Selenium status and its correlates in a British national diet and nutrition survey: people aged 65 years and over.

Plasma selenium (Se) concentrations were measured in 1134 British people aged 65 years and over, living in mainland Britain during 1994-1995. Eight hundred and eighty-three lived in the community ("free-living"), while the remainder lived in institutions (residential and nursing homes). The overall mean plasma Se concentrations was 0.90 micromol/l (inner 95% range 0.50-1.36 micromol/l). Free-living people had significantly higher values than counterparts living in institutions, and there was an overall inverse correlation with age. Plasma Se varied with season (lower from October to December than at other times of the year), while values were higher in southern Britain than in the North. Socio-economic associations also existed (lower plasma Se in people receiving state benefits or with poorer educational attainment), while smokers had lower values than non-smokers. Several indices of poor health status or of medicine use were correlated with lower plasma Se, which was also predicted by several biochemical or haematological indices of infection or inflammation. Plasma Se was strongly and directly correlated with plasma albumin, zinc, cholesterol, vitamin C, several carotenoids, alpha-tocopherol, pyridoxal phosphate, and with blood haemoglobin and several anthropometric indices. All relationships were independent of age and gender, and are consistent with the view that plasma Se levels are higher in fit and well-nourished elderly people and lower in those who are frail, poorly-nourished and unwell. Whole-blood glutathione peroxidase generally did not share these relationships, was only very weakly correlated with plasma Se, and appeared to be less useful as a status indicator in this population group.