Herd-level diagnosis for Salmonella enterica subsp. enterica serovar Dublin infection in bovine dairy herds.

Herd-level sensitivities of bacteriological and serological methods were compared in 79 bovine dairy herds, recently infected with Salmonella enterica subsp. enterica serovar Dublin. All farms experienced clinical signs of salmonellosis for the first time and had no history of vaccination against salmonellosis. At the start of the study, infection with serovar Dublin was confirmed with at least one positive bacteriologic culture for serovar Dublin from a clinical case (gold standard for herd infection). Bacteriological culture was done on samples of dung-pits, drinking water, bulk-milk filters, and faeces of animals with current or earlier clinical signs of salmonellosis. Blood samples of all animals and bulk-milk samples were tested using an ELISA.Herd-level sensitivity (HSe) of culture of dung-pits, drinking water, bulk-milk filters, and faeces of animals with current or earlier signs of salmonellosis was 45, 5, 7, and 38%, respectively. HSe for serology of all animals was 100%. If blood samples of all calves 4-6 months old were examined, at least one calf was seropositive on 91% of the infected farms. If serology was performed on samples of animals with current or earlier signs of salmonellosis, at least one animal was seropositive on 80% of the infected farms. HSe for bulk-milk samples was 54%. However, if clinical signs of salmonellosis were observed only in lactating animals, sensitivity of bulk-milk serology was 79%. Interesting combinations of methods were the combination of serology of bulk milk with either serology of animals with current or earlier signs of salmonellosis (HSe=91%), or serology of all calves of 4-6 months old (HSe=99%).