Validation of dual x-ray absorptiometry for body-composition assessment of rats exposed to dietary stressors.

Evidence of the validity and accuracy of dual x-ray absorptiometry (DXA) to measure soft-tissue composition of laboratory rats with altered body composition associated with nutritional perturbations is lacking. We compared DXA determinations made in prone and supine positions with measurements of chemical composition of 49 male, weanling Sprague-Dawley rats that were fed the basal AIN-93 growth diet, were fed the basal diet modified to contain 30% fat, were fasted for 2 d, were limit fed 6 g of the basal diet daily for 1 wk, or were treated with furosemide (10 mg/kg intraperitoneally 2 h before DXA). DXA produced similar estimates of body mass and soft-tissue composition in the prone and supine positions. DXA estimates of body composition were significantly correlated with reference composition values (R(2) = 0.371-0.999). DXA discriminated treatment effects on body mass, fat-free and bone-free mass, fat mass, and body fatness; it significantly underestimated body mass (1% to 2%) and fat-free and bone-free mass (3%) and significantly overestimated fat mass and body fatness (3% to 25%). The greatest errors occurred in treatment groups in which body mass was diminished and body hydration was decreased. These findings suggest that DXA can determine small changes in fat-free, bone-free mass in response to obesity and weight loss. Errors in DXA determination of fat mass and body fatness associated with extra corporeal fluid and dehydration indicate the need for revision of calculation algorithms for soft-tissue determination.