Abstract: The aim of this study was to determine whether iron supplementation in iron-deficient adolescent girls would improve thyroid function. A double-blind clinical trial was performed in a region in southern I.R. Iran. A total of 103 iron deficient participants were chosen. In all, 94 participants successfully completed this study. Participants were randomly assigned to one of two groups and treated with a 300 mg ferrous sulfate 5 times/week (n = 47) and placebo 5 times/week (n = 47) for 12 weeks. Blood samples were collected and assayed for hemoglobin, hematocrit, serum ferritin, iron, total iron binding capacity (TIBC), Thyroid stimulating hormone (TSH), total thyroxine (TT4), total triiodothyronine (TT3), free thyroid hormones (FT4 and FT3), triiodothyronine resin uptake (T3RU), reverse triiodothyronine (rT3), selenium and albumin concentrations. Statistical analysis was performed with parametric and non-parametric methods as appropriate. Data analysis revealed a significant increase in TT4, TT3, T3RU and a significant decrease in rT3 concentration in comparison to initial values in iron treated group (12%, p<0.001; 3.5%, p<0.001; 16%, p<0.05 and 47%, p<0.001, respectively). At 12 week there were significant differences between control and placebo in TT4, TT3, T3RU and rT3 concentrations (9.9 vs 8.4 μg dL-1, 145.2 vs 130.4 μg dL-1, 32.5 vs 28.4% and 23 vs 41 μg dL-1, respectively, all p<0.001). Alterations in FT3 and TSH concentration were not significant, but concentration of FT4 revealed a significant difference between the beginning and the end of the study in iron treated group (10.3 vs 11.4, p<0.001). Iron supplementation improves some indices of thyroid function in iron-deficient adolescent girls.
INTRODUCTION
Children and adolescents differ from adults in many aspects, especially in that they continue to grow. The thyroid gland is one of the most important organs that exerts a broad range of effects on development, growth and metabolism. Several minerals and trace elements are essential for thyroid hormone metabolism, e.g., iodine, iron, selenium and zinc. Deficiencies of these elements can impair thyroid function.
The initial steps of thyroid hormone synthesis that is, iodide incorporation into tyrosine residues of thyroglobulin and covalent linking of the residues to generate the diphenylether structure, are catalyzed by heme-dependent thyroperoxidase. Other iron-containing enzymes (e.g., cytochorom oxidase, myeloperoxidase and succinate-ubiquinone oxidoreductase) are sensitive to iron deficiency (Ackrell et al., 1984; Murakawa et al., 1987) .
Nutritional iron deficiency has been shown to significantly lower the circulating levels of both thyroxine and triiodothyronine in rats (Chen et al., 1983; Beard et al., 1984; Brigham and Beard, 1995) and human studies. Martinez-Torres and co-workers (Martinez-Torres et al., 1984), reported 10% lower T3 levels in human subjects with moderate to severe iron deficiency anemia, and Beard and co-workers (Beard et al., 1990b), showed that in iron-deficient subjects serum T3 and T4 levels significantly decreased.
On the other hand peripheral metabolism of thyroid hormones is a critical component of the impact these hormones have on intracellular function. Thyroid hormone can be metabolized in peripheral tissue by deiodination enzyme reactions. Alterations in these metabolic pathway might significantly impact the quantity of specific thyroid hormone metabolites influencing function at cellular level.
Rats with iron deficiency and moderate iron deficiency anemia have reduced conversion of T4 to T3 (Dillman et al., 1980) and lower serum T3 and T4 concentrations compared to controls (Tang et al., 1988). Iron deficiency rats have significantly lower hepatic T4-5’-deiodinase activity, with hepatic production of T3 only 46% of controls (Beard et al., 1990a). Weaning rats fed iron-deficient diets have significantly reduced turnover of serum T3 (approximately 50% lower than control) and lower hepatic thyroxine-5'-deiodinase activity (reduced by 41%) (Beard et al., 1989).
Studies performed to date are highly suggestive of a causal relationship, but because of limitation in research design there is no conclusive evidence that iron deficiency with or without anemia results in abnormal thyroid function or hypothyroidism. Therefore the aim of this randomised, double-blind, placebo-controlled trial was to examine the effects of iron treatment on thyroid function in iron-deficient adolescent girls in southern Iran, in which iron deficiency is quite prevalent (Djazayery, 2000).
