Article ID Journal Published Year Pages File Type
2737524 Radiography 2009 8 Pages PDF
Abstract

BackgroundBecause of its high accuracy, dual-energy X-ray absorptiometry (DXA) has become one of the most frequently used methods for estimating human body composition. One limiting factor concerning measuring obese people with the DXA technique is the size of the scanning area.ObjectiveTo explore the possibility of estimating whole body composition from half-body scans before and after weight reduction, and compare the results with densitometry measurements.DesignIntervention study of 15 obese adults (age 47.2 ± 13.4; BMI 35.9 ± 3.1) who were measured with full- and half-body DXA scans before and after a 7-week weight loss program. On both occasions, body composition was also assessed with air-displacement plethysmography (ADP).ResultsThe mean weight loss at follow-up was 14.9 ± 4.1 kg (5.0 kg/m2), corresponding to a 14% decrease in body weight. When comparing the results from full- and half-body DXA, between 96% and 98% of the variance was explained. At baseline, %Body Fat (%BF) did not differ significantly between full and half-body measurements (0.6, −0.1−1.3), but the half-body method overestimated it by 1.0% (0.2–1.8) at follow-up. On the contrary, the difference between DXA and ADP in the assessment of %BF was both significant and of large magnitude (5.2; 2.4–8.0) at baseline, while non-significant and near zero (0.4; −1.3–2.2) at follow-up when the subjects had lost a significant amount of weight.ConclusionThe results obtained from half-body DXA scans can accurately predict whole body composition, as measured by full-body DXA, before and after significant weight reduction, in obese patients who barely fit into the scanning area. However, increasing discordance between DXA and ADP with increasing adiposity was seen, indicating that the measurements might not be as reliable on extreme obese subjects as on normal and overweight ditto.

Related Topics
Health Sciences Medicine and Dentistry Radiology and Imaging
Authors
, , , , ,