pooling for Salmonella spp. detection in foods

- Paper validates the pooling of 375 g test portions for the detection of Salmonella.
- No differences were found for the 5 methods tested regardless of the technology.
- Maximum incubation time for pre-enrichment showed a better performance.
- Differences in pH after pre-enrichment did not affect the detection of Salmonella.

Pathogen monitoring programs play a crucial role in the verification of the effectiveness of implemented hygiene control measures. Sampling and testing procedures included in pathogen monitoring involve the analysis of multiple test portions where all samples must be negative for the presence of pathogens for a certain test portion size. Many food safety programs require increased testing due to the risks that a pathogen may be present. Analyzing more than one test portion could prove to be expensive and labor intensive. When more than one test portion for a specified food item is to be tested, the test portions could be combined to form a pooled test portion to reduce laboratory workload, costs of reagents and further confirmatory steps, but only when evidence is available that pooling does not affect on the number of false negative results for different matrices. This study has been performed to demonstrate the equivalence of test portion pooling for Salmonella detection with five different methods using cultural, ELISA and Real Time PCR technologies. Twenty-three (23) different food items including confectionary products, meal components, infant formula, pet food and powdered beverages were validated. Other complementary parameters like impact of minimum and maximum incubation time for pre-enrichment, temperature profile, pH and Salmonella concentration after the pre-enrichment and background flora have also been considered in the study. The results showed that pooling test portions up to 375 g for Salmonella detection is valid for the methods that were tested. Relative level of detection (RLOD50) values for 22 of the food items tested were acceptable (i.e. lower than 2.5) when comparing the reference sample size (25 g) against the alternative pooled sample size (375 g), provided the enrichment broth was pre-warmed and maximum incubation time is respected.