Usefulness of indicator bacteria as potential marker of Campylobacter contamination in broiler carcasses.

According to Regulation (EC) No 2017/1495, amending Regulation (EC) No 2073/2005, a slaughterhouse process hygiene criterion based on a limit of 103 CFU/g of Campylobacter for no more than 20 (for the years 2018 and 2019), 15 (for the years 2020-2024) and 10 (starting from 2025) of 50 neck skin samples of broiler carcasses could be an effective measure to reduce the incidence of human campylobacteriosis. In order to stimulate the poultry industry to improve the control of Campylobacter along the slaughter-line, the quantification of indicator bacteria such as Escherichia coli or Enterobacteriaceae could be a useful strategy. The aims of this study were: a) to investigate the possible relationship between Campylobacter and indicator bacteria counts at two different points of the broiler slaughter-line and b) to evaluate the probability that carcasses have Campylobacter counts above 103 CFU/g in relation to indicator bacteria counts. E. coli, Enterobacteriaceae and Campylobacter were simultaneously enumerated on neck skin samples of broiler carcasses sampled at the post-evisceration (n = 75) and at the post-chilling points (n = 75) of three Italian poultry slaughterhouses. In general, the log counts of all the investigated microorganisms were significantly lower at the post-chilling point, and the indicator bacteria (E. coli and Enterobacteriaceae) counts were significantly higher than Campylobacter counts at both sampling points. A multilevel linear mixed model, relating the Campylobacter log10 counts with the E. coli and Enterobacteriaceae log10 counts, showed that the Campylobacter log10 counts increased significantly for every additional E. coli log10 count, and this was more evident at the post-chilling than at the post-evisceration sampling point. With regards to the Enterobacteriaceae, the increase was similar at the two sampling points. An additional model, developed to assess the probability that carcasses would have Campylobacter counts above 3 log10 CFU/g, showed that this probability increased significantly if the level of E. coli count also increased. Carcasses classified as Campylobacter > 3 log10 CFU/g according to the observed results of E. coli ≥ 4 log10 CFU/g did have high Campylobacter counts. Conversely, it was not possible to conclude anything for carcasses with E. coli < 4 log10 CFU/g. These findings support the hypothesis that the monitoring of poultry carcasses for E. coli load could be useful to identify those heavily contaminated with Campylobacter, in order to implement control measures on the farms of origin of such batches, or improving the slaughter process of the plants where the heavily contaminated batches are found.