Validity of optical device lipometer and bioelectric impedance analysis for body fat assessment in men and women.

The aims of this study were to validate different subcutaneous adipose tissue layers (SAT-layers) measured by lipometer for body fat percentage (BF%) assessment with dual-energy X-ray absorptiometry (DXA) and to compare the validity of lipometer and bioelectrical impedance analysis (BIA). The subjects were 21 male (18-60 years) and 19 female (23-54 years) healthy Estonian volunteers. SAT-layers were measured by lipometer using 15 standardized SAT-layers. Sum of arms, legs and trunk SAT-layers were calculated and compared with arms, legs and trunk fat percentage measured by DXA. BF% was calculated by BIA using the equations of Lukaski et al. and Chumlea et al. for both genders and the equations of Segal et al. for males and Van Loan and Mayclin for females. BF% measured by DXA was significantly higher than calculated by Lukaski et al. and Chumlea et al. in both genders. The correlation was highest between the BF% measured by DXA and using Segal et al. equation in males (r = 0.94) and Van Loan and Mayclin equation in females (r = 0.84). High relationship was observed between BF% measured by DXA and sum of 15 SAT-layers (r = 0.88 in males and r = 0.91 in females). Stepwise multiple regression analysis indicated that two selected SAT-layers explained 85.9% and 86.7% (R2 x 100) of the total variance in BF% measured by DXA in males and females, respectively: [BF% = 1.308 neck + 0.638 hip + 6.971 (males; SEE = 2.59) and BF% = 1.152 hip + 1.797 calf + 12.347 (females; SEE = 3.46)]. In conclusion, lipometer and BIA give a similar mean estimation of BF% when compared with DXA. However, there is a wide range of variance for the upper and lower limits of agreement between the methods, and the methods are not interchangeable. Lipometer seems to be superior to BIA.