Salmonella serology compared to body weight for finishers and sows

Abstract

The relation between live body weight and Salmonella OD% for meat juice samples was investigated for finishers. The result was a seagull shaped curve indicating that the risk of being seropositive declines with increasing body weight from 60 to approximately 100 kilos and then rises again. The seagull curve is thought to be a combination of different trends: 1) A declining seroreaction caused by early Salmonella infection and 2) A higher background level of cross-react- ing antibodies with increasing age, induced by other Enterobactericeae. A study of 414 sows at slaughter revealed a significantly higher seroprevalence among sows than among finishers. This is probably caused by the continuous housing system resulting in repeated boosting of Salmonella antibodies. The relatively low bacteriological Salmonella excretion rate among sows indicates that the higher seroprevalence among sows is not necessarily reflecting a higher level of Salmonella excretion among sows. Introduction The purpose of the present study was to investigate the connection between Salmonella serology and live body weight (LBW) in pigs at slaughter and to test the seropreva- lence among sows in comparison to finishers. The vast majority of finishers are slaughtered at a LBW close to 100 kilos, but the LBW varies from under 80 to more than 130 kilos of LBW at slaughter. Serological data and carcass weight for finishers was recorded in a central register. For sows, serological data and LBW (140-300 kilos) were recorded in a pilot study. Materials and Methods According to the Danish surveillance and control program (Nielsen et al. 2001), 572,096 finishers were randomly selected for meat juice sampling in 2004. By using the public Zoonoses Register in the national Danish Central Herd Register, the carcass weight, and individual number for 542,270 of the finishers from the 17 largest slaughterhouses was obtained. The carcass weight was transformed into LBW with the formula: LBW = Carcass weight x 1.19 + 7.01 (Personal comm., Danish Crown, 2005). In a cross-sectional study, blood samples from 550 sows were taken at a slaughterhouse. The carcass weight was recorded for 414 sows. For sows, the LBW corresponds to carcass weight x 1.28 (Personal comm., Danish Crown, 2002). All meat juice and blood samples were analysed for Salmonella antibodies in the mix-ELISA used in the surveillance program (Nielsen et al., 1995), and the results were expressed as OD% (Nielsen et al., 1998). The dependency of Salmonella serology on LBW and thus indirectly age, (Vestergaard, unpub- lished results) was studied using partly the OD% as observed and partly the dichotomized values obtained by applying two cut-off values used in practice (20 OD% and 40 OD%). The use of a cut- off corresponds to the state of each animal being regarded as 'positive' or 'negative'. For the analysis of OD%, simple quadratic regression was used to obtain an OD% versus LBW curve. Likewise the slope in both sides of the curve could be tested. For the analysis of pos- itive/negative results, logistic regression was applied in the analogous way. The effect of sow ver- sus finisher was evaluated using a two-level indicator.