MATERIALS AND METHODS
Participants: A double-blind randomized placebo-controlled trial was carried out in the city of Lar and its outskirts in southern Isiamic republic of Iran (altitude; 800 meters above see level), an area in which iron deficiency is prevalent. The study was based at 8 high schools for girls. In the first step, 431 iron-deficient participants (with or without anemia) were selected by systematic random sampling among 2038 students in grades 1 to 4 from the Lar high schools for girls and its vicinity. In the second step 103 participants who met all of the inclusion criteria were chosen. Of 103 girls who enrolled, 94 completed the intervention and 9 dropped out of the study. Of this 9, three girls disliked the tablets and 6 girls claimed that side effects (constipation, gastric discomfort and headache) were the reason for withdrawal. Criteria for case inclusion were: a) absence of any systemic diseases, except for iron deficiency without anemia (hemoglobin>12 g dL-1, serum ferritin <12 μg L-1 and transferrin saturation<16%); b) body mass index>19 kg m2; c) serum albumin within the normal range: 3.5 to 5.5 μg L-1 and d) age within the range of 14 to 18-years-old.
Demographic data, menstruation, any concurrent illness history including surgery, severe bleeding, haemorrhoid, haemophilia and parasitic infections, medication and vitamin and mineral supplementation were collected by interviews and anthropometric indices were determined for each participant. Anthropometric assessments included measurement of weight and height. Body weight was measured to the nearest 0.1 kg using the Seca 713 scales while participants were minimally clothed. Height was determined using measuring tape without shoes and subsequently body mass index, was calculated by dividing weight (kg) by squared height (m2).
Participants were given an oral and written explanation of the study, including its benefits and procedure and at the beginning of the study, participants’ parents or guardian were asked to read and sign an informed consent document.
Clinical trial: This 12 weeks clinical trial was conducted between winter 2002 and summer 2003. The selected participants were randomly assigned in a double-blind fashion to 2 groups. One group received 300 mg oral ferrous sulfate (containing 60 mg elemental iron) as 5 times/week for 12 weeks; the second group received identical-looking placebo tablets. All iron tablets and their placebo had the same colour and shape and were produced by Daroupakhsh company (Tehran, Iran). Participants were asked not to take any vitamin or iron supplements during the trial. The tablets for each girl were packed in a small, same label plastic bag. Student were asked to not drink tea or coffee when taking the tablets which could have inhibited iron absorption and suggest to take it after meal to reduce side effects of ingested tablets. Researcher and the mothers of selected girls supervised ingestion of supplements. If any girl reported any unpleasant side effects of ingested tablets during study, she was encouraged to continue the study but to take supplement with a snack.
Biochemical analysis: At the beginning and the end of the 12 week supplementation trial, 10 mL fasting venous blood samples were drawn from the arm. Blood was collected in two tubes; 2 mL were placed in the EDTA tube for measurement of hemoglobin and hematocrit and 8 mL were placed in another tube for determination of serum albumin, TIBC, iron, ferritin, total and free thyroxin, total and free triiodothyronine, thyrotropin, triiodothyronine resin uptake and reverse triiodothyronine. Iron deficiency without anemia was defined to be present if the Hb concentration was >12 μg dL-1, ferritin<12 μg dL-1 and transferrin saturation<16%. Hemoglobin was measured using the cyanomethemoglobin method, while serum iron, total iron binding capacity and albumin, were measured by the colorimetric method (Dacie and Lewis, 1975) (Zist Chimie Company Lot. No. 11-514, Lot. No. 12-515 and Lot. No. 10-502, respectively). Transferrin saturation was calculated by dividing serum iron by the TIBC and multiplying by 100. Serum ferritin, TSH, TT4, FT4, TT3, FT3, T3RU and RT3 were determined by radioimmunoassay (Henry, 1996), using commercially available kits (Belgium ZenTec for rT3 and American DSL for the rest).
Statistics: Data processing and statistics were done with the use of SPSS version 11 for Windows (SPSS Inc., Chicago, 2001). Normally distributed data were expressed as mean (±SD) and were compared by Independent Student’s t-test and Paired t-test and not normally distributed data were compared using non-parametric tests (Wilcoxon Signed Ranks Test and Mann-Whitney Test). The Simple Linear Regression Model was used to test for possible association (s) and Multiple Linear Regression analysis using stepwise methods was performed to determine the most significant predictors of changes in rT3 concentrations. Significant was set at p<0.05. The correlations were considered significant and clinically important when r was between ≥0.4 or ≤-0.4.
Ethical aspects: The study protocol was reviewed and approved by the Human Ethics Committee of Research Council of the Dean of Research Affair of Tehran University of Medical Sciences and participants’ parents or guardian were asked to read and sign an informed consent document.
