Validity of the portable ultrasound bodymetrix™ Bx-2000 for measuring body fat percentage

Abstract

Bodymetrix™ Bx-2000 (IntelaMetrix, Livermore, CA, USA) has been introduced as one of the alternatives and portable methods to estimate body fat percentage. However, inconsistent results between protocols built-in the Bodymetrix™ may be compelling the question of its validity. Thus, this study first investigated the possible errors between protocols and evaluated the validity of body fat percentage (BF%) compared to the gold standard method (dual-energy X-ray absorptiometry, DEXA). One hundred and five collegiate males, aged 20.01 ± 2.11 years, body height, 174.81 ± 6.01 cm, body mass, 73.26 ± 13.60 kg, and body mass index, 23.91 ± 3.77 kg m-2 participated in the present study. Participants’ body fat percentage was estimated by built-in nine different protocols in the Bodymetrix™ BX-2000 using A-MODE ultrasound. Pearson correlation (r), Mean absolute percentage errors (MAPEs), Bland & Altman plots, and Equivalence testing were used to examine the validity of each protocol by comparing it to the criterion measure (i.e., DEXA). The results indicated good potential for almost all of the protocols in correlation (Min: r = 0.79, Max: r = 0.92)., MAPEs (Min: 20.0%, Max: 33.8%), and Bland-Altman (Min diff: 16.7, Max diff: 41.4). Particularly, the estimated BF% from protocol 7 (4-sites by Durnin & Wormersley) and protocol 9 (9-sites Parllo) were completed within the equivalence zone (-10% of the mean). The estimates measured by protocol 7 and protocol 9 identified as the most valid methods for estimating BF% using a Bodymetrix™ Bx-2000, compared to the DEXA. Our findings provide valuable information when applying in young male individuals, but future studies with other populations such as female or adolescents may be required to suggest a valid protocol within the instrument.