Prader-Labhart-Willi syndrome (PWS)-characterized by severe obesity, short stature, hypogonadism, and muscle hypotonia-appears to be an interesting model for body-composition abnormalities. Twenty-seven PWS patients (15 males and 12 females) aged 6-22 y underwent total-body analysis by dual-energy X-ray photon absorptiometry (DXA). For each PWS patient two age- and sex-matched control subjects were studied: one obese subject with a relative body weight (RBW > 120%) and body mass index (BMI) similar to that of the patient and one normal-weight subject (RBW < 120%). Percentage body fat was significantly greater in PWS patients than in obese subjects (47.4 +/- 7.2% compared with 41.9 +/- 9.9%, P < 0.0001) and the same difference was evident for arms and legs but not for the trunk. Lean mass was significantly lower in PWS patients (26.4 +/- 8.2 kg) than in normal-weight subjects (32.9 +/- 10.2 kg) and even more so than in obese subjects (40.3 +/- 13.2 kg) (P < 0.0001). The most affected regions were limbs; thus, the ratio of lean mass in the trunk to that in the limbs was significantly higher in PWS patients (1.19 +/- 0.15) than in obese (1.07 +/- 0.13) and normal-weight (1.07 +/- 0.09) subjects (P < 0.002). The ratio of fat mass to lean mass was significantly higher in PWS patients than in obese subjects (0.90 +/- 0.32 and 0.74 +/- 0.27, P < 0.05). Bone mineral content (BMC) was significantly lower in PWS patients (1503 +/- 46 g) than in normal-weight (1876 +/- 677 g) and obese (2322 +/- 773 g) subjects (P < 0.0001); this difference was most pronounced in the limb region. Bone mineral density (BMD) in PWS patients (0.993 +/- 0.116 g/cm2) did not differ significantly from that of normal-weight subjects (1.033 +/- 0.147 g/cm2) but was significantly lower than that of obese subjects (1.154 +/- 0.139 g/cm2). The influence of age on body composition was assessed by comparing two age subgroups (< 12 y, n = 10; and > or = 12 y, n = 17). The older PWS patients had higher adiposity, lower BMC, and dramatically lower BMD. Also, the lean mass deficit increased with age so that the ratio of fat mass to lean mass was close to 1. In conclusion, PWS patients showed a peculiar body composition, to some extent similar to that found in subjects deficient in growth hormone or even to sedentary and elderly people. These results suggest the importance of an accurate analysis of body composition in PWS patients.

Fat and muscle areas can be calculated from equations on the basis of upper arm circumference (C) and triceps skinfold thickness (TS). These equations assume a circular limb and muscle compartment and a symmetrically distributed fat rim: total upper arm area (TUA) = C2/(4 pi), upper arm muscle area (UMA) = [C - (TS x pi)2]/(4 pi), and upper arm fat area (UFA) = TUA - UMA. This traditional method underestimates the degree of adiposity. We propose that the unrolled fat rim is a rectangle whose length = C and width = TS/2. The following new indexes are based on this assumption: upper arm fat area estimate (UFE) = C x (TS/2), and upper arm muscle area estimate (UME) = TUA - UFE. To validate these equations, areas were measured with magnetic resonance imaging (MRI) in 28 children aged 9-15 y (17 control subjects and 11 obese subjects). Correlations between MRI and UFA and MRI and UFE were similar (r = 0.96 for both correlations in the control group and r = 0.84 and 0.82, respectively, in the obese group), but the areas assessed by MRI (13.8 cm2) were closer to UFE (12.4 cm2) than to UFA (11.2 cm2) in the control group as well as in the obese group (MRI = 48.7 cm2, UFE = 46.6 cm2, and UFA = 38.5 cm2). The limits of agreement between MRI and anthropometry were 5.7 +/- 5.8 cm2 for UFA and 0.6 +/- 5.0 cm2 for UFE, showing that UFA is not acceptable in most cases, whereas UFE measurements are close to MRI measurements. In conclusion, UFE and UME are simple and accurate indexes to assess body composition. French reference values are available from 1 mo to 17 y of age.

Excess fat and fat-free mass have been extensively described in obese children, whereas few data about bone mineral content (BMC) variations are available in children. Dual-energy X-ray absorptiometry (DXA) allows a direct and accurate measurement of three body compartments (fat, lean, and BMC), subdivided into three regions (arms, trunk, and legs). The aim of our study was to evaluate the influence of body compartments on total BMC (TBMC) and regional BMC (RBMC) in obese and normal-weight subjects. Sixty-five obese and 50 normal-weight children and adolescents (age range: 5-18 y relative body weight: 160 +/- 23% and 101 +/- 12%, respectively), matched for sex and pubertal stage underwent a DXA total-body analysis. Obese subjects had significantly greater fat and lean compartments than normal-weight subjects (P < 0.0001). TBMC was larger in obese children (1930 +/- 670 g compared with 1480 +/- 490 g, P < 0.0001) as was RBMC (arms: 182 +/- 81 g compared with 151 +/- 65 g; trunk: 560 +/- 223 g compared with 433 +/- 169 g; legs: 788 +/- 341 g compared with 539 +/- 231 g, P < 0.0001). We found lean mass to be the best correlate with TBMC (r = 0.91 in obese and 0.94 in normal-weight children). Multiple-regression analysis confirmed lean mass as one of the major determinants of TBMC and RBMC in children. However, differences in TBMC and RBMC were no longer present after correction for age, sex, and body-composition variables. There were no differences in TBMC and RBMC between obese and normal-weight children after correction for the confounding variables age and sex.