RESULTS
Anthropometric data indicated a normal population excluding chronically undernourished girls. Hematological and thyroid values show that all participants at baseline were iron-deficient and had normal thyroid function. Both groups were well matched. Baseline haematological and thyroid indices were similar in both groups at the beginning of study and were no significant differences between two groups (Independent Sample t-test and Mann-Whitney Test, all p>0.05) (Table 1).
As Table 2 shows, some of thyroid indices significantly correlated with serum ferritin. Also regression model showed that when serum ferritin concentration was used as an independent variable, participants with higher T3/T4 ratio had a lower ferritin concentration. Further investigation into the changes in rT3 concentration in these participants, was carried out using multiple regression analysis in which the independent variables included were: TSH, TT4, TT3, FT4, FT3, T3RU, T3/T4 ratio, ferritin and transferrin saturation. Using stepwise regression procedure, only ferritin contributed significantly to the rT3 concentration (r = -0.34, b = -1.3, r2 = 0.12, p<0.01).
Table 3 shows the changes in haematological and thyroid indices after 12 weeks intervention in two groups. Both groups had depleted iron stores as indicated by multiple indicator of iron status. Iron supplementation was effective in correcting the iron deficiency in iron treated group and significantly improved the concentration of hemoglobin, serum ferritin and transferrin saturation (p<0.001 for all three indices). These changes were significant when compared to placebo group (p<0.001 for all three indices). Also, we observed a significant changes regarding these iron indices in the placebo group at the end of the study (p<0.001 for all three indices), although these changes were minimal when compared to changes in the iron treated group.
Serum TT4, TT3 concentrations and T3RU were significantly increased by iron supplementation compared with values in placebo group (p<0.001, p<0.001and p<0.01, respectively) and before intervention (p<0.001 for all three indices). Although the level of FT4 in iron treated group was significantly increased after treatment (p<0.001), but this increase was not significant when compare to placebo group.
| Table 1: | Baseline characteristics of selected iron deficient participants |
| *Mean±standard deviation, **Median; range in parentheses | |
| Table 2: | Pearson correlation coefficient and result of fitting simple
regression line between thyroid hormones concentrations and serum ferritin
in 94 iron deficient participants |
1Statistically significant (p<0.001), that not
clinically important. 2Statistically significant (p<0.001)
and clinically important. 3Not significant |
|
| Table 3: | Hematological and biochemical parameters before and after intervention in two groups |
**Median; range in parentheses, a = p<0.001 and b = p<0.01.
Significantly different from before intervention; Paired t-test, c = p<0.001.
Significantly different from placebo group; Student's t-test, d = p<0.001.
Significantly different from before intervention; Wilcoxon test, e = p<0.001.Significantly
different from placebo group; Wilcoxon test |
|
The changes in the level of TSH and FT3 before and after study and between two groups were not significant.
Observing changes in levels of rT3 in two groups show that the rT3 concentration at the end of study in iron treated group decreased significantly (p<0.001) and this change was significant in comparison to placebo group (p<0.001).
DISCUSSION
This investigation demonstrates that iron deficiency significantly alters thyroid function of adolescent girls, as previously reported in animal and human studies (Chen et al., 1983; Brigham and Beard, 1995; Martinez-Torres et al., 1984; Beard et al., 1984, 1990a, b, 1998). Before group assignment, our data support for the contention that, decrease in TT4 and increase in rT3, TSH concentrations and T3/T4 ratio were significantly correlated with decrease in serum ferritin concentration (Table 2). Although normal thyroid function, as defined by normal levels of thyroid indices, was preserved in our participants with iron deficiency.
Hemoglobin and iron indices in two groups showed a significant improvement at the end of study, although these improvements in placebo group in comparison to iron treated group were very minimal (Table 3). Significant increase in iron indices and hemoglobin concentrations in placebo group could be due to change of season and more access to fruits and vegetables (specially lemon and citrus fruit in this specific area). So significant increase in iron status and hemoglobin in placebo group compared with baseline could be likely due to increase intake of vitamin C, an enhancer of iron absorption.
Extensive data from animal and human studies indicate that iron deficiency impairs thyroid metabolism. Animal studies have documented that weanling rats fed iron-deficient diets have significantly lower T3 and T4 compared to rats fed adequate iron (Chen et al., 1983; Beard et al., 1984; Brigham and Beard, 1995). Rats with iron deficiency and moderate iron deficiency anemia have reduced conversion of T4 to T3 and lower serum T3 and T4 concentrations compared to controls (Tang et al., 1988). On the other hand, iron-deficient rats have significantly lower hepatic 5-deiodinase activity, with hepatic production of T3 only 46% of control (Beard et al., 1990a), reduced turn over of serum T3 and blunted TSH response to exogenous TRH (Beard et al., 1989).
