The aim of this study was to examine the relationship between the load of Salmonella spp. in caeca and the carcass contamination in an Italian slaughterhouse. The sampling scheme was designed to be representative of the pigs slaughtered in a day and to estimate a 12% prevalence of pigs highly contaminated by Salmonella spp. (HCP, cecal load ≥3log). Environmental swabs were taken before slaughter. Cecal contents and carcass swabs were collected from the same pig. Salmonella MPN were estimated according to ISO6579- 2:2012/A1 and ISO7218:2007/E. The overall Salmonella prevalence were 34.64% and 7.19% for ceca and carcasses respectively, with S. Derby and S. 4,[5],12:i:- being the prevalent serotypes. The HCP prevalence was 11.44%. 7/59 environmental swabs tested positive; when the same serotype was isolated from the environment and from carcasses, the samples were excluded from further analysis. Statistical analysis was performed to investigate the relationship between Salmonella spp. loads in the cecum and contamination of the carcass of the same pig and the prevalence of HCP and the contamination of carcasses on the same day. For this purpose, the days were classified as “high prevalence days” depending on the proportion of caeca resulted positive (≥36%) and as “high load” days depending on the prevalence of HCP (≥10%). A correlation between the contamination of carcasses and the cecal Salmonella loads of the same animal was found (Spearman’s correlation coefficient: 0.2254; p-value=0.0001). No correlation was found between the contamination of carcasses and the categorization of the day of sampling as “high prevalence day”. Conversely, a correlation was found between the contamination of carcasses and the “high load” category of the sampling day (Wilcoxon test, p=0.0011). Notably, not the prevalence of pigs carrying Salmonella spp. but the prevalence of highly contaminated pigs was shown to be related to the contamination of carcasses.