Iodine status of Turkish pregnant women and their offspring: A national cross-sectional survey

Journal of Trace Elements in Medicine and Biology 63 (2021) 126664

Available online 7 October 2020

0946-672X/© 2020 The Authors. Published by Elsevier GmbH. This is an open access article under the CC BY-NC-ND license (http://creativecommons.org/licenses/by-nc-nd/4.0/).

Clinical studies

Iodine status of Turkish pregnant women and their offspring: A national

cross-sectional survey

Mehmet Vural

_*

_{, Esin Koc}

_{, Olcay Evliyaoglu}

_{, Hazal Cansu Acar}

_,

Abdurrahman Fatih Aydin

_{, Canan Kucukgergin}

_{, Gozde Apaydin}

_{, Ethem Erginoz}

_,

Xanim Babazade

_{, Sabina Sharifova}

_{, Yildiz Perk}

_{, Turkish Iodine Survey Group}

c_{Department of Public Health, Cerrahpasa Faculty of Medicine, Istanbul University-Cerrahpasa, Kocamustafapasa, Fatih, 34098, Istanbul, Turkey} d_{Department of Medical Biochemistry, Istanbul Medical Faculty, Istanbul University, Çapa, Fatih, 34104, Istanbul, Turkey}

A R T I C L E I N F O

Keywords:

Iodine deficiency

Urinary iodine concentration New-born

Pregnant women

A B S T R A C T

Background: This national cross-sectional survey aimed to assess the iodine status in pregnant women and their

offspring, and also to demonstrate regional differences by measuring urinary iodine concentration (UIC). For each woman and her newborn a questionnaire was prepared with basic facts as age, parity number or birth weight and additional information regarding thyroid diseases, use of iodized salt in the household, extra iodine supplementation during pregnancy, education level and wage income.

Methods: The target population represented 1444 pregnant women who gave birth between January 1 st, 2018

and 2019, and their offspring. Iodine deficiency for pregnant women and their offspring were defined as urine iodine level <150 μg/L and <100 μg/L, respectively. Results are given as median (25th–75th percentile).

Results: The median UIC in the group of pregnant woman was 94 (52–153) μg/L. Within the sample of 1444

pregnant women, UIC indicative of mild iodine deficiency (100− 149 μg/L) was present in 21 % (n = 306), moderate deficiency (50− 99 μg/L) in 30 % (n = 430), and severe deficiency (<50 μg/L) in 23 % (n = 337). This

study showed a prevalence of 74 % of iodine deficiency in Turkish pregnant woman. The median UIC in the group of offspring was 96 (41− 191) μg/L. Within the new-borns, UIC indicative of mild iodine deficiency (50− 99 μg/L) was present in 22 % (n = 323), moderate deficiency (20− 49 μg/L) in 15 % (n = 222), and severe

deficiency (<20 μg/L) in 13 % (n = 192). This survey showed a prevalence of 51 % of iodine deficiency in

Turkish new-borns. Pregnant women with lower socioeconomic and education level, lower access to household iodized salt, lower rates of exposure to povidone-iodine containing skin disinfectant, higher parity and higher iodine deficiency had higher rates of iodine deficiency in their offspring. Regional differences were observed both in mothers and their offspring concerning their iodine status.

Conclusions: Our findings suggest that iodine deficiency is still an important public health problem in Turkey.

More drastic measures should be taken to decrease these important iodine deficiencies, both in pregnant women and in their offspring.

1. Introduction

Iodine is a trace element - essential for the synthesis of thyroid hormones which are involved in growth and development mainly of the neuronal system. Intrauterine growth and brain development depend on sufficient transfer of maternal thyroid hormones during the first trimester, and, from the 16th to 20th pregnancy week, by onset of fetal

thyroid hormone production, they depend on placental iodine transfer [1]. Severe iodine deficiency in a pregnant woman will cause maternal and foetal hypothyroidism with irreversible neurodevelopmental dam-age in the offspring. Iodine deficiency is the single most important preventable cause of severe brain damage [2]. Regarding the iodine deficiency frequency on a global measure, the situation was significantly improved, mainly by the application of iodized salt in the majority of

* Corresponding author.

