Glutathione peroxidase, selenoprotein P and selenium in serum of elderly subjects in relation to other biomarkers of nutritional status and food intake

The relation of the intake of selenium and different food groups to serum levels of selenium, glutathione peroxidase and selenoprotein P, and urinary selenium was studied in 50- to 69-year-old subjects (101 men, 105 women). Blood was sampled sir times during the course of I pear, and during the same time the subjects performed six 3-day weighed dietary records. The concentration of glutathione peroxidase in serum was higher among men (4.3 (0.74) mg/L) (mean (SD)) than among women (4.0 (0.76) mg/L, P < 0.03), whereas men had lower serum selenium (1.10 (0.17) mu mol/L) vs. 1.17 (0.19) mu mol/L, P < 0.01). The intake of selenium among men was calculated to be 36 (18-54) mu g/day and among women 29 (13-48) mu g/day (geometric means (90% central range)), but there was no significant gender difference when the selenium intake was corrected for energy intake. No difference between men and women was observed for serum selenoprotein P (1.47 (0.25) a.u. versus 1.47 (0.24) a.u. (mean(SD)) or urinary selenium excretion (0.31 (0.09) mu mol/d vs. 0.27 (0.08) mu mol/d). Serum selenium was significantly associated with selenoprotein P in both sexes but with glutathione peroxidase only in men. Serum selenium had a marked association with urinary selenium excretion. Selenium intake was significantly associated with serum selenium in men, and with selenoprotein P and urinary selenium in women. Among II major food groups, the intakes of fish and milk products ,cere significantly associated with biomarkers of selenium status among women. Calculation of the association between biomarkers of selenium status and variables on intake and serum content of nutrients using multiple regression analysis resulted in most cases in low explanatory power. The variables most consistently related to more than one variable of selenium status were serum levels of retinol, alpha-tocopherol and eicosapentaenoic acid, and intake of protein. The study shows that individual biomarkers of selenium status are associated differently to variables on dietary intake and nutritional status. (C) Elsevier Science Inc. 1997.