Glutathione peroxidase is not a functional marker of selenium status in the neonatal period

Background: The antioxidant enzyme glutathione peroxidase is a selenoprotein that, in adults with low selenium intakes, has a strong linear relationship with blood selenium and hence is used as a functional indicator of selenium status. Our aim was to evaluate glutathione peroxidase as a functional marker of selenium status in preterm infants. Methods: Erythrocyte glutathione peroxidase activity and plasma and erythrocyte selenium were measured between days 1-5 and then weekly until discharge in 63 preterm infants with mean +/- standard error birth weight and gestation of 1572 +/- 60g and 30.7 +/- 0.3 weeks. A healthy reference group of term infants (n = 46) was assessed at day 5 and at 6 weeks. Results: In preterm infants, over the first 3 months, the association of glutathione peroxidase activity with erythrocyte selenium was weak and inconsistent and nonexistent with selenium intake or plasma selenium. No correlations between any of these indicators were evident for term infants. In preterm infants, plasma and erythrocyte selenium declined over the first 6 weeks (p < 0.01), while glutathione peroxidase activity increased (p < 0.05). In term infants, plasma selenium increased (p < 0.001), but there was no change in erythrocyte selenium or glutathione peroxidase activity. For preterm infants, glutathione peroxidase activities at weeks 4 and 6 were associated with maximum inspired oxygen concentration, ventilator pressure, and days of ventialtion. Conclusions: This data is consistent with animal and in vitro evidence that glutathione peroxidase may be confounded by oxygen. We conclude that erythrocyte glutathione peroxidase activity is not a reliable functional marker of preterm selenium status in the neonatal period.