Women show higher vitamin C plasma concentrations than men, but the reasons for this observation still require elucidation. The objective of the present study was to investigate whether sex differences in vitamin C plasma concentrations are present in elderly subjects and whether these differences are due to sex-specific lifestyles, total antioxidant status (TAOS) and/or body composition. Fasting plasma concentrations of vitamin C were assessed by photometric detection in a cross-sectional study of 181 women and eighty-nine men aged 62–92 years. Body composition was determined by bioelectrical impedance analysis. Vitamin C intake was assessed with a 3 d estimated dietary record. Stepwise multiple regression analyses were performed to investigate whether sex is an independent predictor of vitamin C plasma concentrations by controlling for age, vitamin C intake, lifestyle factors, TAOS and body composition. Women showed higher vitamin C plasma concentrations than men (76 v. 62 μmol/l, P< 0·0001). In the multiple regression analysis, male sex was a negative predictor of vitamin C plasma concentrations (β = − 0·214), as long as absolute fat-free mass (FFM) was not considered as a confounder. When absolute FFM was included, sex was no longer a predictor of vitamin C plasma concentrations, whereas absolute FFM (β = − 0·216), physical activity level (β = 0·165), intake of vitamin C supplements (β = 0·164), age (β = 0·147) and smoking (β = − 0·125) affected vitamin C plasma concentrations. The results indicate that a higher absolute FFM, and thus a higher distribution volume of vitamin C, contributes to lower vitamin C plasma concentrations in men than women.

This study investigates age-dependent changes in resting metabolic rate (RMR) considering changes in body composition and fat distribution within the longitudinal study on nutrition and health status in an aging population in Giessen (GISELA), Germany, using three different approaches. In approach 1 cross-sectional data from 358 female and 155 male participants of the GISELA study were evaluated (mean age of 67.4 ± 5.9 and 66.9 ± 5.2 y, respectively). In approach 2 longitudinal data of 107 female and 55 male subjects who participated over a follow up period of 10 years were analysed. In approach 3 all data obtained at a total of 3033 visits from 363 women and 153 men between 1994 and 2006 were evaluated. The mean duration of follow-up was eight years. RMR was assessed by indirect calorimetry. Approach 1: RMR correlates significantly negatively with age in women and men. Considering fat free mass, fat mass, and WHR, age proved to be a significant predictor of RMR in both sexes in multiple regression analysis; RMR falls by 11.2 kJ/d and 34.1 kJ/d per year in females and males, respectively. Approach 2: In males but not in females RMR decreases significantly in the course of the follow up. After ten years measured RMR is significantly lower than expected RMR predicted on the basis of body composition and fat distribution in females and males. Deviations correspond to a decline in RMR by 11.4 and 27.5 kJ/d per year independently of changes in body composition and fat distribution. Approach 3: Results of the mixed linear model show that RMR decreases in the course of aging in both women and men; after considering changes in body composition and fat distribution respective decreases were 8.7 and 30.7 kJ/d per year. These results indicate that the decline in RMR with advancing age cannot be totally due to changes in body composition.