ABSTRACT

Four carcasses from two separate flocks processed each day (the third flock on the first shift and the third flock on the second shift) were removed from the line just after bleed out using the following technique to ensure that no bias was introduced: ’ese flocks were chosen to prevent selection of carcasses from the very beginning of a shift when process waters are clean and do not represent typical processing situations. A carcass was selected visually on the line after bleed out, then the next five carcasses were counted aloud, and the sixth carcass was selected for testing. ’e individual selecting the carcasses was wearing sterile examination gloves. In this way, no visual cues were used to introduce bias. ’e four carcasses were plugged using unscented tampons, and the necks were tied off using zip ties. ’is was done so that the variable of fecal or ingesta leakage was avoided. It is impossible to predict how much fecal material or ingesta will leak from bird to bird during rinsing. ’erefore, it is best to avoid this variable by making sure no leakage occurs. ’is has been detailed in a research study by Musgrove et al. (1997). ’e carcasses were then individually bagged in sterile polyethylene bags and rinsed using 400 mL of sterile buffered peptone solution by conducting the whole-carcass rinse method as employed by the USDA inspectors in processing facilities. ’e rinsate was encoded using a three-digit number (to prevent identification by laboratory employees and the introduction of bias) and sent to the laboratory for evaluation for APC, E. coli counts, and Salmonella prevalence. ’e microbiological methods used are detailed in the microbiological methods section of this report. A total of 96 carcasses were evaluated over 12 days and 24 flocks. ’is allowed the plant to determine the level of APC, E. coli, and Salmonella coming in on the live birds. ’e number of carcasses tested was based on statistical models to ensure a representative number

over the entire grow-out process. Incoming bacterial counts and prevalence were evaluated so that a comparison to fully processed, ready-to-cook carcasses could be made.