Campylobacter jejuni contamination of broiler carcasses: Population dynamics and genetic profiles at slaughterhouse level

•Defeathering and evisceration procedures identified as critical contamination points.•Cross-contamination is possible, but does not seem to be important in carcass contamination.•Water cleansing, cooling and/or freezing reduce Campylobacter carcass contamination.•Escalating level of Campylobacter contamination detected in successive slaughter batches.

Six slaughter batches deriving from six typical industrial broiler flocks were examined for the presence, quantity and genetic characteristics of contaminating Campylobacter jejuni (C. jejuni) during various stages of slaughtering and carcass processing. To assess the contamination dynamics of the carcasses, the analyses were always conducted on neck-skin samples from the same pre-selected and carefully marked carcasses in each batch. The skin samples were taken sequentially at three successive slaughter-line locations in the evisceration room, after three-day refrigeration and after three-day freezing procedure. Caecal samples from the same animals were also tested, as well as samples from the slaughterhouse environment before and after slaughtering. The samples were analysed by the ISO10272 isolation method; campylobacters from neck-skin samples were also quantified. Isolates were species-identified and genotyped by pulsed-field gel electrophoresis (PFGE). On average, the highest C. jejuni skin contamination was detected at the first sampling point (post-plucking), suggesting that the majority of Campylobacter contamination actually occurs before the entrance to the eviscerating room, probably during the preceding plucking stage. In two out of five positive batches, an additional increase in contamination was recorded after the evisceration step. An evident trend of increasing contamination level was detected when successive batches were compared at each of two initial sampling sites in the evisceration room, indicating an accumulation of contaminating C. jejuni at some point before the evisceration room. Three-day refrigeration and three-day freezing caused a 4.5- and 142-fold drop in mean C. jejuni CFU counts, respectively. All pre-slaughtering samples from the slaughterhouse environment were negative and all post-slaughtering samples, except water from the scalding tank, were positive. Pulsotypes were limited: altogether five different types were detected, typically one type per batch. The PFGE results from the slaughterhouse environment isolates indicate that cross-contamination is possible (multiple pulsotypes detected in e.g. eviscerating machine). Nevertheless, this was not confirmed in carcasses: analyses of neck-skin isolates suggest that carcasses are contaminated by their own caecal/farm/flock pulsotype.