Association between serum selenium concentrations and learning disability in a nationally representative sample of US children

Background Oxidative stress has been implicated in the pathogenesis of neurodevelopmental disorders. As an anti-oxidative agent, selenium plays an important role in human health. However, the relationship between selenium status and learning disability (LD), a common neurodevelopmental disorder, is unknown. Objective To examine the association between serum selenium concentrations and learning disability. Design Nationwide, population-based, cross-sectional study. Participants/setting Children aged 4-11 years who have available data on serum selenium concentrations and LD (N = 1,076) from the U.S. National Health and Nutrition Examination Survey 1999-2000. Exposure Serum selenium levels were measured using atomic absorption spectrometry. Main outcome measures Diagnosis of LD was reported by the children's parents. Statistical analyses performed Logistic regression models with survey weights were conducted adjusting for age, race/ethnicity, family income, total energy intake, body mass index, and serum cotinine levels. Results In this study, 8.2% (95% confidence interval [CI] 5.2%-11.2%) of children had a diagnosis of LD. Serum selenium concentration was lower among children with LD than those without LD (geometric mean +/- standard error, 107.7 +/- 2.7 ng/mL vs. 112.8 +/- 1.0 ng/mL, P for difference = 0.08). The adjusted odds ratio (OR) of LD comparing the highest with lowest tertile of serum selenium concentrations was 0.39 (95% CI 0.19-0.82). Each 10 ng/mL increment in serum selenium concentrations was associated with 31% (OR 0.69, 95% CI 0.51-0.93) lower odds of LD. Conclusions Higher serum selenium concentration was associated with a lower risk of LD in U.S. children. The causal relationship between selenium and LD and the underlying mechanisms warrant further investigation.