Salmonella Control in Poultry

Direct plating typically occurs by preparing a homogenate of the test sample and using a portion to inoculate agar plates selective for a given pathogen. Background organisms that can grow on the same agar can confound the process, making it difficult to isolate the target pathogen. Also, injured or stressed cells may not grow at all. Thus, the reliability of this method varies based on the type of sample, state of the target pathogen, and presence of other microbial flora.

Both of these enumeration methods are time-consuming because of culture growth requirements, confounding factors, and requirement for hands-on technical interpretation. Food processing needs rapid, cost-effective methods that provide quantitative information on the contamination level to better support process control and pathogen management.

Evaluating a Quantitative Solution

Based on the association of Salmonella with raw poultry products and the limitations of conventional quantitative methods for Salmonella evaluation, processors face a big challenge in understanding overall process control and potential for high contamination levels. Despite limitations in the science for direct enumeration of low pathogen levels, a threshold determination based on little or no enrichment is possible with existing detection technology.

Vanguard Sciences (formerly ­AEGIS Food Testing Laboratories, Inc.) provides technical consultation and microbiological testing services. Vanguard Sciences in collaboration with Bio-Rad Laboratories and a poultry processor evaluated the feasibility of Bio-Rad’s iQ-Check Salmonella II standard real-time polymerase chain reaction (PCR) method for direct detection of Salmonella at a threshold of 100 colony-forming units (CFU)/milliliter (2.00 logCFU/ml). Feasibility and method (matrix) validation trials were performed using poultry carcass rinsate samples prepared as part of a routine Salmonella sampling and testing program in a commercial poultry processing facility.

For the feasibility trials, 10 carcass rinsate samples that tested Salmonella negative were shipped via overnight courier to Vanguard Sciences under refrigeration for inoculation within 24 hours of receipt. Samples were analyzed for background flora based on aerobic plate count (APC) by direct plating on 3M PetriFilm following AOAC method 990.12. In addition, Salmonella-negative status was verified with the iQ-Check direct PCR standard method prior to inoculation. A poultry isolate of Salmonella was used to inoculate 10 prepared poultry carcass rinsates and samples were serially diluted to above and below 100 CFU/ml (6 concentrations per rinsate; n=60). They were also directly analyzed without enrichment for the presence of Salmonella using the iQ-Check Salmonella II standard real-time PCR method.

After trials one and two, optimizations to the assay were made to improve accuracy and sensitivity, including adjusting the sample size and volume of lysate buffer. All inoculated rinsates were simultaneously direct plated on Bio-Rad’s RAPID’Salmonella chromogenic agar at the time of testing to verify the presence of Salmonella at the target threshold level. Of the 60 rinsates analyzed during trial three using the optimized protocol, a total of 20 had Salmonella present on the plates, indicating levels of 20 to 1,000 CFU/ml (1.30 to 3.00 logCFU/ml). Of those samples, 18 were positive for Salmonella using the iQ-Check Salmonella II standard real-time PCR assay, ranging between 30 to 1,000 CFU/ml (1.48 to 3.00 logCFU/ml). Two samples at 80 and 20 CFU/ml (1.9 and 1.3 logCFU/ml) were PCR negative. A total of 10 rinsates were not inoculated and served as negative controls, and all were negative by direct plating and PCR. Background flora (APC) ranged from 10 to 8,600 CFU/ml. These data indicate the feasibility of this approach as a direct detection method for Salmonella at a specified threshold level.

Upon feasibility determination, a matrix validation was performed whereby representative bone-in, skin-on thighs (n=20) were inoculated with Salmonella and subjected to standard rinsate procedures to allow for recovery of approximately 100 CFU/ml (2.00 logCFU/ml) in the rinsate. These rinsates were then subjected to analysis using the optimized protocol for the iQ-Check Salmonella II real-time PCR assay in conjunction with a reference method following USDA-FSIS procedures published in the Microbiology Laboratory Guidebook. In addition, samples were direct plated on RAPID’Salmonella chromogenic media. A total of five samples were processed as negative controls. All samples were also processed for indigenous Salmonella and background APC.