E-mail address: drmehmetvural@gmail.com (M. Vural).

Contents lists available at ScienceDirect

Journal of Trace Elements in Medicine and Biology

journal homepage: www.elsevier.com/locate/jtemb

https://doi.org/10.1016/j.jtemb.2020.126664

countries. In 1999, World Health Organization recorded iodine defi-ciency in 130 members among 191 member states, whereas in 2019 the Global scorecard of iodine nutrition showed only 18 countries with deficiency [2–5]. Despite this progress in populations` iodine supply, as a rule, the most vulnerable groups for iodine undersupply and deficiency like, expectant and nursing mothers and their offspring, are insuffi-ciently considered. Irrespective of the mentioned strong hypothyroid-ism, it is not clear how mild maternal iodine deficiency may influence child neurobehavioral development [6]. In observational studies, mild iodine deficiency during pregnancy seemed to be associated with lower child intelligence quotient (IQ) or educational assessment scores [7,8]. UIC is a good marker of very recent dietary iodine intake [9]. It may vary up to threefold during a day and cannot be used to diagnose iodine deficiency in an individual with a spot urine sample [10]. The procedure of relating urinary iodine to creatinine (UIC/Cr) has been studied in order to minimize variations caused by differences in urine volume and dilution but is considered to be cumbersome, expensive and unnecessary [11]. Median urinary iodine concentration (UIC) is an acceptable indi-cator for monitoring iodine status in populations, and national population-based monitoring is recommended in school-age children (aged 6–12 years) every five years [11]. On the other hand, it is still not clear and further investigation is needed to decide whether iodine status in this group of age serves as proxy for the general population, and especially for the pregnant women. No universal rule can be given on the association between UIC in school children and in pregnant women and no consistent pattern in the median UIC by trimester can be derived either [12]. Pregnancy is a different period of life with diverse physio-logic changes. Iodine requirements increase due to doubling of serum thyroxine-binding globulin levels under the influence of rising oestrogen and elevated concentrations of human chorionic gonadotropin, enhanced iodide renal clearance, and passage of iodine from the mother to the foetus. The UIC may differ from region to region, even in the same country, depending on different factors such as accessibility to iodized salt or dairy products (main source of iodine in general population) and changing nutritional habits [12].

Household iodized salt use might not be a good indicator of sufficient daily iodine consumption. A study from UK showed that the use of household iodized salt was unlikely to protect individuals from iodine deficiency or contribute meaningful amounts of iodine to the diet [13]. This study underlined that UK iodine intakes were dependent entirely on food choice and good sources, such as milk and fish which might not be consumed sufficiently. Dairy and seafood products are the richest sources of iodine and their consumption is essential to support adequate iodine status. The increasing iodine through the diet might be possible if iodine-rich foods get repositioned in the diet [14]. In Turkey, no study exists showing a correlation between different food choices and iodine intake or deficiency, neither no data exists on the estimations of quan-tities of the consumed food iodine sources. There is no national Iodine supplementation strategy for pregnant women, either.

The WHO/UNICEF/International Council recommended 250 μg/d of iodine intake for pregnant women which could be met by consuming two portions of fish per week and 2 glasses of milk plus one yoghurt and a cheese serving daily [11,14]. Salt iodization is not accepted unani-mously as a good practice and there is lack of legislation in some of the countries like UK. On the other hand, Australia, managed to increase iodine intake in pregnancy by the fortification of bread with iodized salt [15]. An adequate maternal iodine intake should be targeted during pregnancy for healthier generations.

