Salmonella Dublin associated with abortion in dairy cattle in Algiers and comparison of different diagnostic methods.

Background: In cattle, many serotypes of Salmonella enterica are responsible for a wide variety of clinical manifestations, which can cause considerable economic loss. Some serotypes can cause cows to abort sporadically, such as the Dublin serotype. This study was carried out on different cattle farms in the Algiers region to determine the prevalence of Salmonella Dublin using bacteriological and immunological methods. Methodology: The prevalence of Salmonella was determined by bacteriological analysis in accordance with the reference method AFNOR NF U 47-100 on faecal samples collected from 184 cattle belonging to 19 different farms, and serotyping for S. Dublin. Immunological analysis by enzyme-linked immunosorbent assay (ELISA) for S. Dublin was carried out on milk samples collected from 91 cattle. A survey of case (n=5) and control (n=14) farms for comparative analysis was performed to demonstrate a link between abortion in cows and prevalence of S. Dublin with both bacteriological and immunological methods. Sensitivity, specificity, Cohen Kappa coefficient, McNemar test odds ratios, and confidence intervals were calculated using Winepiscope 2.0 and StatA 9.1 software, and p<0.05 was considered as statistically significant. Results: The bacteriological results showed a prevalence of 7.6% (95%CI: 3-10), for Salmonella and serotyping revealed a prevalence for S. Dublin of 2.7%. The immunological analysis of milk by the ELISA technique revealed a prevalence of 13.2% (95%CI: 5-20) for S. Dublin. The comparative study between immunological results from milk and bacteriological results from faeces for detecting S. Dublin showed poor agreement between the two tests (k=0.25), with enzyme immunoassay being significantly more sensitive than the bacteriological test (p<0.05). The results of the survey did not demonstrate a clear association between bacteriological detection of S. Dublin in faeces and abortion in cows (OR=8.66, 95%CI: 0.58-130.12). However, with the immunological analysis of milk for S. Dublin, there was a significant positive association (OR=62.33, 95%CI: 2.13-18.22) between a positive antibody response to S. Dublin in milk and the presence of abortions on the farm. Conclusion: In view of these results, we can conclude that Salmonella infections should systematically feature in the differential diagnosis of abortions in dairy cattle in Algeria. [ABSTRACT FROM AUTHOR]