Abstract

Salmonella-contaminated poultry house dust plus 10 g chicken faeces inoculated with Salmonella Enteritidis and then frozen for storage and transport were used as candidate external quality assurance test samples. Variations in faeces sample preparation, storage and culture were examined initially. This indicated that, within modest limits, the age of the inoculating culture and of the faeces did not affect detection, nor did swirling the pre-enrichment culture or extending its duration. Under optimal conditions of preparation and storage, Salmonella numbers of 70 colony-forming units (CFU) and above were reliably detected at the originating laboratory. A ring trial was performed, involving 13 external UK laboratories plus the originating laboratory. Faeces samples inoculated with Salmonella Enteritidis were frozen, transported on dry ice and tested by the ISO 6579:2002 (Annex D) method. Detection by the originating laboratory was consistent with the previously established lower limit for reliability of 70 CFU. However, the sensitivity of detection by the external laboratories was apparently poorer in several cases, with significant interlaboratory variation seen at the lowest inoculum level, using Fisher's exact test. Detection of Salmonella in poultry house dust appeared to be more sensitive and uniform among laboratories. Significance and impact of the study: Salmonella surveillance and control regimes in the European poultry industry and elsewhere require sensitive culture detection of Salmonella in environmental samples, including poultry faeces. A ring trial was conducted, and the results highlighted that some of the participating laboratories failed to identify Salmonella. This suggests that contaminated frozen faeces cubes could be beneficial to assess proficiency, according to the results of this preliminary study. The data obtained in this study can be used as an indication for the design of realistic external quality assurance for laboratories involved in official testing of Salmonella in poultry flocks.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call