Abstract #354

# 354
A cohort study on the carcass weight loss associated with bovine leukemia virus infection in dairy production in Hokkaido, Japan.
S. Nakada*1,2, Y. Fujimoto2, J. Kohara3, Y. Adachi4, K. Makita2, 1Hokkaido Higashi Agriculture Mutual Aid Association, Nakashibetsu, Japan, 2Veterinary Epidemiology Unit, Graduate School of Veterinary Medicine, Rakuno Gakuen University, Ebetsu, Japan, 3Animal Research Center, Agricultural Research Department, Hokkaido Research Organization, Shintoku, Japan, 4Hayakita Meat Inspection Center, Iburi Sub-Prefectural Bureau, Hokkaido Prefectural Government, Hokkaido, Japan.

Bovine leukemia virus (BLV) is widely prevalent in Japanese dairy farms. To understand the burden of BLV infections in Japan, it is important to estimate the economic loss. The objective of our study was to assess the loss of carcass weight (CW) for dairy culled cows due to BLV infection. In total, 226 Holstein culled cows from 12 commercial dairy farms participating in the routine BLV infection monitoring scheme in the Nemuro and Kushiro regions of Hokkaido, Japan were studied. Information on age and the last delivery day was collected. BLV infection status was categorized according to the proviral load (PVL) which was measured using 2 different but well correlated tests according to the farms. Those cows whose PVL was not detected in the regular blood test just before import slaughterhouse were categorized into non-infected culled cows. Those cows whose PVL was under 2,500 copies/50 ng DNA or 60,000 copies/105 Cell by continuous 2 times blood tests just before sending to slaughterhouse was categorized as low PVL (LPVL), while those with PVL higher than 2,500 copies/50 ng DNA or 60,000 copies/105 Cell were high PVL (HPVL). CW information and postmortem examination results of the culled cows were provided by meat companies and meat inspection centers, respectively. A Generalized Linear Mixed Model (GLMM) framework with Gamma errors were used to estimate CW, age, lactation stage, status of BLV, and their interaction term as explanatory variables and owner was random effect. Additionally, GLMM with Poisson errors were applied to reckon the abnormal finding of the postmortem examination (AFPE), status of BLV was explanatory variable, and owner was random effect. GLMM with Gamma errors were practiced calculation CW, AFPE was explanatory variable, and owner was random effect. The mean CW of HPVL culled cows was 38.52 kg smaller than that of non-infected (P = 0.039). As the number of AFPE increased by 1, mean CW decreased 14.96 kg (P < 0.001). The mean number of AFPE of HPVL culled cows was significantly larger by 1.01 than non-infected cows (1.60, P = 0.018). In summary, BLV infection causes decrease in CW, and this may be due to the pathology of internal organs in persistent high viremia cows.

Key Words: bovine leukemia virus (BLV), carcass weight, postmortem examination