Interrelation between Plasma Concentrations of Vitamins C and E along the Trajectory of Ageing in Consideration of Lifestyle and Body Composition: A Longitudinal Study over Two Decades

Abstract

Although the interrelation between vitamins C and E has been demonstrated on an experimental level, its impact on biomarkers in community-dwelling subjects along the trajectory of ageing has not yet been shown. The present longitudinal study investigates the determinants and interrelation of vitamins C and E plasma concentrations in 399 subjects aged ≥60 years with a median follow-up time of 12 years. Linear mixed-effects models were used to analyze the influence of age, sex, body composition, dietary intake, physical activity, smoking and supplement/drug use on plasma vitamin C, plasma α-tocopherol and α-tocopherol/total cholesterol ratio. At baseline, median plasma concentrations of vitamin C and α-tocopherol were 74 and 35 µmol/L. Absolute fat-free mass, physical activity, use of supplements, and plasma α-tocopherol were main determinants of plasma vitamin C in the course of ageing. For the α-tocopherol/total cholesterol ratio, age, use of supplements, use of lipid-modifying drugs, and plasma vitamin C were main determinants. The results reveal a stable positive interrelation between plasma concentrations of vitamins C and E along the trajectory of ageing independent of the other identified determinants. The possible regulatory mechanisms that could explain this robust positive interrelation remain to be elucidated.

Keywords: ascorbic acid; body composition; diet; elderly; longitudinal changes; physical activity; tocopherol; vitamin C; vitamin E.

Figure 1

Flowchart. This flowchart illustrates the numbers and reasons for excluding subjects (indicated by IDs) and/or records. In total, 587 subjects participated in the GISELA study from 1994 to 2014. After applying the illustrated exclusion criteria, the final sample size for the present investigation was 399 subjects who had complete data records on relevant parameters on at least three visits during the study period.

Figure 2

Longitudinal associations of vitamin C plasma concentrations with age, vitamin C intake, absolute fat-free mass and physical activity index separated by sex if appropriate. This figure illustrates the changes in vitamin C plasma concentrations with increasing age (A), vitamin C intake (B), fat-free mass (C) and physical activity index (D) after controlling for the cofactors included in the linear mixed-effects model 3 (n = 399). The thick lines represent the estimated means and the respective colored areas reflect the 95% confidence intervals. The associations are illustrated in magenta color for females (n = 278) and blue color for males (n = 121). For physical activity index, no sex-specific associations are displayed because no such interaction term was investigated.

Figure 3

Longitudinal associations of the log α-tocopherol/total cholesterol ratio with age, intake of α-tocopherol equivalents, absolute fat mass and vitamin C plasma concentrations separated by sex. This figure illustrates the changes in the log α-tocopherol/total cholesterol ratio with increasing age (A), intake of α-tocopherol equivalents (B), fat mass (C) and vitamin C plasma concentrations (D) after controlling for the cofactors included in the linear mixed-effects model 3 (n = 399). The thick lines represent the estimated means and the respective colored areas reflect the 95% confidence intervals. The associations are illustrated in magenta color for females (n = 278) and blue color for males (n = 121). Abbreviation: TOC/chol ratio = α-tocopherol/total cholesterol ratio.