|Emenheiser, Joseph - Virginia Tech|
|Tait Jr, Richard|
|Notter, David - Virginia Tech|
|Lewis, Ron - Virginia Tech|
Submitted to: Journal of Animal Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/7/2014
Publication Date: 9/1/2014
Publication URL: http://handle.nal.usda.gov/10113/62206
Citation: Emenheiser, J.C., Tait, Jr., R.G., Shackelford, S.D., Kuehn, L.A., Wheeler, T.L., Notter, D.R., Lewis, R.M. 2014. Use of ultrasound scanning and body condition score to evaluate composition traits in mature beef cows. Journal of Animal Science. 92(9):3868-3877. DOI: 10.2527/jas.2014-7920.
Interpretive Summary: Real-time ultrasound is used extensively in the U.S. to evaluate young beef seedstock candidates for body composition traits. However, effectiveness of ultrasound to evaluate body composition traits in mature cows has not been evaluated. Mature cows destined for slaughter were repeat body condition scored and ultrasound scanned for fat and muscling traits by two experienced technicians to evaluate the repeatability and accuracy in evaluating body composition / carcass traits in cows. This work demonstrates that real-time ultrasound is an effective tool for measuring body composition traits in cows with very repeatable, accurate measures of fat and muscle traits. Criteria used to certify ultrasound technicians when scanning young beef animals are generally achievable when scanning mature cows. Ultrasound is a more repeatable and more trait specific (fat vs. muscle) assessment of cow composition than body condition scoring. Scientists and beef producers may identify situations where the increased precision of ultrasound over body condition scoring for measuring cow body composition is warranted.
Technical Abstract: The experiment was designed to validate the use of ultrasound to evaluate body composition in mature beef cows. Both precision and accuracy of measurement were assessed. Cull cows (n = 87) selected for highly variable fatness were used. Two experienced ultrasound technicians scanned and assigned BCS to each cow on two consecutive days. Ultrasound traits were backfat thickness (UBFT), LM area (ULMA), body wall thickness (UBWT), rump fat depth (URFD), rump muscle depth (URMD), and percentage intramuscular fat (UIMF). Cows were then harvested. Carcass traits were HCW, backfat thickness (CBFT), LM area (CLMA), body wall thickness (CBWT), and predicted marbling score (CPMS). Correlations between consecutive live measurements were greatest for subcutaneous fat (r > 0.94) for the two technicians, and lower for BCS (r > 0.74) and URMD (r > 0.66). Repeatability bias differed from zero for only 1 technician for URMD and UIMF (P < 0.01). The two technicians differed in repeatability SE for only ULMA (P < 0.05). Correlations between live and carcass measurements were high for UBFT and UBWT (r > 0.90), and slightly less for UIMF and ULMA (r = 0.74 to 0.79). Both technicians underestimated all carcass traits with ultrasound, but only CBFT and CBWT prediction bias differed from zero (P < 0.05). Technicians had similar prediction SE for all traits (P > 0.05). Technician effects generally explained <1% of the total variation in precision. After accounting for technician, animal effects explained 50.4% of remaining variation for BCS differences (P < 0.0001) but were minimal for scan differences. When cows with mean BCS <4 or >7 (9-pt scale) were removed, the portion of remaining variation defined by animal effects increased (to 15.7 to 23.6% from 1.8 to 11.2%) for UBFT, ULMA, URFD and URMD and was significant for UBFT and URFD (P = 0.03). Technician also defined trivial variation in accuracy (P > 0.24). Animal effects explained 87.2, 75.2, and 81.7% (P < 0.0001) of the variation remaining for CBFT, CLMA, and CBWT prediction errors, respectively, and remained large and highly important (P < 0.0001) in the reduced data. We conclude that experienced ultrasound technicians can precisely and accurately measure traits indicative of composition in mature beef cows. However, animal differences define substantial variation in scan differences and prediction errors. Implications for technician certification, carcass pricing, and genetic evaluation are discussed.