Growth of Salmonella Typhimurium DT104 at 30°C is not affected by anatomical location on the chicken carcass
Posted: December 31st, 2011 - 10:47am
Source: Journal of Food Protection®, Volume 75, Number 1, January 2012 , pp. 164-168(5)
Development of models for growth of Salmonella in the chicken food matrix is time-consuming and expensive. The current study was undertaken to examine growth of Salmonella on different anatomical locations of the chicken carcass. The purpose was to determine whether anatomical location should be included as an independent variable in predictive models for chicken. Eleven anatomical locations were studied: skin (wing, breast, drumstick, and thigh), meat surface (wing, breast, drumstick, and thigh), and meat interior (breast, drumstick, and thigh). Background microflora, pH, and growth (lag time, λ; growth rate, μ; and time for a 3-log increase, t3) at 30°C for a small inoculum size (0.92 ± 0.30 log per portion) of Salmonella Typhimurium DT104 were examined. Four or six replicate storage trials were conducted per anatomical location (n = 46 growth curves). Portion sizes were 1.12 ± 0.17 g (mean ± standard deviation) for meat and 0.25 ± 0.08 g for skin. A two-phase linear model was used to determine λ and μ. The effect of anatomical location on dependent variables was assessed by one-way analysis of variance. pH values differed (P < 0.001) among anatomical locations, with skin (6.86 ± 0.20). dark meat (6.39 ± 0.20) . white meat (5.97 ± 0.20). Background microflora (4.32 ± 1.66 log per portion) was variable and not affected (P > 0.05) by anatomical location. Likewise, λ (1.90 ± 0.75 h), μ (0.648 ± 0.120 log/h), and t3 (6.71 ± 0.82 h) at 30°C were not affected (P > 0.05) by anatomical location. Although there were differences in pH among anatomical locations, these differences were not sufficient to affect growth of Salmonella Typhimurium DT104 at 30°C. If this observation holds for other storage conditions and strains, then anatomical location does not need to be included as an independent variable in predictive models for chicken. This would save significant time and money for the predictive microbiologist.