S. Dexter, K. Christison, C. Dumke, FACSM

University of Montana, Missoula, MT

The use of ultrasound has become a popular method to evaluate body composition as it claims to remove technician errors that can often occur in skinfold assessments. PURPOSE: To evaluate the validity of using ultrasound (ULTRA) to determine body composition compared to hydrostatic weighing (HW) and skinfold (SK) techniques. METHODS: Twenty two subjects (18 male, 6 female, 22.1 ± 0.6 yrs, 78.9 ± 3.5 kg, 177.7 ± 2.0 cm) underwent body composition analysis using SK, ULTRA, and HW in the same session while being a minimum of 4 hours fasted. Both SK and ULTRA assessments were taken using the 3-site Jackson and Pollock method for males (chest, abdomen, and thigh) and a 4-site Jackson, Pollock, and Wade analysis for females (tricep, suprailiac, abdomen, thigh). HW was measured using a hydrostatic weighing tank with three force transducers. Residual volume was estimated from the subject's height, age, and weight. Subjects were submerged and weighed repeatedly until 3 consistent measurements were recorded within 0.1 kg. Body fat percentage (BF), fat mass (FM), and fat free mass (FFM) were recorded. Data was analyzed using paired sample t-tests and a one-way ANOVA. Data is presented as mean ± SEM. RESULTS: Across the measurement modalities, no significant differences were seen in BF (13.9 ± 1.2 % ULTRA, 13.4 ± 1.3 % SK, 12.3 ± 0.7 % HW, p =0.57), FM (10.7 ± 0.9 kg ULTRA, 10.3 ± 0.9 kg SK, 9.7 ± 0.7 kg HW, p = 0.75), and FFM (68.2 ± 3.5 kg ULTRA, 68.6 ± 3.5 kg SK, 69.2 ± 3.1 kg HW, p = 0.98). There was a significant difference in subcutaneous skinfold thickness measured between SK and ULTRA (p < 0.009). Since ULTRA measures one subcutaneous fat layer while SK measures two, a ratio factor was determined at each site (1.49 ± 0.08 abdomen, 2.26 ± 0.09 thigh, 1.69 ± 0.08 chest, 1.90 ± 0.19 tricep, 2.07 ± 0.16 suprailiac). CONCLUSION: Based on these results, ULTRA appears to be a valid measure compared to SK and HW in accurately assessing body composition. However, the assumption of a doubling of the subcutaneous fat layer is not consistent across the anatomical sites. It remains unclear how ULTRA accounts for this variance in their proprietary algorithm.

This document is currently not available here.