Reliability and accuracy of segmental bioelectrical impedance analysis for assessing muscle and fat mass in older Europeans: a comparison with dual-energy X-ray absorptiometry.

Segmental determination of muscle and fat mass (MM, FM) attains growing importance for judging effects of training and malnutrition in older people. This study evaluated the reliability and accuracy of segmental bioelectrical impedance analysis (sBIA) for use in older people. In 72 (40 men, 32 women) healthy elderly (mean age 69.0 ± 4.8 years), the MM and FM of right and left arm (RA, LA), right and left leg (RL, LL), and trunk were determined by sBIA (BC-418-MA, Tanita) and dual-energy X-ray absorptiometry (DXA) as a reference method. The sBIA provided in both sexes reliable values for limb and truncal MM and FM, except for MM of RL in women. The accuracy of sBIA displayed sex-specific bias. For MM, accurate values were noted for men's trunk and women's limbs (except LA). By contrast, MM was significantly underestimated in men's limbs by 6-18% and overestimated in women's LA (13%) and trunk (14%). Estimates of FM were accurate for men's arms as well as women's legs and trunk. However, FM was significantly overestimated in men's legs (34-37%) and trunk (60%), but underestimated in women's arms (27-35%). The proportional deviations of sBIA estimates from DXA values for limbs and trunk were significantly related to the respective MM or FM. The sBIA tends to underestimate MM in men and to overestimate in women. The reverse occurs for FM. The actual equations of the Tanita device may not completely represent the European older population and should be partly revised.