Tropical and Subtropical Agroecosystems (Jun 2009)
COMPARISSON OF DIFFERENT IN VIVO ESTIMATORS OF BODY FAT AND MUSCLE CONTENT IN ADULT CREOLE GOATS
Abstract
The objective of the present study was to compare the effectiveness of ultrasound measurements (ULT), body condition score (BCS) and body mass index (BMI) to predict body fat and muscle in goats. Twenty-four adult, ovariectomized, creole goats were fed individually to achieve dissimilar stable body weights (26 to 66 kg). After a six-week period of stabilization BCS was evaluated (average of 4 separate evaluations) and BMI calculated (BW/(length x height)x10; average of 2 independent length and height measurements). Fat and muscle coverage over the dorsal (12-13th), lumbar (1st) and coxal (1st) vertebrae were measured by ultrasound imaging. Animals were slaughtered and visceral fat dissected and weighed. Left half carcass weights were obtained and muscle and fat were dissected and weighed. Left half carcass data were adjusted to whole carcass data. Simple  regression analyses were performed with ULT, BCS and BMI as predictors and visceral (VF), carcass (CF) and total fat (TF), and muscle (MU) as response variables. Coefficients of variation (CV) were calculated for BCS and BMI estimations. ULT measurement of total tissue (fat + muscle) coverage over dorsal vertebrae presented the highest predictive value for ULT measurements; thus this measurement is the only reported. Coefficients of determination (R2) for VF, CF and TF were .55, .76 and .68; .78, .82 and .87; .81, .81 and .88 for ULT, BCS and BMI as predictors. R2 for MU were .67, .84 and .76 using ULT, BCS and BMI as predictors. R2 for proportions of VF/BW, CF/BW and TF/BW were .47, .73 and .65; .75, .78 and .88; .74, .71 and .85; when ULT, BCS and BMI were used as predictors. R2 for MS/BW were ≤ .10 using ULT, BCS and BMI as predictors. CV was greater in BCS estimation as compared to BMI estimation (9.7 vs. 3.1 %). BCS and BMI were more accurate in vivo predictors than ULT for body fat content, fat as proportion of BW and body muscle content. ULT, BCS and BMI were completely worthless for predicting muscle as proportion of BW. BMI estimation was less variable than BCS estimation. This study, for the first time, provides a predictive value of BMI for body fat and muscle content in goats.