According to the World Bank, Turkey is a country situated partly in Europe and Asia, with a population of 83 million, with 1.25 million live births per year, a fertility rate of 1.99 and GDP per capita 9505 USD in 2018 [16]. Although a national household salt iodization program had been installed since 1998 (industrial salt is not iodized), and the rate of Turkish pregnant women using iodized household salt was very near to the objectives determined by the WHO, regional studies continued to demonstrate important levels of iodine deficiency in the Turkish

population [11,17–19].

The aim of our study was to describe iodine status in pregnant women and their offspring while identifying the different levels of iodine deficiency in both groups (classified as severe, moderate and mild) in 12 regions and on the national level. Further objectives were: 1 To assess the relationship between maternal UIC and the mother’s characteristics detected by a questionnaire (mother’s age, parity, thyroid disease, socioeconomic level, education level, household salt type, iodine supplementation during pregnancy, mode of delivery, povidone-iodine exposure with skin disinfectant before birth) 2 To assess the relationship between offspring’s UIC and the

charac-teristics of newborns and mothers detected with the same question-naire (sex, gestational week, birth weight, birth height, head circumference, mode of delivery, umbilical cord care with disinfec-tants containing povidone-iodine, mother’s age, mother’s socioeco-nomic and education level, household salt type, iodine supplementation during pregnancy, povidone-iodine exposure with skin disinfectant before birth, parity) and also between offspring’s UIC and maternal UIC.

2. Material and methods

2.1. Participants

The target population represented pregnant women who gave birth between January 1st, 2018, and January 1st, 2019, and their offspring. Birth statistics was gathered from Turkish Statistical Institute and 1,248,847 births were registered in 2018 [20]. Stratified sampling was performed for 12 NUTS (Nomenclature of Territorial Units for Statistics) regions defined in 2002 in agreement with Eurostat (Fig. 1) [21].

The sample size was calculated based on an estimated 30 % [17,19] prevalence of iodine deficiency of new-borns (UIC < 100 μg/L) and

pregnant women (UIC < 150 μg/L), with 95 % confidence intervals, a

design effect of one, and an absolute precision of 3%, and it resulted in a sample size of 2812 new-borns and pregnant women for the 12 NUTS regions. The sample size was distributed according to the number of births in each region to determine the sample size in the regions sepa-rately [20]. Designated hospitals, included all the pregnant women-offspring pairs who matched the criteria after the start of the study. The recruitment was stopped once the calculated sample size for the hospital was reached. Only mothers with Turkish nationality were included, in order to eliminate the effect of the recent Syrian refugee inflow. Term and appropriate-for-gestational-age new-borns without congenital malformations were included in the study. Individuals were excluded from the study (Fig. 2) if the urine sample from the mother or offspring was not available and in the event that the UIC was higher than 500 μg/L (considered as heavily contaminated with disinfectants

con-taining povidone-iodine) [22,23].

Multicentre ethics approval was obtained from Istanbul University - Cerrahpasa Medical Faculty (Reference: 83045809-604.01.02). The centres were asked to add the registry database including sociodemo-graphic and socioeconomic characteristics of the mothers using an on-line, standard, patient-specific electronic case report form. Each mother- offspring pair had a number including the code of the region. The identity of the mother was not transmitted and was completely unknown to the researchers.

2.2. Definition of iodine deficiency

Iodine deficiency for pregnant women was defined as UIC < 150 μg/

L. UIC between 100− 149 μg/L, 50− 99 μg/L, and less than 50 μg/L were

defined as mild, moderate, and severe iodine deficiency for the mother, respectively [11].

Iodine deficiency for new-borns was defined as UIC < 100 μg/L [24].

defined as mild, moderate, and severe iodine deficiency for the new-born, respectively.

2.3. Analyses

Urine samples were collected from the mothers just before delivery and from the new-born in the first 48 h of life. Measurements of UIC were performed between July 2018 and July 2019 in Istanbul Medical Faculty, Department of Medical Biochemistry. The Sandell-Kolthoff re-action in a modified microplate method was used according to the rec-ommendations of the US CDC (Centers for Disease Control and Prevention) with minor modifications as described by Tang et al. [25]. Between-run coefficients of variation were 13.1 % for the low (60 μg/L)

and 8.4 % for the high (370 μg/L) level internal quality control samples.

