Metabolites and physical scores as possible predictors for postpartum culling in dairy cows

The purpose of the study was to explore the associations of serum non-esterified fatty acids (NEFA) and β-hydroxybutyrate (BHBA) concentrations with the body condition score (BCS) and rumen fill score (RFS) in order to predict the risk of postpartum culling, and to further investigate effective monitoring stages during the dry period. From October 2012 to March 2014, clinically healthy Holstein heifers and cows were sampled once before calving, and the occurrence of culling within 60 days in milk (DIM) was investigated. The discriminatory ability of each parameter was evaluated using receiver operating characteristic (ROC) analysis. Of the 720 cows sampled between 14 and 2 days before the actual day of calving in the study, 42 cull cows (mean DIM ± SE: 22.0 ± 2.6) were confirmed. The areas under the curve (AUC) of the ROC for predicting culling using serum NEFA concentrations were 0.6 and 0.7 at 14 to 2 and 7 to 2 days before calving, respectively. The AUC for the RFS was 0.7 for both periods, indicating the same diagnostic level as the serum NEFA concentration. Both the serum NEFA concentration and RFS were possible predictors in cows with ≥ 2 parities, but not in cows with 0–1 parity sampled even at 7 to 2 days before calving. The serum BHBA concentration and BCS were not suitable predictors of culling for any period or parity. These results indicate that RFS has a discriminatory ability comparable to the serum NEFA concentration for predicting culling within 60 DIM.