Methods to sample processed broiler carcasses for the presence of Salmonella can differ among countries. Neck-skin excision and whole-carcass rinsing are differing methods used to detect Salmonella prevalence on commercial broiler carcasses before and after chilling.
In the United States, the U.S. Department of Agriculture (USDA) Food Safety and Inspection Service (FSIS) guidance dictates that whole broiler carcasses should be rinsed by whole-carcass rinse with 400 mL of 1 percent buffered peptone water (BPW) and then 30 mL of the resulting rinsate be used for Salmonella analysis.
In the European Union, 25 g samples composed of neck skin from three carcasses are macerated along with 1 percent BPW in a stomacher and then analyzed.
Whole-carcass rinse and neck-skin excision methods studied
In this study, from the sampling step forward, the laboratory methods were standardized for both whole-carcass rinse and neck skin and followed the FSIS standard microbiological laboratory guidelines for Salmonella enrichment, plating and isolation to reduce variables associated with sampling results for Salmonella. This standardized approach was used with the two sampling methods because laboratory procedures can contribute to isolation sensitivity.
The purpose of this study was to evaluate Salmonella prevalence from commercial broiler carcasses obtained pre-chill (before the inside-outside bird washer and immersion chilling) and post-chill (after the chill tank) using the whole-carcass rinse method and evaluation of the corresponding neck skin from the same carcass.
A total of 357 naturally contaminated broiler carcasses were obtained from commercial broiler processing plants. For each carcass, 8.3 g of neck skin was aseptically collected, added to 83 mL of sterile 1 percent BPW and macerated with a stomacher for 30 seconds. The corresponding carcass was placed in a bag with 400 mL of BPW and mechanically shaken for one minute, then 30 mL of the carcass rinsate was removed and 30 additional mLs of BPW was added. After incubation, the FSIS cultural procedure for the isolation and identification of Salmonella was used for the neck skin and whole-carcass rinse methods.
No significant difference observed between methods
The whole-carcass rinse and neck skin procedures produced results that were not significantly different (P>0.05) for Salmonella isolation when the same carcass was sampled by both methods (Table 1). However, there was a significant difference in Salmonella prevalence for carcasses sampled pre- and post-chill, with pre-chill carcasses having a much higher prevalence than post-chill samples. In the U.S., processing plants use immersion chilling, and a variety of antimicrobials may be applied in the chilling process and in the plant. In the EU, air chilling is more common and antimicrobials are currently not in use.
Overall, there was no significant difference observed between the two methods (whole-carcass rinse and neck skin). However, in some cases, when Salmonella was detected on a carcass by one sampling method it was not detected by the other method, indicating that both methods can produce false negative results and by the nature of the methods, this should be obvious.
Both methods produced false negative results
With whole-carcass rinse only 7.5 percent (30 mL) of the rinsate is analyzed and this contributes to false negative results. In a 2008 Australian study (King, et al., 2008) using the same sampling method and the same carcass, Salmonella prevalence ranged from 20 percent to 48 percent depending on the volume of rinsate analyzed. When 30 mL was tested, the detection rate was 20 percent, when 50 mL was analyzed it was 38 percent, and when approximately 400 mL was analyzed it was 48 percent positive detection. In addition to this, firmly attached Salmonella will not be recovered by the whole-carcass rinse.
However, the neck skin method is equally insensitive because only 3 percent to 4 percent of the total carcass skin is used for analysis. The total surface and visceral cavity are not part of the sample.
Differing methods yielded comparable results
Even though some insensitivity does exist, the fact that results with these two methods are similar allows one to compare Salmonella prevalence on broiler carcass studies conducted in the U.S. and the EU and other parts of the world using either method.
N. A. Cox, U.S.D.A,, A.R.S., Russell Research Center, Athens, Georgia, 30605; J. A. Cason, retired; M. E. Berrang, U.S.D.A,, A.R.S., Russell Research Center; C. Hofacre, University of Georgia; D. E. Cosby, U.S.D.A,, A.R.S., Russell Research Center; R. Biggs, Tegel Foods, New Zealand; S. Wagener, Ministry for Primary Industries, New Zealand; A. M. Lammerding, AML Consulting, Canada; and M. P. Doyle, Center for Food Safety and Department of Food Science, University of Georgia, Griffin, GA 30223