Limit of detection and limit of quantification of the method were 4.8 μg/L and 15.9 μg/L, respectively. The laboratory in which UIC

an-alyses were performed was a member of the CDC EQUIP (Ensuring the Quality of Urinary Iodine Procedures) program. During the study period, samples provided in the CDC EQUIP program were analysed periodically by the laboratory and all results were within acceptable limits. Acceptable limits are defined by the program as target value ± 30 %, 25 %, 20 %, and 15 % for UIC values <50 μg/L, 50− 100 μg/L,

100− 200 μg/L, and >200 μg/L, respectively.

2.4. Statistical methods

Statistical analysis was performed using the SPSS software version

21.0 (SPSS Inc., Chicago, IL, USA). Descriptive statistics (mean, standard deviation, kurtosis and skewness), histograms, and the Kolmogorov- Smirnov test were used for assessing normality. Continuous variables were presented using median and 25th-75th percentiles, and categorical variables were presented as frequency and proportion. Mann-Whitney U and Kruskal-Wallis tests were used to compare continuous variables between the groups. Chi-square test was used to analyze categorical variables. The possible factors identified with univariate analyses (p values < 0.2) were entered into the logistic regression analysis using the backward method to determine independent predictors of outcome. Hosmer-Lemeshow goodness-of-fit statistics were used to assess model fit (p > 0.05 means a model with good predictivity). Results are pre-sented as Odds Ratio (OR) and 95 % Confidence Intervals (95 % CI). A p- value <0.05 was accepted for statistical significance.

3. Results

The median UIC in the group of pregnant woman was 94 (52–153)

μg/L, indicating moderate iodine deficiency. Among the 1444 pregnant

women, UIC indicative of mild iodine deficiency (100− 149 μg/L) was

present in 21 % (n = 306), moderate deficiency (50− 99 μg/L) was seen

in 30 % (n = 430), and severe deficiency (<50 μg/L) was observed in 23

% (n = 337). This study showed a prevalence of 74 % of iodine defi-ciency in Turkish pregnant woman. The severity of iodine defidefi-ciency among the regions is shown in Table 1. The prevalence of iodine defi-ciency in pregnant women showed differences among regions (p = 0.001). In West Marmara region (no. 2), where lowest maternal

Fig. 1. Twelve NUTS regions in Turkey. Iodine deficiency rates are given for each region (pregnant women; offspring).

iodine deficiency rate is detected, the prevalence is found to be 58 %, while the highest rate (84 %) is detected in East Marmara region (no. 4). In West Marmara region (no. 2) and in West Black Sea region (no. 8), mothers had statistically lower, and in East Marmara region (no. 4) statistically higher iodine deficiencies, compared with other regions (Table 1). Highest severe iodine deficiency rates are found to be 30 % in East Black Sea (no.9) and Northeast Anatolia (no.10).

Table 2 shows the characteristics of the pregnant women according to UIC levels. Online patient-specific electronic case reports questioned the age, parity, maternal thyroid disease, familial socioeconomic level, maternal education level, household-salt type consumption, iodine- containing supplements use during pregnancy, mode of delivery, pres-ence of povidone-iodine exposure with skin disinfectant before birth. Among these further criteria studied, only exposure to povidone-iodine containing disinfectant showed a significant effect on women UIC. Among 1444 pregnant women, 1286 (89 %) were consuming household iodized salt. In mothers who had an access to iodized salt and in those did not, iodine deficiency rates were 74 % and 75 % (p > 0.05) (Table 2) while their UIC were 95 μg/L and 87 μg/L (p > 005), respectively.

