The Effects Of Dual-Energy X-Ray Absorptiometry-Derived Body Volumes On Percent Body Fat.

Abstract

Body volume (BV) is typically measured using hydrostatic weighing (HW) or air-displacement plethysmography (ADP). However, new BV prediction equations using Dual-Energy X-Ray Absorptiometry (DXA) have been developed and require validation. PURPOSE: Compare DXA-based BV predictions to HW and ADP and the effects on percent fat using a two-compartment (2C) and four-compartment model (4C). METHODS: Sixty physically inactive (<30 min. physical activity/wk), men (n=33) and women (n=27), 18-43 years of age, completed the study (25 +/- 7yr, 79.1 +/- 16.0kg, 171.0 +/- 9.0 cm, BMI 26.8 +/- 4.1). Body fat percentage (%Fat) was assessed using BV-based 2C models (DXA-BV, ADP, HW) and 4C models that included bone mineral content, BV, and total body water using deuterium oxide. DXA-BV, ADP, and HW BV values were also compared in addition to DXA-derived %Fat (DXA-%Fat). RESULTS: For men, DXA-BV estimations (84.3 +/- 15.1L) were no different than ADP (84.2 +/- 15.0L) or HW (84.2 +/- 14.8L) (p = 0.29). Percent Fat values for both 2C and 4C models were no different between DXA-BV (2C: 26.6 +/- 8.6, 4C: 28.6 +/- 8.0), ADP (2C: 26.2 +/- 7.2, 4C: 28.4 +/- 7.0), and HW (2C: 26.2 +/- 6.4, 4C: 28.4 +/- 6.6) (p > 0.4). However, significant differences were found between DXA-%Fat (31.0 +/- 7.3) and HW, ADP, and DXA-BV 2C and 4C models (p < 0.01). For women, significant differences (p < 0.01) were observed for DXA-BV estimations (68.1 +/- 12.1L) compared to ADP (67.5 +/- 11.9L) and HW (67.6 +/- 12.1L). Percent Fat values for both 2C and 4C models were also significantly different between DXA-BV (2C: 35.2 +/- 5.6, 4C: 37.0 +/- 5.0), ADP (2C: 31.7 +/- 3.8, 4C: 34.9 +/- 3.9), and HW (2C: 31.7 +/- 3.8, 4C: 34.9 +/- 4.1) (p < 0.01). Significant differences were also found between DXA-%Fat (38.2 +/- 4.5) and HW, ADP, and DXA-BV 2C and 4C models (p < 0.01). CONCLUSIONS: For both men and women the DXA-BV 2C and 4C %Fat values improved on the DXA-%Fat values. However, for women the DXA-BV 2C and 4C %Fat values were significantly higher than ADP and HW 2C and 4C models. For men DXA-BV appears to be an alternative method for calculating BV compared to ADP and HW-based 2C and 4C models. For women, the %fat values from DXA-BV-based 2C and 4C models is an improvement over standard DXA-%Fat values, but may not be suitable as an alternative for measuring BV compared to ADP and HW.