European Journal of Clinical Nutrition (2016) 70, 929–934 © 2016 Macmillan Publishers Limited, part of Springer Nature. All rights reserved 0954-3007/16 www.nature.com/ejcn ORIGINAL ARTICLE Association of serum selenium with thyroxin in severely iodine-deficient young children from the Amhara region of Ethiopia D Gashu1, BJ Stoecker2, A Adish3, GD Haki4, K Bougma5, FE Aboud6 and GS Marquis5 BACKGROUND/OBJECTIVES: Selenium (Se) as part of glutathione peroxidase and iodothyronine deiodinase enzymes influences thyroid metabolism. This study investigated the association of serum Se levels with thyroid metabolism of severely iodine-deficient young children from the Amhara region of Ethiopia. SUBJECTS/METHODS: In a cross-sectional study, Se, thyroid-stimulating hormone, total thyroxin, total triiodothyronine and thyroglobulin in serum of children (N = 628) 54–60 months of age from the Amhara region, Ethiopia, were analyzed. In addition, iodine in urine and household salt was analyzed, and the presence of goiter was assessed. RESULTS: The median serum Se concentration was 61.4 μg/l (10.7–290.9 μg/l). Selenium deficiency (serum Se o70 μg/l) was detected in 57.8% (N = 349) of the children. The median urinary iodine concentration (UIC) was 9.8 μg/l. The majority (86.6%, N = 449) of children had UIC below the recommended value (100 μg/l). In addition, 59.8% (N = 310) of children were severely iodine deficient (UICo20 μg/l). Only 12.7% of salt samples had iodine. Goiter was present in 44.6% (N = 280) of the children. Selenium- deficient children had higher serum thyroxin (T4) than children with normal serum Se concentration (Po0.001). CONCLUSIONS: Serum Se was negatively associated with T4 level in young children from the Amhara region of Ethiopia and may endanger the effectiveness of the salt iodization program. European Journal of Clinical Nutrition (2016) 70, 929–934; doi:10.1038/ejcn.2016.27; published online 16 March 2016 INTRODUCTION of Se deficiency on thyroid metabolism of severely iodine- The synthesis and normal metabolism of thyroid hormones deficient children are limited. The presence of Se deficiency in 10 require adequate intake of Se. This is because the Se-containing northwestern Ethiopia has been reported. In addition, iodine 11 iodothyronine deiodinase enzymes are involved in the conversion deficiency (ID) is a severe public health problem in the country. of the inactive thyroid hormone, thyroxin (T or 3, 5, 3′, Therefore, this study investigated the association of serum Se 4 fi 5′-tetraiodothyronine), into the biologically active thyroid hor- levels with thyroid metabolism of severely iodine-de cient young children from the Amhara region of Ethiopia. mone, triiodothyronine (T3 or 3, 5, 3′-triiodothyronine), or in to the inactive metabolite reverse triiodothyronine (rT3) and 3,3′-di-iodothyronine (T ). In addition, thioredoxin reductase, 2 MATERIALS AND METHODS which is another Se-dependent enzyme, protects the autoxidation of the thyroid gland by hydrogen peroxide (H2O2) produced Study participants during the synthesis of thyroid hormones.1,2 This study was part of a large randomized cluster trial ‘Effect of Iodized Salt The activity of selenoenzymes could be reduced in response to on Child Development in Amhara Region, Ethiopia’, which had mental inadequate Se nutrition and could in turn impair thyroid development of children 5 years of age and under as the primary outcome metabolism despite adequate iodine intake. Reports from (Clinicaltrials.gov #NCT01349634) in 60 randomly selected districts. In all selected districts, one village had been randomly selected and a census Uganda,3 Turkey4 and Iran5,6 attributed the persistence of goiter fi team enumerated all households with children of the required age and prevalence to Se de ciency despite of iodine replenishment. In children were further classified into three age groups (6–11, 18–22 and 54– addition, low serum Se level was associated with an increased 60 months) for the purpose of the larger study outcome (cognitive 7 concentration of T4. A large epidemiological study in populations development). Mothers of 32 children in the age group 54–60 months with adequate iodine nutrition from two counties in China refused to participate. Sample size in this part of the study was determined similar except their Se status reported a significantly higher risk by single population proportion with 50% prevalence of Se deficiency, 95% 8 confidence interval, 5% margin of error, design effect of 1.5 and 5% non- of thyroid diseases in the participants from the low-Se area. In an – fi response. Baseline data from all eligible children (N = 628), 54 60 months area of mild iodine de ciency, serum Se concentration was of age, from randomly selected 26 rural kebeles (smallest administrative negatively associated with thyroid volume. In that same study, low division) of six administrative zones (West Gojjam, East Gojjam, South serum Se concentration was reported to be a significant risk Gonder, North Wollo, South Wollo and Wagehmera) of the Amhara region factor for thyroid enlargement.9 Human studies on the association were included (Figure 1). 