Mycobacterium avium (MA) is a potential food safety hazard in pigs. Blood samples of slaughtered pigs in the Netherlands and Germany were tested for the presence of MA antibodies to estimate the serological prevalence in the tested population. In the Dutch and German population 1.0% and 1.7% samples were positive, and 0.5% and 17.4% of the herds were at risk for having a MA infection respectively. The validity of the applied MA-ELISA was evaluated under field conditions. The specificity of the MA-ELISA was high (>98.4%). The average herd sensitivity was 18%. In the affected herds on average 50% of the animals were tested bacteriological positive for MA. It can be concluded that serological screening for the presence of MA antibodies is capable of identifying pig populations that are at risk for a MA infection.

Mycobacterium avium, both subspecies hominissuis (MAH) and subsp. avium (MAA), are considered a significant zoonotic hazard in pigs. Therefore special attention is given to detect the presence of this hazard in pigs during post mortem meat inspection. Herds delivered at slaughter were monitored on blood antibodies against MAH. Herds with an antibody respons against a MAH infection were visited. Initially a questionnaire assessing relevant risk factors for MAH was applied. Additionally to the questionnaire in several herds intracutaneous tuberculination was carried. Positive results in tuberculination in 3 different herds in the Netherlands, Belgium and Germany are presented; two farms where compost was used and one farm where the pig holding was adjacent to a big broiler farm. Twice the presence of MAH and once MAA was bacteriologically confirmed. When the supply of compost was stopped in two herds no positive tuberculination was present anymore. The other herd with the adjacent broiler flock ceased its activities as a pig producer. When preventive measures are an active part of daily farm management MAH can be controlled at farm level effectively. Screening blood of slaughter pigs on the presence of MAH antibodies can be used to identify true positive herds. Serological surveillance is presently applied in the newly developed supply chain meat inspection in Germany and The Netherlands.

Salmonella serology is used for classifying pig herds in risk categories in several national quality programs. Meat juice is used as test matrix in most of these programs. Two studies were done to compare the salmonella ELISA test results from meat juice with blood serum as a reference. Pig blood and meat samples for these studies were collected in one slaughterhouse. ELISA tests were done with a commonly applied commercial test. In the first study paired blood serum and meat juice samples from 182 pigs were collected and tested in two different laboratories. In the second study meat and blood samples were collected from 470 herds, over 20.000 samples for each matrix. The first study showed a linear relation between all matrices, but the OD values in meat juice were significantly lower than in blood serum. To obtain comparable outcomes in serum and meat juice, the blood serum OD%-values had to be reduced with 20 to 40%, depending on the lab that applied the test. This underestimation was confirmed in the second study. When the diagnostic cut off, OD10%, was applied on the blood samples, over 57% of the tested pigs showed antibodies and none of the slaughtered herds had fully negative serology, whereas with meat juice and a cut off at OD40% only 7,5% pigs were positive. It is concluded that meat juice testing for Salmonella antibodies can heavily underestimate the proportion of pigs that have encountered a salmonella infection. Consequently, pigs from herds that are categorised as low risk may be infected with salmonella. These pigs may therefore contaminate the lairage and the slaughter line. Monitoring results based on blood serology can not be compared with results based on meat juice, without taking care of the observed differences.