Abstract
Deficiency of essential trace element, Se, has been implicated in adverse birth outcomes and in child linear growth because of its important role in redox biology and associated antioxidant effects. We used data from a randomised controlled trial conducted among a cohort of pregnant and lactating women in Dhaka, Bangladesh to examine associations between Se biomarkers in whole blood (WBSe), serum and selenoprotein P (SEPP1) in maternal delivery and venous cord (VC) blood. Associations between Se biomarkers, birth weight and infant growth outcomes (age-adjusted length, weight, head circumference and weight-for-length z-scores) at birth, 1 and 2 years of age were examined using regression analyses. WB and serum Se were negatively associated with birth weight (adjusted β, 95 % CI, WBSe delivery: −26·6 (–44·3, −8·9); WBSe VC: −19·6 (–33·0, −6·1)); however, delivery SEPP1 levels (adjusted β: −37·5 (–73·0, −2·0)) and VC blood (adjusted β: 82·3 (30·0, 134·7)) showed inconsistent and opposite associations with birth weight. Positive associations for SEPP1 VC suggest preferential transfer from mother to fetus. We found small associations between infant growth and WBSe VC (length-for-age z-score β, 95 % CI, at birth: −0·05 (–0·1, −0·01)); 12 months (β: −0·05 (–0·08, −0·007)). Weight-for-age z-score also showed weak negative associations with delivery WBSe (at birth: −0·07 (–0·1, −0·02); 12 -months: −0·05 (–0·1, −0·005)) and in WBSe VC (at birth: −0·05 (–0·08, −0·02); 12 months: −0·05 (–0·09, −0·004)). Given the fine balance between essential nutritional and toxic properties of Se, it is possible that WB and serum Se may negatively impact growth outcomes, both in utero and postpartum.