The present study in human participants provides a certain degree of agreement with the animal studies. The results show, iron supplementation could result significant increase in the concentrations of TT3, TT4, FT4 and T3RU in physiological range, with more significant increase in TT4 concentrations. The iron-deficient participants in placebo group showed 16, 9 and 11 lower levels of TT4, TT3 and T3RU and about 50% greater concentrations of RT3, after 12 weeks intervention in comparison to the iron treated group.
Only a few studies performed on humans exist on this subject. Result of these studies showed that in adults, iron deficiency is accompanied by reduced serum T4 and T3, as compared to healthy control (Brigham and Beard, 1995; Martinez-Torres et al., 1984; Dillman et al., 1980). Also Zimmermann et al. (2000), Hess et al. (2002a), showed that iron supplementation improves the efficacy of iodized salt in goitrous children with iron deficiency. In contrast, study of Tienboon (Tienboon and Unachak, 2003), showed that there is no statistical difference in thyroid hormones in the iron deficient-anaemic children before resolution of anemia as, compared to after.
The initial steps of thyroid hormone synthesis-iodide incorporation into tyrosine residues of thyroglobulin and covalent linking of the residues to generate diphenylether structure are catalyzed by heme-dependent tyroperoxidase. Recently Hess et al. (2002b) showed that iron deficiency sharply reduced thyroperoxidase activity and suggest that decreased tyroperoxidase activity contributes to the adverse effects of iron deficiency anemia on thyroid metabolism.
In this study, despite increase in TT3 and TT4 concentrations, TSH concentrations, were unaffected by iron supplementation, although we have seen a negative significant correlation between the ferritin concentration and the TSH level at the beginning of the study and before group assignment.
Zimmermann et al. (2000) showed that at 15 and 30 weeks after iodine supplementation, median TSH values were significantly lower in non anemic goitrous children compared with anemic goitrous children. Also, Beard et al. (1989) has shown that iron deficiency blunts the thyrotropic response to exogenous TRH. On the other hand, Scrimshaw (1984), has reported that in iron deficient women even in situation of cold stress, no variation in the level of TSH secretion is observed. The pituitary production of TSH is under complex feed-back control from intracellular T3, control from TRH and other neurohormonal modulations (Spira and Gordon, 1986).
This discrepancy is not easily explained; however, perhaps our use of only adolescent girls with a certain age and rapid growth period contributed to these different results.
Present study provides support for the contention that decrease in rT3 is related to changes in iron status. As a result of iron supplementation and therefore improvement iron status, serum level of rT3 showed a significant decrease in iron treated group and a significant difference compared to placebo group. Also before group assignment, multiple regression analysis showed that the increased level of rT3 is inversely correlated only with changes in serum ferritin concentration.
Iron deficiency decreased plasma concentrations of T3 and T4 and increased in vitro hepatic rT3 deiodination suggesting that the iron deficient animals tend to metabolize thyroid hormone via a deactivating pathway ( Smith et al., 1993). Presumably, a small fraction of T4 was converted to T3 and a larger proportion metabolized to a physiologically inactive metabolite, rT3. It is not yet clear how iron deficiency exerts its effects on deiodinase activity. Kaplan and Utiger (1978) have shown that the outer ring deiodinase activity is not affected by either ferrous or ferric ions in an in vitro incubation method. This of course, does not rule out the possibility that iron needs to be incorporated into the enzyme during synthesis.
In summary, the effect of iron treatment on thyroid function may have been greater if the follow up had been longer. We did not extend the study past 12 weeks because we wanted to limit the delay in iron treatment of the iron deficient adolescent in the placebo group.
This experiment demonstrates a functional consequence of iron deficiency on thyroid function and indicates that, in proportion to improvement of the iron status of the individuals we notice also a significant improvement in some indices of thyroid hormones. Thus, if iron deficiency is a nutritional factor that influences the alteration in thyroid hormone status, supplementation of this element for these populations will be recommended and especially in area that iodine deficiency is a problem, a combined fortification of salt with iron and iodine, may be valuable.
ACKNOWLEDGMENTS
The authors are indebted to the Dean of Research Affairs of Tehran University of Medical Sciences for financial support and the participating students, their mothers and teachers, for their compliance and patience.