Iodine status of breastfed infants and their mothers' breast milk iodine concentration

Abstract Iodine is an essential nutrient for growth and development during infancy. Data on iodine status of exclusively (EBF) and partially breastfed (PBF) infants as well as breast milk iodine concentration (BMIC) are scarce. We aimed to assess (a) infant iodine nutrition at the age of 5.5 months by measuring urinary iodine concentration (UIC) in EBF (n = 32) and PBF (n = 28) infants and (b) mothers' breast milk iodine concentration (n = 57). Sixty mother–infant pairs from three primary health care centres in Reykjavik and vicinities provided urine and breast milk samples for iodine analysis and information on mothers' habitual diet. The mother–infant pairs were participants of the IceAge2 study, which focuses on factors contributing to infant growth and development, including body composition and breast‐milk energy content. The median (25th–75th percentiles) UIC was 152 (79–239) μg/L, with no significant difference between EBF and PBF infants. The estimated median iodine intake ranged from 52 to 86 μg/day, based on urinary data (assuming an average urine volume of 300–500 ml/day and UIC from the present study). The median (25th–75th percentiles) BMIC was 84 (48–114) μg/L. It is difficult to conclude whether iodine status is adequate in the present study, as no ranges for median UIC reflecting optimal iodine nutrition exist for infants. However, the results add important information to the relatively sparse literature on UIC, BMIC, and iodine intake of breastfed infants.

participants of the IceAge2 study, which focuses on factors contributing to infant growth and development, including body composition and breast-milk energy content. The median (25th-75th percentiles) UIC was 152 (79-239) μg/L, with no significant difference between EBF and PBF infants. The estimated median iodine intake ranged from 52 to 86 μg/day, based on urinary data (assuming an average urine volume of 300-500 ml/day and UIC from the present study). The median (25th-75th percentiles) BMIC was 84 (48-114) μg/L. It is difficult to conclude whether iodine status is adequate in the present study, as no ranges for median UIC reflecting optimal iodine nutrition exist for infants. However, the results add important information to the relatively sparse literature on UIC, BMIC, and iodine intake of breastfed infants.

K E Y W O R D S
diet, human, infant, iodine, lactation, milk

| INTRODUCTION
Iodine is essential for production of thyroid hormones that are crucial for growth and numerous processes of neural and cognitive development (Bougma, Aboud, Harding, & Marquis, 2013;Leung, Pearce, & Braverman, 2011;Zimmermann, Jooste, & Pandav, 2008). Structural brain changes start in foetal life, involving neuronal proliferation and development of axons, synapses, and myelination, and continue well after birth. The plasticity of the central nervous system in the first years of life makes it especially vulnerable to external influences such as nutritional deficiencies and excesses. Thyroid hormone production rates are higher in infancy than any other life stage. Consequently, iodine deficiency can have detrimental effects in the prenatal stage, when cortical formation is at its peak, and postnatally due to neuronal plasticity (Berbel et al., 2009;Delange, 2007;Gothié, Demeneix, & Remaud, 2017;Horn & Heuer, 2010;Leung et al., 2011;Rovet, 2014;Trumpff et al., 2013;Zimmermann, 2007Zimmermann, , 2011Zimmermann, , 2012. The mother is the foetus' only source of thyroid hormones during the first half of pregnancy until its own thyroid starts functioning in the second half. From then on and during exclusive breastfeeding (EBF), the infant relies solely on the mother for iodine to produce its own thyroid hormones.
In the past, iodine status in Iceland has been found to be optimal among adults, adolescents, and pregnant women (Gunnarsdottir et al., 2010;Gunnarsdottir et al., 2013;Gunnarsdottir & Dahl, 2012;Gunnarsdottir, Gustavsdottir, & Thorsdottir, 2009;Nyström et al., 2016). The main reason was the relatively high intake of fish and dairy, together providing 76% of the total iodine intake in the adult Icelandic diet (Þorgeirsdóttir et al., 2011). The dietary habits of the Icelandic population are, however, changing. As a consequence of decrease in fish and dairy intake, insufficient iodine status was, in fact, recently observed in a cohort of pregnant women in Iceland (Adalsteinsdottir et al., 2020). Breastfeeding is highly prevalent in Iceland compared with other high-income countries, with 85% of infants EBF in their first week of life and 35% still EBF at 5 months of age (Hörnell, Lagström, Lande, & Thorsdottir, 2013;Sigbjörnsdóttir & Gunnarsdóttir, 2012;Thorisdottir, Thorsdottir, & Palsson, 2011). No recommendations exist for optimal iodine status of infants, but a cut-off of UIC >100 μg/L for sufficiency for children <2 years has been proposed, based on the criteria used for older children (Andersson, de Benoist, Delange, & Zupan, 2007;WHO/UNICEF/ICCIDD, 2007). Furthermore, no official guidelines for the cut-off for median breast milk iodine concentration (BMIC) exist, but a concentration of >75 μg/L has been suggested to be adequate (Azizi & Smyth, 2009). It should be noted that studies from iodine sufficient areas (indicated by median UIC > 100 μg/L in school-aged children and adults) have reported median values that are clearly higher, that is, BMIC 150 to 180 μg/L (Delange, 2007;Dorea, 2002;Semba & Delange, 2001). The recommended iodine intake of children <2 years of age is 90 μg/day (WHO/UNICEF/ICCIDD, 2007).
The aim of this study was to assess infant iodine nutrition at the age of 5.5 months, by measuring infant UIC in EBF and partially breastfed (PBF) infants as well as BMIC.

