Anthropometric indices and selenium status in British adults: the U.K. National Diet and Nutrition Survey.

Recent studies have raised concern over possible associations between high selenium (Se) status and excess adiposity, known to be linked to adverse cardiometabolic outcomes. Studies of Se status in relation to adiposity are scarce in the United Kingdom. This study examined cross-sectional associations of anthropometric indices with Se-status biomarkers in a nationally representative sample of 1045 (577 female, 468 male) British Caucasian adults ages 19-64 who participated in the 2000-2001 National Diet and Nutrition Survey. Median (first, third quartile) values for whole-blood glutathione peroxidase (GPx) activity and plasma and erythrocyte Se concentrations were 120.0 (103.0, 142.4) nmol mg Hb(-1) min(-1), 1.08 (0.98, 1.20) µmol/L, and 1.62 (1.38, 1.91) µmol/L, respectively. For males, values were 119.0 (100.0, 141.0) nmol mg Hb(-1) min(-1), 1.09 (0.99, 1.22) µmol/L, and 1.54 (1.34, 1.79) µmol/L, respectively; for females 121.0 (105.0, 145.0) nmol mg Hb(-1) min(-1), 1.07 (0.97, 1.18) µmol/L, and 1.71 (1.43, 1.99) µmol/L, respectively. Multivariate adjusted mean differences (95% CI) in whole-blood GPx between the highest (>30 kg/m(2)) and the lowest (<25 kg/m(2)) categories of body mass index and the highest (96.5-139.2 cm) and the lowest (52.2-78.1cm) quartiles of waist circumference (WC) were -7.9 (-13.2, -2.7) and -9.7 (-16.2, -3.2) nmol mg Hb(-1) min(-1), respectively. Difference (95% CI) in plasma Se between the third (87.5-96.4 cm) and the lowest quartiles of WC was -0.04 (-0.08, -0.03) µmol/L. Difference (95% CI) in red blood cell (RBC) Se between the highest (0.91-1.11) and the lowest (0.53-0.76) quartiles of waist-to-hip ratio (WHR) was 0.10 (0.00, 0.20) µmol/L. Similar results were observed in gender and menopausal-status subgroup analyses. The inverse association between plasma Se and WC and the positive association between RBC Se and WHR will need confirmation. The findings suggest associations between low whole-blood GPx activity and higher measures of general and central adiposity. Further experimental and randomized studies are needed to deduce the mechanisms and infer causality.