Electrical stunning, hot boning, and quality of chicken breast meat.

The first experiment was conducted to determine the effects of varying voltage, 20, 40, 80, and 100 V at 60 Hz, on stunning efficiency, blood loss, and carcass defects. In the second experiment, the same parameters were evaluated to determine the effects of varying frequency, 60, 200, 350, 500, and 1,000 Hz at 40 V. A control group for both experiments was not stunned. At 40V, 30 to 50 mA, 90% of the birds were unconscious, as shown by no response to comb piercing, and blood loss was maximized (55.3%). When varying the stunning frequency, maximum blood loss (73.1%), 90% of the birds were unconscious, and minimum carcass defects were observed at 1,000 Hz, 40 V. In the third experiment, birds were stunned at 40 V, 1,000 Hz and deboned immediately after defeathering (hot boning) and chilled or deboned after passing through all stages of a commercial abattoir operation (conventional boning). Control lots were unstunned and followed normal abattoir stages. Average shear value was significantly lower for stunned compared to unstunned birds (6.0 vs. 7.1 kg/g), although tenderness scores, as measured by a trained panel, were not significantly different (6.6 for stunned birds vs. 6.1 for unstunned). Scores for juiciness were also not significantly different (5.5 for stunned vs. 5.8 for unstunned). Average shear value was also significantly lower for conventionally boned birds (5.2 kg/g) than for hot boned birds (7.9 kg/g). Sensory analysis confirmed the shear value results. Conventionally boned breasts had an average tenderness score of 7.4 vs. an average of 5.3 for hot boned breast. No statistical differences were observed with respect to juiciness, although a score of 6.2 was observed for conventionally boned breast meat vs. a score of 5.1 for hot boned breast meat.

Occurrence of Escherichia coli O157:H7 in hamburgers produced in Brazil.

The occurrence of Escherichia coli O157:H7 was evaluated in 886 samples of hamburgers produced by eight manufacturers in the south and southeast of Brazil from January to September of 1997. The pathogen was not detected in any of the samples analyzed, although 17 (1.9%) showed the presence of E. coli strains, which presented agglutination with O157 antiserum. These strains were not confirmed, since a subsequent identification indicated that they were fermentation positive for sorbitol, incapable of agglutination with the antisera of other commercial brands, and nonproducers of verocytotoxins. A high rate of false-presumptive-positive tests was observed with strains of enterobacteria when isolated by the culture method and when using a commercially purchased enzymatic immunoassay. This high rate of false presumptive positives was likely because the initial screening for sorbitol-negative strains was carried out using phenol red sorbitol 4-methylumbelliferyl-beta-D-glucuronide agar, which was shown to be inadequate for this purpose.