| Subjects
Mother-infant pairs were recruited during routine postnatal visits at infant age 5 months in two primary health care centres in Reykjavik and one in the capital vicinity. The eligibility criteria were Icelandic mother, singleton birth, gestational age 37-42 weeks, birth weight > 2,500 g, EBF >2 months, no diseases or defects likely to affect growth or body composition, and either EBF from birth until at least 5.5 months (EBF group) or PBF from 3 to 5 months and receiving at least 100 g or ml of complementary foods in addition to breast milk at 5.5 months (PBF group). All mothers provided written informed consent.

Key messages
• Iodine status of breastfed infants in Iceland was unknown prior to this study.
• The study adds important information to the relatively sparse literature on infant UIC and BMIC.
using sterile pad inserts for diapers and the validation of the method. We measured possible iodine content of the cotton pads and possible entrapment of iodine in cotton fibres, as described below. Mothers then collected a 10-to 15-ml sample of breast milk at the start of a single feed (fore milk) into a 100-ml container.
Breast milk was stored together with the urine sample in household fridges until collected by research staff. The research staff transferred the collected breast milk into 2 mL vials. The samples were then frozen at −80 C at Landspitali University Hospital in Reykjavík, Iceland. Information on sociodemographic characteristics was collected, including marital status, maternal and paternal education, and parity and information on supplement and medication use and whether or not mothers ate from all food groups. Mothers also answered a food frequency questionnaire, providing information on their habitual consumption of key iodine containing foods.

| Iodine concentration and estimated intake
When all urine and breast milk samples had been collected, they were transported on dry ice to Helsinki, Finland. Samples were assayed at the National Institute for Health and Welfare. Iodine concentration of the samples was measured by inductively coupled plasma mass spectrometry (ICP-MS), using an Agilent 7800 ICP-MS system (Agilent Technologies Inc., Santa Clara, CA, USA). This is the recommended method for the determination of trace elements in urine and more complex matrices like breast milk because of its specificity, sensitivity, and low detection limits . For iodine measurements, 100 μl of urine/breast milk sample was extracted by using 2% (w/v) ammonium hydroxide. Iodine was scanned on m/z = 127 and Tellurium was used as an internal standard. All samples were analysed in duplicate.
Iodine contamination tests on the cotton pads used to collect urine during the study period were performed as well as recovery tests (all three batches). Ultrapure water and control urine (with known iodine concentration) were added to the cotton samples. For the purpose of accuracy, the sampling protocol in IceAge2 study required that parents check their child's diaper frequently every 30 min. This means that urine was aimed to be in the cotton pads for a maximum of 30 min. Therefore one half of the control urine samples waited for 0-3 min in the cotton pads and the other for 30-33 min.
The test using ultrapure water revealed iodine concentrations below the detection limit of the ICP-MS method, which is 2 μg/L. The test using control urine showed normal and accurate iodine concentrations of urine samples. Therefore, the cotton pads were considered a reliable method to collect urine in our study. Infant iodine intake was estimated based on UIC and assuming an average urine volume of 300-500 ml/day (Andersson et al., 2007) and compared with rec-

| Statistical analysis
Data were analysed using IBM SPSS for Windows, version 24 (IBM Corp., Armonk, N.Y., USA). Normally distributed data were presented as means (standard deviation), but other data were reported as medians (25th-75th percentile) or numbers (%). UIC and BMIC are presented as medians (25th-75th percentile). Spearman's rho was used to assess possible correlation between UIC, BMIC, and maternal intake of dietary sources of iodine (dairy and fish). Student's t test was used to compare means and Mann-Whitney U test was used to test for differences between groups when data was non-parametric. Pearson Chi-square test and Fisher's exact test were used on categorical data.

| Ethical considerations
The study was approved by the National Bioethical Committee (VSN 13-146) in Iceland. Additional approval was granted for the analysis of UIC and BMIC (VSN 13-146-V3 and V6).

