The importance of malnutrition as a risk factor in osteoporosis is emphasized by the evidence that patients with fractures of the proximal femur are often undernourished. In this study, nutritional status, bone mineral mass and its association with body composition were investigated in underweight and normal weight elderly subjects. Moreover the hypothesis that malnutrition in elderly is associated with a higher risk of osteoporosis was tested. The participants were 111 elderly subjects divided into two groups according to body mass index (BMI): 51 patients were underweight (BMI < 22 kg/m2) while in 60 subjects BMI ranged from 22 to 30 kg/m2. In all patients anthropometric parameters and blood indices of malnutrition and of bone turnover were measured. Fat-free soft mass (FFSM), fat mass (FM), bone mineral content (BMC) and bone mineral density (BMD) 'total body' and at the hip were obtained by dual-energy X-ray densitometry. Dietary intake was evaluated with the diet history method, while resting energy expenditure (REE) was measured by indirect calorimetry. Underweight subjects had other signs of malnutrition, such as low visceral proteins, sarcopenia, and an inadequate energy intake. Moreover they showed a significant reduction of BMC and BMD compared with normal subjects. In men with BMI <22 kg/m2, T-score was below -2.5 (-3 at femoral neck and -2.7 at total hip) while men in the control group had normal bone mineral parameters. T-score at different sites was lower in underweight women than in underweight men, always showing values under -3.5, with clear osteoporosis and a high fracture risk. In healthy women the T-score values indicated the presence of mild osteoporosis. In underweight subjects, low levels of albumin (< 35 g/l) were associated with higher femoral bone loss. Using a partial correlation model, BMC, adjusted for age, bone area, knee height and albumin showed a significant association with FM in women (r = 0.48; p < 0.01) and with FFSM in men (r = 0.48; p < 0.05). Albumin, when adjusted for other variables, was significantly correlated (r = 0.52; p < 0.05) with femoral neck BMC only in women. In conclusion, the underweight state in the elderly is associated with malnutrition and osteoporosis; other factors occurring in malnutrition, besides body composition changes, such as protein deficiency, could be involved in the association between underweight and osteoporosis. Moreover bone mineral status seems to be related to fat-free soft mass tissue in men while in women it is much more closely associated with total body fat.

OBJECTIVE: To investigate the changes of visceral fat, as compared with total and subcutaneous adipose tissue (AT) in obese patients operated with laparascopic adjustable silicone gastric banding (LAP-BAND). SUBJECTS: Six premenopausal morbid obese (body mass index range: 41.4 ± 44.2 kgam 2 ) women, aged 38 ± 42 y, operated with LAP-BAND, evaluated before, 8 weeks after, and 24 weeks after surgery. MEASUREMENTS: Fat distribution was analysed by total body multi-slices MRI. Total AT, gluteo-femoral subcutaneous AT, abdominal subcutaneous AT, and abdominal visceral AT volumes were measured. FM was calculated from MRI-determined total AT volume and AT density. RESULTS: A weight loss of 9.9AE AE3.8 kg was observed in the ®rst 8 weeks after LAP-BAND (0 ± 8 weeks), and a further weight loss of 7.1AE AE4.9 kg in the subsequent 16 weeks (8 ± 24 weeks). Total AT showed a statistically signi®cant reduction of 6.2AE AE4.0 l in 0 ± 8 weeks and a further signi®cant reduction of 7.7AE AE3.9 l in 8 ± 24 weeks (P`0.01 from baseline). A similar trend was observed for both abdominal and gluteo-femoral subcutaneous AT. Visceral AT showed a statistically signi®cant reduction of 1.0AE AE0.9 l in the 0 ± 8 weeks (P`0.05) and a further non-signi®cant reduction of 0.6AE AE0.7 l in 8 ± 24 weeks (P`0.05 from baseline). In 0 ± 8 weeks, the relative reduction of visceral AT was higher than the relative reduction of both total AT and gluteo-femoral subcutaneous AT. A highly signi®cant correlation was observed between the reduction of total AT and the reduction of both abdominal and gluteo-femoral subcutaneous AT. By contrast, in 0 ± 8 weeks, the reduction of total AT and the reduction of visceral AT were not correlated. In a subsequent analysis, both observations collected in the ®rst 8 weeks after LAP-BAND and observations collected in the last 16 weeks are simultaneously considered, leading to a total of 12 time periods (two time periods for each individual patient). In order to identify factors associated with preferential visceral fat reduction, we calculated for each of the 12 time periods the difference between the percentage changes of visceral AT and the percentage changes of total AT. The relationship between this difference and several other variables were investigated by simple correlation analysis. The only variables found to be associated were the initial visceral AT volume, the absolute level of weight loss (kg) per week of observation, and the relative level of weight loss (%) per week of observation. CONCLUSION: In the phase of rapid weight loss following LAP-BAND, a preferential mobilization of visceral fat, as compared with total and subcutaneous AT, can occur. However, this preferential visceral fat reduction occurs only in those patients presenting higher levels of visceral fat deposition at baseline and higher levels of weight loss.

Background: In the literature there are several body impedance analysis (BIA) prediction equations generally determined in younger populations and their accuracy in the elderly has not been adequately confirmed. Objective: We verified the reliability of the BIA method in a body composition study in the elderly. Methods: To assess the accuracy of bioelectrical impedance analysis we compared this method with dual photon absorptiometry (DPA), assumed as a gold standard; body composition was predicted by seven BIA prediction equations in 24 healthy elderly individuals. Results: The best equation in fat-free mass (FFM) estimation is the RJL System formula (published by Segal in 1988); nevertheless, the wide range of the error in FFM estimating may limit its clinical application. The FFM hydration variability seems to be the principal variable which explains the error in FFM estimation by BIA prediction equations. Conclusion: These findings indicate that the use of BIA equations is not interchangeable, when FFM is calculated in an elderly population, and more validation studies are necessary in this age group to evaluate the clinical application of this method.

Underweight elderly subjects show a hypometabolism due to a reduction of both FFM quantity and its metabolic activity. Functional status in ADLs comes out as an important predictor of REE independently from FFM. The limited physical activity might be the underlying determinant of this hypometabolism, but further investigations are necessary to confirm this issue.