Simulation model for enumeration of Salmonella on chicken as a function of PCR detection time score and sample size: Implications for risk assessment

Abstract

A data gap commonly identified in risk assessments is the lack of quantitative information on the contamination of food with pathogens. A simulation model that predicts the incidence and distribution of Salmonella contamination on chicken as a function of PCR detection time score and sample size was developed with data from challenge studies with preenrichment samples that were composed of 25 g of chicken and 225 ml of buffered peptone water inoculated with 100.7 to 106 Salmonella and incubated at 37°C. At 0, 2, 4, 6, 8, 10, 12, and 24 h of incubation, subsamples were collected and tested for Salmonella by PCR, and a PCR detection time score based on the widths of the bands in the electrophoresis gel was obtained for each preenrichment sample. Standard curves relating PCR detection time score to initial density of Salmonella inoculated were developed for sterile and nonsterile preenrichment samples. Presence of other microorganisms in the preenrichment sample decreased the PCR detection time score at low (<10 2 per 25 g) but not at high (>102 per 25 g) initial densities of Salmonella and resulted in a nonlinear standard curve rather than the linear standard curve obtained for sterile samples. The predicted incidence and distribution of Salmonella contamination on chicken increased in a nonlinear manner as sample size increased from 25 to 500 g. The new method reduced the time and cost of Salmonella enumeration by eliminating the selective enrichment, selective plating, and confirmation steps of the traditional most-probable-number method. Results are useful for risk assessment because they consider the uncertainty of the standard curve predictions and because they provide distributions of Salmonella contamination for different size samples of chicken that can be directly used in risk assessment.