| RESULTS
Sixty mother-infant pairs participated in the study, thereof 32 in the EBF group and 28 in the PBF group. Characteristics of mothers and infants are shown in Table 1. Majority of the mothers had older children (60%) and were married or cohabiting (97%). No significant difference was found between characteristics of infant-mother pairs based on infant feeding methods (EBF vs. PBF).
All 60 mothers provided spot urine samples from their infants.
Based on the UIC, estimated median iodine intake of the infants in the present study was found to be 52-86 μg/day (Andersson et al., 2007).
Median maternal intake of fish and dairy, the two main dietary sources of iodine in the Icelandic diet, was 1.2 times per week and 1.2 times per day, respectively. Table 2 shows median UIC, median BMIC, and infant iodine intake estimated by using UIC and BMIC, in categories of maternal adherence to Icelandic food based dietary guideline on fish (≥two portions per week) and diary (≥two portions per day) intake. More frequent intake of diary, products was associated with higher BMIC (r = .36, p < .008). Other associations between maternal intake and UIC or BMIC were not found.

| DISCUSSION
The median UIC of infant in the present study of 152 μg/L seems to indicate sufficiency in the population studied when compared with proposed cut-off for median UIC >100 μg/L for children <2 years (WHO/UNICEF/ICCIDD, 2007). However, this cut-off is based on criteria for assessing iodine nutrition of older children (WHO/UNICEF/ICCIDD, 2007), and currently, no ranges for median UIC reflecting optimal iodine nutrition exist for infants. Estimated median iodine intake of the infants in the present study was 52-86 μg/day based on UIC and 67-71 μg/day based on BMIC.
When compared with the recommended iodine intake of children <2 years of 90 μg/day (WHO, 2007), insufficient iodine intake cannot be excluded.
While most previous studies on infant UIC have either been done in areas of deficiency, reporting considerably lower UIC than in the present study (Mulrine et al., 2010;Stinca et al., 2017) or from areas of excess, reporting higher UIC than in the present study (Aakre et al., 2016;Nepal et al., 2015;Osei et al., 2016), there are at least two studies that can be found in the literature from sufficient areas (Gordon et al., 2014;Leung et al., 2012). In both studies, the UIC was found to be higher than in the present study; the 39 EBF infants  Leung et al. (2012) reported no information on maternal dietary sources of iodine, but Gordon et al. (2014) reported maternal use of iodized table salts, kelp, iodine containing supplements, and recent consumption of common iodine containing foods. Although historically, Iceland has been known to be an iodinesufficient population, recent data suggest insufficient iodine status in pregnant women in Iceland where median UIC was found to be 89 μg/L (Adalsteinsdottir et al., 2020). Median maternal intake of fish (1.2 times per week) and dairy products (1.2 portions per day) in the present study was similar to intake reported by pregnant women in the study by Adalsteinsdottir et al., (2020) Therefore, it is possible that maternal iodine status in the present study might have been insufficient.
Studies have shown considerable variation in BMIC within and between individuals, and although no official guidelines exist for the cut-off for median BMIC, >75 μg/L has been suggested as adequate (Azizi & Smyth, 2009). However, higher cutoffs have been proposed (Fisher, Wang, George, Gearhart, & McLanahan, 2016), and studies from areas of sufficiency have reported median BMIC 150 to 180 μg/L (Delange, 2007;Dorea, 2002;Semba & Delange, 2001). The results of the present study suggest that a median BMIC of 84 μg/L might not be sufficient for 5-6 months EBF and PBF infants.
Fifty-three mothers in the present study answered a food frequency questionnaire with semiquantitative response options for dairy and frequency options for fish as a main meal. The median con- between the groups easy. Finally, the analysis of UIC and BMIC were done using the ICP-MS method, which is considered the best available method for analysing iodine in urine and breast milk. Despite its considerable strengths, there are some limitations. First, the study is limited by its small sample size regarding the questions on iodine status.
A larger sample size would have made comparison between groups possible, based on the consumption of dietary iodine sources and use of iodine containing supplements. Second, recruiting of participants was dependent on recruitment for IceAge2, in which evaluation of the iodine status was not one of initial objectives. Therefore, the results may not be extrapolated as being valid for the general public.
This is the first study assessing iodine status of breastfed infants in Iceland. It provides new insights into iodine nutrition of infants that have been exclusively or PBF since birth until they are 25 weeks of age or 6 months old. It is difficult to conclude whether iodine status is adequate in the present study, as no ranges for median UIC reflecting optimal iodine nutrition exist for infants. However, the results add important information to the relatively sparse literature on UIC, BMIC, and iodine intake of breastfed infants.

ACKNOWLEDGMENTS
We would like to thank all the infants and their mothers for participating in the study. Furthermore, we are grateful to the primary health care centres for their part in the recruitment process.