RESUMEN
OBJECTIVES: Plasma selenium may not reflect selenium status in critically ill patients because it transiently decreases inversely with the magnitude of the systemic inflammatory response. The decision to supplement selenium should ideally be based on laboratory measurements that reliably reflect selenium status. We hypothesized that erythrocyte selenium, unlike plasma selenium, is not affected by the systemic inflammatory response in critically ill children. METHODS: In a prospective study of 109 critically ill children, plasma and erythrocyte selenium concentrations were evaluated on admission, and plasma selenoprotein P was evaluated on days 1, 2, and 3 of the ICU stay. The main outcome was the effect of systemic inflammation on the erythrocyte and plasma selenium concentrations. The magnitude of the systemic inflammatory response was measured using serum C-reactive protein (CRP) and procalcitonin levels. The covariates were age, sex, anthropometric nutritional status, diagnosis of severe sepsis/septic shock, and clinical severity on admission. Multiple linear regression and generalized estimating equations were used for statistical analysis. RESULTS: Erythrocyte selenium levels were not influenced by the magnitude of the inflammatory response or by the patient's clinical severity. Procalcitonin (ß coefficient=-0.99; 95%CI: -1.64; -0.34, p = 0.003) and clinical severity (ß coefficient= -11.13; 95%CI: -21.6; -0.63), p = 0.038) on admission were associated with decreased plasma selenium concentrations. Erythrocyte selenium was associated with selenoprotein P in the first three days of ICU stay (ß coefficient=0.32; 95%CI: 0.20; 0.44, p < 0.001). CONCLUSION: Unlike plasma selenium, erythrocyte selenium does not change in children with an acute systemic inflammatory response and is associated with selenoprotein P concentrations. Erythrocyte selenium is probably a more reliable marker than plasma selenium for evaluating the selenium status in critically ill children.