Feeding management practices and feed characteristics associated with Salmonella prevalence in live and slaughtered market-weight finisher swine: a systematic review and summation of evidence from 1950 to 2005.

This review summarizes evidence for associations between Salmonella prevalence in market-weight swine and changes in feeding management practices or feed characteristics. A systematic review of the topic was conducted with the goal of minimizing the impact of bias on the review conclusions. Potential interventions included feed withdrawal from swine prior to slaughter, feed acidification, heat treatment of feed, pelletized feed versus mash, course versus fine grind, and wet versus dry feeds. In the reviewed literature, Salmonella prevalence was measured either by culture or by the presence of antibodies to Salmonella. The evidentiary value of studies was assessed, and studies that failed to meet predetermined standards were excluded. 7694 potentially relevant references were identified by an extensive literature search; however, 2623 references that were not published in English were excluded, because funds for translation were not available. Of the remaining references, only 277 were considered relevant to the review topic by two independent reviewers, and assessed for methodological quality. During quality assessment, 233 references were excluded because they failed to report design features that limit the introduction of bias or were conducted in a non-target population such as gnotobiotic, neonatal, nursery, or recently weaned pigs and sows. Forty-four publications passed the quality assessment conducted by 2 independent reviewers, but only 15 of the 44 publications reported studies that tested hypotheses associated with feeding management practices and feed characteristics and Salmonella prevalence in market-weight swine. The most common study design was cross-sectional (7/15). The included studies failed to provide strong evidence of an association between any of the potential interventions and Salmonella prevalence, due to the potential for confounding, and the failure to document temporal association between the intervention and Salmonella prevalence. The strongest evidence of an association was found for feed form, i.e. the use of non-pelleted may be potential interventions associated with reduced Salmonella prevalence. The uncertainty is primarily based on studies containing moderate to low evidentiary value or insufficient numbers of tested individuals, resulting in a low degree of confidence that results could be extrapolated to the target population. Therefore, the conclusion of the review is that there should be a low level of comfort among qualified scientists that the claimed association between non-pelleted feed and reduced Salmonella prevalence is scientifically valid. There is no strong evidence regarding associations between presence of Salmonella and the other feed characteristics examined.