Associations between reproductive performance and seropositivity for bovine leukemia virus, bovine viral-diarrhea virus, Mycobacterium avium subspecies paratuberculosis, and Neospora caninum in Canadian dairy cows.

Our objective was to investigate effects of seropositivity for bovine leukemia virus (BLV), Type 1 bovine viral-diarrhea virus (BVDV), Mycobacterium avium subspecies paratuberculosis (MAP), and Neospora caninum (NC), and their possible interactions, on reproductive efficiency (specifically, first-service conception [FSC], and calving interval [CI]) in dairy cows. The sample population included up to 30 randomly selected animals from 179 randomly selected farms in five provinces in Canada, from which 23 farms did not meet the inclusion criteria for the final dataset. Serum samples were tested for antibodies against the stated pathogens using commercially available diagnostic tests. A Cox proportional hazards model with shared (herd-level) frailty was utilized to analyze the CI data. In this model, BLV-seropositive cows had a 7% lower rate of conception compared to seronegative cows (P=0.06). Mixed logistic regression models of CI>484 days, CI>534 days, and CI>584 days were built to explore factors of long CIs. These cut-offs were selected to represent calving-to-conception intervals of >200 days, >250 days, and >300 days. BLV-seropositive cows had higher odds of having a CI>484 days compared to BLV-seronegative cows, and BLV serostatus interacted with lactation number in this model, with 1st lactation seropositive cows being more likely to have a CI>484 days than older seropositive cows. NC-seropositive cows had a 1.27 times higher odds of exhibiting a CI>484 days, a 1.37 times higher odds of a CI>534 days, and a 1.54 times higher odds of a CI>584 days, compared to NC-seronegative cows. Neither BVDV nor MAP seropositivity showed any significant effect in these models. For the FSC models, a first service was classified successful (pregnancy=1) if it was the cow's last service and she calved 270-290 days later. A mixed logistic regression model of FSC revealed an interaction between NC and BVDV-seropositivity at the herd level, with odds ratios of 0.64, 1.06 and 0.85 for NC-seropositive cows (compared to NC-seronegative cows) in BVDV-seronegative, BVDV-seropositive and BVDV-missing herds, respectively. BLV and MAP seropositivity had no significant impact on FSC. All models controlled for herd-clustering effects, and included parity, linear score of somatic cell counts, peak milk, and province to control for confounding. The overall FSC was 51%, the average CI was 393 days, and 18%, 9% and 5% of lactations had CI>484 days, >534 days, and >584 days, respectively.