The use of ultrasonography in the transition period to estimate adipose tissue depots and their association with risk of early postpartum hyperketonemia in Holstein dairy cattle

An elevated prepartum BCS is a risk factor for postpartum hyperketonemia (HYK) and elevated nonesterified fatty acid (NEFA) concentrations in dairy cattle. The association between different adipose tissue depots, such as subcutaneous (SCAT) as well as visceral adipose tissue (VAT) depots, and HYK and elevated NEFA concentrations remains unknown. The objective of this 2-part study was to describe SCAT and VAT depots using ultrasonography during the transition period and to associate them with metabolic markers of negative energy balance in early postpartum dairy cows. Multiparous Holstein cows were enrolled in a prospective observational cohort study with 2 cohorts in subsequent study years (cohort 1: n = 31; cohort 2: n = 28). At −6, −4, −2, and 2 wk relative to calving (cohort 1) or −2 and 2 wk relative to calving (cohort 2) BCS was determined and cows were examined via transcutaneous ultrasound at 6 locations for prediction of adipose tissue mass in 5 depots: (1) abdominal adipose tissue (AAT), (2) retroperitoneal adipose tissue (RPAT), (3) omental adipose tissue (OMAT), (4) mesenteric adipose tissue (MAT), and (5) subcutaneous adipose tissue (SCAT). Postpartum serum NEFA and BHB concentrations were determined twice weekly from 1 to 14 DIM. Cows were categorized as having HYK or high NEFA concentrations (NEFA_H) if ≥1 sample resulted in a BHB ≥1.2 mmol/L or NEFA ≥720 µEq/L, respectively. Critical thresholds associated with HYK or NEFA_H for each depot at −2 wk relative to calving and the time of maximum estimated depot size were evaluated using logistic regression and a receiver operator characteristic analysis. Abdominal AT increased from −6 to −2 wk relative to calving in cohort 1 but did not differ from −2 to 2 wk relative to calving. Subcutaneous AT did not change from −6 to −2 wk but decreased from −2 to 2 wk relative to calving in cohort 1 and 2. Omental AT accurately (area under the curve [AUC] = 0.77) predicted HYK in cohort 1 but was noninformative for the prediction of HYK in cohort 2 (AUC = 0.49). Predicted AT depots were not informative for the prediction of NEFA_H (AUC ≤0.59) in either cohort. Results from this study suggest that AT can be monitored during the transition period using the described technique; however, estimated prepartum AT depots at −2 wk relative to expected calving were inaccurate in distinguishing between postpartum NEFA_H and non-NEFA_H or HYK and non-HYK.