Multivariable analysis to investigate factors associated with day 14 to 16 pregnancy rate in Thoroughbred mares on an intensively managed UK stud farm

Reproductive losses limit the output of the Thoroughbred mare. Identifying factors that influence pregnancy rates could help inform practices aimed at maximizing productivity. The most recent UK-based study to identify factors influencing day 14-16 pregnancy rates utilized only univariable analysis and was conducted on data from over 20 years ago. The aim of this study was to use multivariable analysis to investigate associations between mare and cycle level factors and day 14-16 pregnancy rates (PRs) on an intensively managed UK Thoroughbred stud farm. A retrospective cohort study was performed on all available reproductive data from farm and veterinary records for the 2017-2021 breeding seasons. Information was collected on season, month of cover, stallion, mare age, status, parity, location at cover, covered cycle number (within the season), number of coverings and ovulations (within the covered cycle), and all reproductive therapeutics (estrous induction agents, ovulatory agents, and pre-and post-covering treatments) administered. Multivariable mixed effect logistic regression modelling, with mare included as a random effect in all analyses, was used to identify factors associated with a positive pregnancy diagnosis at days 14-16 post-cover. Data were available for 1098 covered estrous cycles in 360 mares over 5 breeding seasons. The overall day 14-16 PR (proportion of coverings with a positive pregnancy diagnosis at day 14-16 post-cover) for the study period was64.5% (708/1098; 95% confidence interval (CI) =61.6, 67.3). In the final multivariable model (adjusted for the effect of the mare (Theta 0.39, P=0.003) increasing mare age (OR 0.93 per year increase in mare age, 95% CI 0.89, 0.98 P=0.01), a course of altrenogest pre-cover (administered to n=61/833 cycles OR 0.47, 95% CI 0.26, 0.85 P=0.01), treatment with intrauterine hyperimmune plasma pre-cover (n=61/829 OR 0.39, 95% CI= 0.21, 0.75 P=0.005), and the use of sole -oxytocin therapy post-cover (n=183/829 OR 0.59, 95% CI 0.40, 0.87 P=0.007) were all associated with reduced likelihood of a positive pregnancy diagnosis at days 14-16 post-cover. Although therapeutic agents could directly influence PRs, it is also likely that their use represents conditions such as acyclicity, subfertility and/or post-breeding endometritis, which may themselves affect PR. Controlled studies to further evaluate the efficacy of these treatments, particularly concerning the use of oxytocin and/or carbetocin, are therefore recommended. In conclusion, this study has provided up-to-date estimates of day 14-16 PRs in Thoroughbred broodmares in a UK setting and enhanced understanding of factors that may be associated with these rates. In addition, knowledge gaps around the rationale and efficacy of some therapeutic agents have been identified; these can help inform future research priorities.