Comparison of two hydrodensitometric methods for estimating percent body fat

This study compared the two following hydrodensitometric methods for estimating percent body fat (%BF): 1) estimation of residual volume (RV) by helium dilution before and after measurement of immersed mass at RV, and 2) determination of immersed mass at a comfortable level of expiration (approximately functional residual capacity) with measurement of the associated gas volume by oxygen dilution. Twelve men [27.9 ± 7.5 (SD) yr; 79.32 ± 12.79 kg; 180.5 ± 9.9 cm] were tested for %BF via both methods on each of two separate visits within 3 days by using a counterbalanced design. The two helium dilution measurements yielded a technical error of measurement of 0.2% BF and an intraclass correlation coefficient of 0.999. Corresponding values for the oxygen dilution method were 0.4% BF and 0.999, respectively. There was no difference (P = 0.80) between the helium dilution (16.9 ± 9.3% BF) and oxygen dilution (16.9 ± 9.4% BF) methods, and the individual differences ranged from -0.7 to 0.6% BF. The interclass correlation coefficient between the two methods was 0.999 with a SE of estimate of 0.4% BF. Whereas both methods were precise and reliable and yielded similar results, the oxygen dilution technique was more expedient and was preferred by the subjects because they were not required to exhale to RV.