Reliability of Compartmental Body Composition Measures in Adults Using GE iDXA: Implications for Nutrition Research

Nutrition intervention studies are increasingly using DXA to measure changes in body composition over time. The reliability and precision of any instrument is important in order to interpret whether measured changes in body compartments are significant and/or clinically relevant. This study aimed to measure the precision and reliability of body compartment measures obtained from the GE iDXA over a 3 month period. Weight stable adults (±2 kg) aged 18–65 years participated in total body composition scans, performed 3 times with at least 1 month between visits. Despite weight-stability, the coefficient of variation was high for each of the body compartment measures over 3 months (range 0.8–5.9%). Precision-scans conducted on the same day with subject re-positioning only resulted in <2% variation. The coefficient of variation (CV%) for visceral adipose tissue (VAT) was 31.6%, significantly higher than other components. The least significant change for VAT mass over 3 months was 206.3 g. VAT CV% was negatively correlated with BMI (−0.537, p < 0.001), total fat mass (−0.489, p < 0.001), and VAT mass (−0.795, p < 0.001), indicating that individuals with higher body fat and BMI > 25 kg/m2 have a reduced variation in VAT measurement over 3 months. To accurately measure change in VAT via iDXA, it is recommended that participants require a BMI > 25 kg/m2, or VAT mass > 500 g. Any observed changes in VAT mass < 500 grams should be interpreted with caution due to the lack of precision.