1Center for Food Science and Nutrition, Addis Ababa University, Addis Ababa, Ethiopia; 2Department of Nutritional Sciences, Oklahoma State University, Stillwater, OK, USA; 3Micronutrient Initiative Africa, Addis Ababa, Ethiopia; 4Department of Food Science and Technology, University of Botswana, Botswana College of Agriculture, Gaborone, Botswana; 5School of Dietetics and Human Nutrition, McGill University, Sainte Anne-de-Bellevue, QC, Canada and 6Department of Psychology, McGill University, Montreal, QC, Canada. Correspondence: Dr D Gashu, Center for Food Science and Nutrition, Addis Ababa University, PO Box 1176, Addis Ababa 1176, Ethiopia. E-Mail: [email protected] Received 2 August 2015; revised 29 January 2016; accepted 12 February 2016; published online 16 March 2016 Selenium predicts thyroxin in young children D Gashu et al 930 Figure 1. Study location. Blood collection and analysis mineral-free gloves and pipette tips were used during analysis of the 12 Blood was collected by experienced phlebotomists from the pediatric serum samples. Selenium deficiency was defined as serum Se o70 μg/l. department of the regional referral hospital (Felege Hiwote Hospital at Serum thyroid-stimulating hormone (TSH), T4,T3 and thyroglobulin (Tg) Bahir Dar) after refresher training. Blood was drawn by disposable butterfly were analyzed by electrochemilumenescenseimmunoassay using an vacutainer needles (BD Vacutainer butterfly needles, 23G, 3/4″, Franklin Elecsys2010 clinical analyzer (Cobas e411, Roche Diagnostics GmbH, Lakes, NJ, USA) by venipuncture using standard safety measures. Blood Mannheim, Germany) at the EPHI. The instrument for analyzing was collected into trace metal-free vacutainer blood tubes (BD Vacutainer, thyroid function biomarkers was calibrated using elecsys reagents every royal blue, trace element tube, 13 mm × 100 mm). The collected blood was time a fresh reagent was used as recommended by the manufacturer. In allowed to clot at an ambient temperature and centrifuged in the field addition, PrciControl Universal 1 and 2 reagents for T3,T4, TSH and Tg within an hour. The serum was separated and transferred into trace were used every 24 h or every time a fresh reagent was used for quality mineral-free vials using disposable plastic pipettes. The samples were control purpose. During the analysis of thyroid biomarkers, the inter- and transported to Bahir Dar (550 km from the capital Addis Ababa) in an intra-assay CV, respectively, for the serum samples (N = 20) was calculated icebox and kept at − 20 °C until they could be transferred to Addis Ababa’s to be 5.9 and 6.53% for T4, 7.6 and 9.5% for T3, 4.9 and 5.1% for TSH, Ethiopian Public Health Institute (EPHI) on dry ice, for storage at − 80 °C and 7.6 and 8.7% for Tg. Furthermore, the laboratory was participating in external quality assurance program ‘one world accuracy’.The until analysis of the thyroid metabolism markers. Duplicate vials of each normal reference range as recommended by the manufacturer was frozen sample were shipped on dry ice to the Department of Nutritional 0.27–4.2 mU/l for TSH, 12.9–181.4 nmol/l for T , 1.2–3.0 nmol/l for Sciences at Oklahoma State University, USA, for Se analysis. 4 T and 1.4–78.0 μg/l for Tg. Serum Se was analyzed by inductively coupled plasma mass spectro- 3 meter (PerkinElmer, ELAN9000, Norwalk, CT, USA). Serum samples were diluted in 0.1% trace metal grade HNO3 (Fisher Scientific, Fair Lawn, NJ, Household salt collection and analysis USA). A series of calibration standards were prepared by diluting 1000 p.p. Mothers or care givers were asked to give a small sample of salt for iodine m. pure Se stock solution (PerkinElmer) in 0.1% HNO3 and 0.5% Triton determination. The collected salt samples were packed in plastic bags X-100 (Sigma Aldrich, St Louis, MO, USA) solution. Working standards were and labeled with the identification code of the respective child. In the prepared daily. All samples and standards were spiked with 10 μg/l gallium field, small portions of the collected salt samples were taken for (PerkinElmer) as an internal standard. The calibration curves were fairly qualitative iodine analysis using a rapid test kit containing iodate linear (r = 0.99). A trace mineral reference standard of freeze-dried human reagents (MBI Kits International, Chennai, Tamil Nadu, India). Later, serum (Utak Laboratories, Inc., Valencia, CA, USA) containing Se was iodine-positive salt samples were analyzed quantitatively for iodine analyzed in each analytical batch to verify method performance. Reported concentration at EPHI using a standard iodometric titration method as results from the analysis of the reference standard were in the acceptable described in WHO/UNICEF/ICCIDD.13 For quality control purpose, food- range (105.13–112.66 μg/l) and were comparable with the verified value grade salt sample of ⩽ 2 mm grain size was iodized in the laboratory with (111 μg/l) set by the supplier.