Vaccination for Johne's disease with killed inactivated vaccine in cattle herds has shown variable success. The vaccine delays the onset of disease but does not afford complete protection. Johne's disease vaccination has also been reported to interfere with measurements of cell-mediated immune responses for the detection of bovine tuberculosis. Temporal antibody responses and fecal shedding of Mycobacterium avium subsp. paratuberculosis, the causative agent of Johne's disease, were measured in 2 dairy cattle herds using Johne's disease vaccine (Mycopar) over a period of 7 years. Vaccination against Johne's disease resulted in positive serum M. avium subsp. paratuberculosis antibody responses in both herds, and the responses persisted in vaccinated cattle up to 7 years of age. Some vaccinated animals (29.4% in herd A and 36.2% in herd B) showed no serological reactivity to M. avium subsp. paratuberculosis. M. avium subsp. paratuberculosis-specific antibody responses were also detected in milk from Johne's disease-vaccinated animals, but fewer animals (39.3% in herd A and 49.4% in herd B) had positive results with milk than with serum samples. With vaccination against M. avium subsp. paratuberculosis, fecal shedding in both dairy herds was reduced significantly (P < 0.001). In addition, when selected Johne's disease-vaccinated and -infected animals were investigated for serological cross-reactivity to Mycobacterium bovis, no cross-reactivity was observed.