Abstract

Background: Lean soft tissue (LST) of limbs is quickly measured by bioelectrical impedance analysis (BIA). This study, compared the accuracy of different standing multi-segments BIA devices by dualenergy X-ray absorptiometry (DXA) reference in older community-dwelling Taiwan adults.

Methods: The subjects recruited elderly people over 60 years old (46 males and 55 females) in Taiwan. The study used Tanita BC418MA (BC418), Biospace InBody230 (InBody230), Tanita MC-780MA (MC780), and DXA to measure the LST of upper limbs, lower limbs, and extremities. Ordinary least product regression and mean absolute percent error (MAPE) was taken as the overall effect indicators of the study.

Results: In the study group the average percent body fat (PBF%) of 101 elderly people was 34.4 ± 7.1%. The correlation coefficient (r) and standard error of the estimate (SEE) of BC418, InBody230, and MC780 relative to the LST (LST_app) of DXA in the appendicularwere 0.873, 0.734, 0.885, and 2.08 kg, 2.73 kg, 2.60 kg, respectively. The average differences between the three BIA devices and DXAwere 5.677 kg, 3.343 kg, and 4.511 kg, respectively. The corresponding limits of agreement (LOA) were 0.320 to 11.053 kg, -2.841 to 9.528 kg, and -0.274 to 9.296 kg, respectively. The MAPE were 38.05–47.82%, 22.40–30.99% and 31.30–39.38%.

Conclusion: The standing multi-segment BIA device is a convenient device for measuring the LST in the limbs of older community-dwelling adults having different measurement accuracy and limitations due to different brands and models. When applied to clinical evaluation, the application needs to be carefully used.