Changes in thoracic gas volume with air-displacement plethysmography after a weight loss program in overweight and obese women

OBJECTIVE: This study was designed to compare measured and predicted thoracic gas volume (V (TG)) after weight loss and to analyze the effect of body composition confounders such as waist circumference (WC) on measured V (TG) changes. DESIGN: Prospective intervention study. SETTING: Outpatient University Laboratory, Lisbon, Portugal. SUBJECTS: Eighty-five overweight and obese women (body mass index = 30.0+/-3.5 kg/m(2); age = 39.0+/-5.7 years) participating in a 16-month university-based weight control program designed to increase physical activity and improve diet. METHODS: Body weight (Wb), body volume (Vb), body density (Db), fat mass (FM), percent fat mass (%FM) and fat-free mass (FFM) were assessed by air-displacement plethysmography (ADP) at baseline and at post-intervention (16 months). The ADP assessment included a protocol to measure V (TG) and a software-based predicted V (TG). Dual-energy X-ray absorptiometry (DXA) (Hologic QDR 1500) was also used to estimate FM, %FM and FFM. Maximal oxygen uptake (VO(2) max) was assessed with a modified Balke cardiopulmonary exercise testing protocol with a breath-by-breath gas analysis. RESULTS: Significant differences between the baseline and post-weight loss intervention were observed for body weight and composition (Vb, Db, %FM, FM and FFM), and measures of V (TG) (measured: Delta=0.2 l, P<0.001; predicted: Delta=0.01 l, P<0.010) variables. Measured V (TG) change was negatively associated with the change in the WC (P=0.008), controlling for VO(2) max and age (P=0.007, P=0.511 and P=0.331). Linear regression analysis results indicated that %FM and FM using the measured and predicted V (TG) explained 72 and 76%, and 86 and 90% respectively, of the variance in %FM and FM changes using dual-energy x-ray absorptiometry. CONCLUSIONS: After weight loss, measured V (TG) increased significantly, which was partially attributed to changes is an indicator of body fat distribution such as WC. Consequently, measured and predicted V (TG) should not be used interchangeably when tracking changes in body composition. The mechanisms relating the reduction of an upper body fat distribution with an increase measured V (TG) are worthy of future investigation.