Background

Salmonella is a significant concern to the beef industry as ground beef products have been implicated in a number of outbreaks. Various interventions have been implemented and proven effective to remove Salmonella on carcass surfaces. Despite the success of carcass surface interventions, however, FSIS testing results indicate little reduction of Salmonella contamination in ground beef during the past decade. These findings imply other potential sources of ground beef contamination, such as the lymph nodes. Recent studies indicate that lymph nodes (LN) may be an important source of Salmonella contamination in ground beef.

Evaluating the relative contribution of Salmonella contamination from beef carcass surface or lymph nodes through risk analysis is beneficial in prioritizing control measures targeted at Salmonella contamination based on the currently available data. We developed a stochastic simulation model covering the pre‐ to post‐harvest stages to assess the relative contribution of lymph node as compared with carcass surface, to Salmonella in ground beef, and the impact of various pre‐ and post‐harvest interventions. We also performed scenario analysis to evaluate different levels of effectiveness for pre‐and post‐harvest interventions impacting the Salmonella load in ground beef.

The objectives of this study were to: 1) Develop a stochastic simulation model to assess the relative contribution of DTLNs to Salmonella contamination of ground beef products and 2) Exercise the model to evaluate the relative impact of various pre‐ and post‐harvest interventions on the reduction of Salmonella contamination in ground beef products.

Methodology

The risk assessment model covers the Salmonella prevalence and concentration at various production and processing points from on‐farm/feedlot (pre‐harvest) to in‐plant (post‐ harvest) stages, with an emphasis on the pre‐harvest stage. The production stage estimates the impact of pre‐harvest interventions on reduction of Salmonella contamination in lymph nodes. The in‐plant stage estimates the prevalence and concentration of Salmonella on cattle hides, carcasses and in ground beef at various processing points and in ground beef.

Model scenarios for ground beef production from fed and cull cattle in high as well as in low Salmonella prevalence seasons were considered. The following outcomes were evaluated in the baseline model: (i) relative contribution of lymph node contamination to the Salmonella CFU load in a 2,000 lb ground beef production lot; (ii) total Salmonella CFU load in a 2,000 lb production lot. In addition, multiple scenario analyses with respect to key parameters were conducted to estimate how changes in their values impact model outcomes.

Linear relationships – based on USDA/ARS studies and published data – were used to relate the changes of Salmonella prevalence and concentration at various processing stages. The decrease in Salmonella concentration due to post‐harvest interventions was estimated from indicator bacteria data in commercial plants and from Salmonella reduction data in lab studies. To model lymph node contamination, statistical distributions for Salmonella prevalence and concentration, and major lymph node weights were estimated using available data and expert opinion. The spread sheet-based simulation model was implemented using Microsoft Excel and @RISK software.

Findings

The baseline model results (Table 1) indicate that on average over 90% of the Salmonella CFU load in a 2,000 lb ground beef production lot originates from lymph node contamination, as compared with carcass surface contamination, for each category of animal type and season combinations. The predicted mean Salmonella CFU load from lymph node contamination was considerably greater in ground beef production lots from fed cattle compared to cull cattle.

Within each animal type, the Salmonella load in a production lot was greater in the high season compared to the low season. Scenario analysis indicates that generic pre‐harvest interventions that can reduce Salmonella prevalence and concentrations in lymph nodes could also result in a significant reduction in the total Salmonella CFU load in a ground beef production lot for fed cattle. However, owing to the lower Salmonella CFUs from lymph nodes of cull cattle, the generic pre‐ harvest interventions were predicted to cause a smaller reduction in the total Salmonella CFU load in this case.

Implications

The model results suggest that most Salmonella load in a ground beef production lot originates from lymph nodes. A rationale for these results is that the carcass surface interventions are effective and very few numbers of Salmonella are predicted to be added to a ground beef production lot from the carcass surface. These results are consistent with previous USDA/ARS studies (Barkocy‐Gallagher et al., 2003; Brichta‐Harhay et al., 2008), where few post‐ intervention carcasses were enrichment positive for Salmonella and an even smaller proportion of those enrichment positive samples were enumerable at low concentrations.

Sensitivity analyses indicated that for fed cattle, the relative contribution of lymph node contamination is fairly robust to changes in model parameters such as the effectiveness of post‐ harvest interventions. There is a greater uncertainty regarding the relative contribution of lymph node contamination for cull cattle and the results were more sensitive to changes in model parameters in this case.

The model results also indicate that ground beef from cull cattle would have a lower Salmonella CFU load relative to fed cattle. This is because, compared with fed cattle, cull cattle contributed a lower number of carcasses per production lot and had a lower Salmonella lymph node prevalence, resulting in fewer contaminated lymph nodes being included in a production lot. Besides, the CFU load per contaminated lymph node was also lower in cull cattle.

Several categories of pre‐harvest interventions, such as vaccines, are being evaluated for reducing Salmonella in lymph nodes of beef cattle (Edrington et al., 2013). Given the limited data on the impact of pre‐harvest interventions, we evaluated the impact of generic pre‐harvest interventions (at different efficacy levels), which could reduce the prevalence, and/or concentration of Salmonella in lymph nodes on the model outcomes. Our scenario analysis indicates that generic pre‐harvest interventions that can reduce Salmonella prevalence and concentrations in lymph nodes could also result in a significant reduction in the total Salmonella CFU load in a ground beef production lot for fed cattle. This is not unexpected since most of the Salmonella CFUs of fed cattle could be attributed to lymph node contamination. However, owing to the lower Salmonella CFUs from lymph nodes of cull cattle, the generic pre‐harvest interventions were predicted to cause a smaller reduction in the total Salmonella CFU load in this case.

There may be potential alternate explanations for the current model results, which have not been considered in the scope of the current analysis. Some of the alternate explanations include cross‐contamination from plant surfaces during fabrication, presence of Salmonella beneath the carcass outer surface due to cuts or bruises, lack of plant data for the effectives of post‐harvest intervention on Salmonella reduction, and internal carcass contamination caused by gut rupture. Evaluation of these potential alternate explanations in the future would improve the robustness of the conclusions.