Paratuberculosis (PTBC) is a chronic intestinal disease of animals caused by Mycobacterium avium paratuberculosis (MAP). MAP infection is diagnosed through indirect tests based on the immune response. The aims of this study were to compare the performance of two milk ELISA for the diagnosis of PTBC and to assess the bulk tank milk (BTM) MAP exposure in dairy cattle in Argentina. A total of 357 fecal, serum, and milk samples were collected. The fecal samples were processed by culture for MAP isolation, while both, serum and milk samples were used for the detection of antibodies by two different ELISA tests, "in-house" and commercial kit. MAP was isolated in 3.9% of fecal samples. For milk ELISA, poor concordances were obtained. Optimized cut-off points were calculated. The highest sensitivity and specificity values (64% and 80% respectively) were obtained with the combination of MAP isolation and commercial milk ELISA. The results indicate that the combination of different techniques to identify of dairy cattle infected with MAP increases the efficiency of diagnosis. In addition, BTM samples (n=98) were evaluated to determine herd status using the commercial kit during two seasons, identifying 33.3% of positive samples in autumn and 35.4% in spring.