Selenium status and its determinants in very old adults: insights from the Newcastle 85+ Study

Abstract

There is a dearth of data on selenium status in very old adults. The aims of this study were to assess selenium status and its determinants in 85-year-olds living in the Northeast of England by measuring serum selenium and selenoprotein P (SELENOP) concentrations and glutathione peroxidase 3 (GPx3) activity. A secondary aim was to examine the interrelationships between each of the biomarkers. In total, 757 participants (463 women, 293 men) from The Newcastle 85+ Study were included. Biomarker concentrations were compared with selected cut-offs (serum selenium: suboptimal 70 µg/L and deficient 45 µg/L; SELENOP: suboptimal 4.5 mg/L and deficient 2.6 mg/L). Determinants were assessed using linear regressions and interrelationships were assessed using restricted cubic splines. Median (IQR) concentrations of serum selenium, SELENOP and of GPx3 activity were 53.6 (23.6) µg/L, 2.9 (1.9) mg/L and 142.1 (50.7) U/L, respectively. Eighty-two percent, and 83 % of participants had suboptimal serum selenium (< 70 µg/L) and SELENOP (< 4.5 mg/L), and 31 % and 40 % of participants had deficient serum selenium (< 45 µg/L) and SELENOP (< 2.6 mg/L), respectively. Protein intake was a significant determinant of selenium status. Additional determinants of serum selenium were gender, waist:hip ratio, self-rated health, and disease, whilst gender, BMI and physical activity were determinants of GPx3 activity. There was a linear association between serum selenium and SELENOP, and nonlinear associations between serum selenium and GPx3 activity and between SELENOP and GPx3 activity. These findings indicate that most participants had suboptimal selenium status to saturate circulating SELENOP.