Selenium in drinking water and cereal grains, and biomarkers of Se status in urine and fingernails of the Main Ethiopian Rift Valley population.

Selenium (Se) plays an important role in human health, yet Se overexposure or deficiency can lead to deleterious health effects. This study aims to determine the concentration of Se in drinking water and staple cereal grain (maize, wheat, and teff) samples from the Main Ethiopian Rift (MER) Valley, and correspondingly, assesses Se biomarkers and their status as measured in the urine and fingernails of 230 individuals living in 25 MER communities.The concentration of Se in drinking water and cereal grain (maize, wheat, and teff) samples, and urine and fingernail samples were measured using Inductively Coupled Plasma Mass Spectrometry (ICP-MS). Demographic, anthropometric, and elemental concentrations were described by their quartiles and mean ± standard deviations. The 5th and 95th percentiles were used to describe the concentrations Se biomarkers ranges. The Se biomarker distributions in different study communities were further characterized according to Se levels found in drinking water, sex, and age using ANOVA, and multivariate regression. We conducted a correlation analysis (with Pearson correlation coefficient) and fitted a regression to evaluate the associations between these variables.The mean concentration of Se in the drinking water samples was 0.66 (range: 0.015-2.64 µg/L; n = 25), and all samples were below the threshold value of 10 μg/L for Se in drinking water set by the World Health Organiation (WHO). In Ethiopia, most rural communities rely on locally produced cereal grains. We found mean Se concentrations (µg/kg) of 357 ± 190 (n = 14), 289 ± 123 (n = 14), and 145 ± 100 (n = 14) in wheat, teff, and maize, respectively. Furthermore, Se concentrations in drinking water showed no significant correlation with biomarker measures, indicating that the primary source of dietary Se is likely from local foods including staple grains. The mean±SD (5th-95th percentiles) of Se concentrations in fingernails and urine among study subjects were 1022 ± 320 (624-1551 µg/kg), and 38 ± 30 (1.9-100 µg/L), respectively.A sizeable share of study participants (31%) fell below the lower limits of what is considered the currently accepted Se range of 20-90 µg/L in urine, though relatively few (only 4%) had similarly low fingernail levels. On the other hand, none of the samples reached Se toxicity levels, and the biomarker levels in this study are comparable to results from other studies that find adequate Se. Our results show that Se toxicity or deficiency is unlikely in the study population.

Duke Scholars

Author Marc A. Jeuland Sanford School of Public Policy