ABSTRACT
BACKGROUND: Information on precision errors and the least significant change (LSC) of dual energy X-ray absorptiometry (DXA)-derived body composition estimates is scarce, particularly for the appendicular lean mass (ALM) and appendicular lean mass index (ALMI). Overnight fasting is recommended for body composition measurements but has not been well tolerated by some elderly patients. This study aimed to establish precision errors and LSC values of body composition estimates in all regions-including visceral adipose tissue (VAT) and ALM-and the ALMI to assess the effect of a small meal on body composition and to estimate the changes it incurred. METHODOLOGY: Our institutional review board approved the study protocol. Altogether, 36 non-obese men aged ≥60 years, having given written informed consent, underwent body composition assessment after fasting overnight except for water. They underwent DXA scans three times, each time with repositioning (to simulate the clinical setting), the last after consuming a standardized meal (210-250 g and 200 cc of water). RESULTS: Precision errors and LSC values of DXA-derived body composition estimates in these elderly men tended to be higher than those in reports on younger subjects. Coefficients of variation (CVs (%)) of total bone mass (Tb.BMC) and total lean mass (Tb.LM) were <1%, whereas those of total fat mass (Tb.FM) and total %fat mass (Tb.%FM) were <2%, with LSCs of 45.8 g, 706.52 g, 731.4 g, and 1.15%, respectively. The CVs (LSC) of VAT, ALM, and ALMI were 8.9% (150.65 g), 0.93% (501 g), and 0.94% (0.19), respectively. After meal consumption, the mean changes in Tb.FM, Tb.BMC, and Tb.LM were -100, -8.2, and 440 g, respectively. CONCLUSIONS: Effects of a small meal on most parameters were trivial, including those for VAT, ALM, and ALMI, where changes were not statistically significant. None exceeded the LSC of ALM and ALMI, suggesting that a small meal is allowable before these measurements.
