Evaluation of electrical broad bandwidth impedance spectroscopy as a tool for body composition measurement in cows in comparison with body measurements and the deuterium oxide dilution method.

Body fatness and degree of body fat mobilization in cows vary enormously during their reproduction cycle and influence energy partitioning and metabolic adaptation. The objective of the study was to test bioelectrical impedance spectroscopy (BIS) as a method for predicting fat depot mass (FDM), in living cows. The FDM is defined as the sum of subcutaneous, omental, mesenteric, retroperitoneal, and carcass fat mass. Bioelectrical impedance spectroscopy is compared with the prediction of FDM from the deuterium oxide (DO) dilution method and from body conformation measurements. Charolais × Holstein Friesian (HF; = 18; 30 d in milk) crossbred cows and 2 HF (lactating and nonlactating) cows were assessed by body conformation measurements, BIS, and the DO dilution method. The BCS of cows was a mean of 3.68 (SE 0.64). For the DO dilution method, a bolus of 0.23 g/kg BW DO (60 atom%) was intravenously injected and deuterium (D) enrichment was analyzed in plasma and whey by stabile isotope mass spectrometry, and total body water content was calculated. Impedance measurement was performed using a 4-electrode interface and time domain-based measurement system consisting of a voltage/current converter for applying current stimulus and an amplifier for monitoring voltage across the sensor electrodes. For the BIS, we used complex impedances over three frequency decades that delivers information on intra- and extracellular water and capacity of cell membranes. Impedance data (resistance of extra- and intracellular space, cell membrane capacity, and phase angle) were extracted 1) by simple curve fit to extract the resistance at direct current and high frequency and 2) by using an electrical equivalent circuit. Cows were slaughtered 7 d after BIS and D enrichment measurements and dissected for the measurement of FDM. Multiple linear regression analyses were performed to predict FDM based on data obtained from body conformation measurements, BIS, and D enrichment, and applied methods were evaluated by cross-validation. The FDM varied widely between cows and was correlated to D enrichment in plasma ( = 0.91, < 0.05). Prediction of FDM by body size measurements was less precise ( = 0.84), but FDM prediction was more accurate using D enrichment in plasma ( = 0.90) and BIS ( = 0.99) data. Therefore, both BIS and D enrichment analysis resulted in similarly good predictions of FDM in cows, and we conclude that BIS could have the potential to predict FDM in dairy cows from 40 to 380 kg.