RESUMO
BACKGROUND: Dual-energy X-ray absorptiometry (DXA) measures are affected by the noise produced by external factors such as textile compression found in loose clothing. The study aimed to determine the effect of a compression bandage (CB) on body composition (BC) assessed by DXA. METHODS: Sixty volunteers (age=21.4±4.7yr.) underwent full-body DXA scans on a control (CTRL) condition and after wearing a 30-mmHg CB on the trunk, legs, and arms. ANOVA (2 genders by 2 experimental conditions) determined mean interactions in BC variables tissue body fat% (BF%), region body fat% (RBF%), body tissue (BT), fat mass (FM), lean mass (LM), bone mineral content (BMC), and total mass (TM). Absolute reliability in BC scores was studied by the typical error of the measurement (TEM), the coefficient of variability (CV), and Bland-Altman plots. RESULTS: ANOVA interactions were found on tissue total BF% (p=0.049), RBF% (p=0.048), android lean mass (p=0.004), and android total mass (p=0.019). The CV was small for tissue BF% (2.61±0.93%, CI95%=0.79, 4.43%), RBF% (2.66±1.78%, CI95%=-0.83, 6.15%), BT (4.82±2.19%, CI95%=0.54, 9.10kg), FM (4.17±2.25%, CI95%=-0.24, 8.58kg), LM (3.25±2.44%, CI95%=-1.53, 8.04kg), BMC (4.81±2.96%, CI95%=-0.99, 10.62kg), and TM (2.84±2.80%, CI95%=-2.65, 8.33kg). Bland-Altman plots showed random error for BF%, LM, and BMC. A bias of 0.5% was observed on BF% in males. CONCLUSION: A CB worn during a full-body DXA scan elicited similar BC scores than not wearing it. The variation in scores was <10% for most BC variables, and a trivial bias of 0.5% in BF% was detected in male scores.