We determined the risk factors for late embryonic mortality in dairy cows. We diagnosed pregnancy at 31 days and then confirmed the diagnosis at 45 days after artificial insemination (AI) via ultrasonography. The presence of an embryo with a heartbeat was the criterion for a positive pregnancy diagnosis. A diagnosis of late embryonic mortality was made when there was no positive sign of pregnancy in cows previously diagnosed as pregnant. The overall incidence of late embryonic mortality among 3,695 pregnancies was 6.9%. Logistic regression analysis revealed that herd size, AI month, synchronization protocol, and postpartum disease were important risk factors for late embryonic mortality. Herd size > 100 (odds ratio [OR]: 0.66, p < 0.05) and 50-100 lactating cows (OR: 0.63, p < 0.01) had lower risks of late embryonic mortality than herd size < 50 lactating cows. Cows inseminated during May-July had a higher risk (OR: 1.49, p < 0.05) of late embryonic mortality than cows inseminated during February-April. Cows inseminated after estrus following $PGF_{2{alpha}}$ treatment also had a higher risk (OR: 1.77, p < 0.001) of late embryonic mortality than cows inseminated following natural estrus. Lastly, cows with postpartum disease tended to have a higher risk (OR: 1.26, p < 0.1) of late embryonic mortality than cows without postpartum disease. In conclusion, late embryonic mortality associated with the herd size, AI month, synchronization protocol, and postpartum disease in dairy cows.

Keywords: dairy cows, late embryonic mortality, risk factors, ultrasonography