Hair mineral analysis using an X-ray fluorescence spectrometer as a method of estimating body fat percentage (BF%) was investigated. Body fat percentage of 24 healthy Japanese, aged 20-27 years, was measured using a hand-held impedance analyzer (BF%IMP). X-ray (Kα-ray) intensities of sulfur, chlorine, potassium, calcium, titanium, and iron (Fe) in hair were measured using an X-ray fluorescence spectrometer. Body fat percentage was also measured using a Hologic whole body dual energy X-ray absorptiometer (BF%DXA) in nine subjects selected from the above 24 subjects based on their BF%IMP. Correlations of the two BF%s with Fe-Kα were significant (BFIMP: r = 0.60 n = 24, p <0.01; BFDXA: r = 0.67 n = 9, p<0.05). The mean (±SD) biases (measured minus estimated using multiple regression equations by Fe-Kα) for BF%IMP and BF%DXA were 2.97 ± 2.25% and 1.77 ± 1.33%, respectively. The SEEs for the two equations for BF%IMP and for BF%DXA were less than 4%. These results suggest that Fe-Kα may be a predictor of body fat percentage. However, the subjects were few and only Japanese in their twenties, so that further investigation is needed for methodological generalization.
Hair mineral analysis using an X-ray fluorescence spectrometer as a method of estimating body fat percentage (BF%) was investigated. Body fat percentage of 24 healthy Japanese, aged 20-27 years, was measured using a hand-held impedance analyzer (BF%IMP). X-ray (Kα-ray) intensities of sulfur, chlorine, potassium, calcium, titanium, and iron (Fe) in hair were measured using an X-ray fluorescence spectrometer. Body fat percentage was also measured using a Hologic whole body dual energy X-ray absorptiometer (BF%DXA) in nine subjects selected from the above 24 subjects based on their BF%IMP. Correlations of the two BF%s with Fe-Kα were significant (BFIMP: r = 0.60 n = 24, p <0.01; BFDXA: r = 0.67 n = 9, p<0.05). The mean (±SD) biases (measured minus estimated using multiple regression equations by Fe-Kα) for BF%IMP and BF%DXA were 2.97 ± 2.25% and 1.77 ± 1.33%, respectively. The SEEs for the two equations for BF%IMP and for BF%DXA were less than 4%. These results suggest that Fe-Kα may be a predictor of body fat percentage. However, the subjects were few and only Japanese in their twenties, so that further investigation is needed for methodological generalization.