The median UIC in the group of offspring was 96 (41− 191) μg/L,

indicating mild iodine deficiency. Among the new-borns, UIC indicative of mild iodine deficiency (50− 99 μg/L) was present in 22 % (n = 323), moderate deficiency (20− 49 μg/L) in 15 % (n = 222) and severe

defi-ciency (<20 μg/L) in 13 % (n = 192). This study showed a prevalence of

51 % of iodine deficiency in Turkish new-borns. Table 3 shows the subgroups in each region. There was a significant difference in iodine- deficiency prevalence between the regions (p < 0.001). Iodine defi-ciency was significantly lower in East Marmara, West Anatolia and East Black Sea regions (no. 4, 5, and 9) (43 %, 35 %, and 33 %, respectively), whereas it was significantly higher in Northeast Anatolia region (no.10) (81 %). Severe neonatal iodine deficiency rates varied between 0% in East Black Sea region (no. 9) and 32 % in Northeast Anatolia (no.10).

Table 4 shows characteristics of the new-borns and the characteris-tics of their mothers according to neonatal UIC levels. There was a sig-nificant difference between infants with and without iodine deficiency in terms of birth weight, the socioeconomic and education level of the mother, iodized household salt consumption, maternal povidone-iodine exposure, maternal iodine status, and mother’s parity. When we compared new-borns whose mothers had an access to iodized salt and those did not, we found that neonatal iodine deficiency rates were 50 %

and 61 % (p = 0.01) and new-borns’ UIC were 100 μg/L and 68 μg/L

(p < 0.001), respectively.

Logistic regression analysis was performed using possible factors (p values < 0.2) identified in univariate analysis of the new-borns. The results are shown in Table 5. In new-borns whose mother’s UIC were

<150 μg/L, the risk of iodine deficiency increased 1.53 times (by 53 %)

compared to those whose mother’s UIC were between 150− 249 μg/L

(p = 0,005). The risk of new-born iodine deficiency increased 1.42 times (by 42 %) when mothers had a parity of more than two compared with a parity of one (p = 0.01). The risk of iodine deficiency increased 1.47 times (by 47 %) in new-borns with a birth weight less than 3000 g compared with those whose birthweight was more than 3600 g (p = 0.015). The level of the mother’s education was a risk factor for iodine deficiency in new-borns. Mothers with an education level lower than secondary school increased the risk of iodine deficiency in their offspring 1.27 times (by 27 %) compared with those who had a higher level of education (p = 0.037). Mothers who were not exposed to povidone-iodine containing skin disinfectants, increased the risk of iodine deficiency 1.37 times (by 37 %) in their offspring (p = 0.005).

There were statistical differences between the regions concerning some indicators. A higher percentage of mothers (p < 0.001) had a parity of more than two in regions 10 (56 %) and 12 (64 %). Econom-ically, families had higher percentage of less than guaranteed minimum wage in regions 10 (49 %), 11 (74 %), and 12 (46 %), the most Eastern part of Turkey (p < 0.001). The percentage of mothers with an education level more than the secondary school was higher (p < 0.001) in regions 7 (88 %), 8 (90 %), and 9 (87 %). Lower percentages of household iodized salt use were detected (p < 0.001) in regions 6 (81 %), 10 (68 %), 11 (73 %), and 12 (84 %). The national household iodized salt consumption rate was found to be 89 % (1286/1444). Povidone-iodine exposure of mothers with skin disinfectant was found to be different between the regions (p < 0.001); it was higher in regions 4 (96 %), 5 (81 %), and 6 (94 %), but lower exposure was detected in regions 10 (34 %) and 11 (28 %).

4. Discussion

Previous epidemiologic studies showed that Turkey was a mild-to- moderate iodine deficiency country [26,27]. There had been no na-tional cross-secna-tional survey on iodine status of pregnant women and

Table 1

Prevalence and severity of iodine deficiency in pregnant women in the 12 NUTS regions in Turkey.*

Regions n UIC

Median (25th_-75th percentile)

Iodine deficiency (UIC < 150 μg/L) No iodine deficiency (UIC ≥ 150 μg/L) Severe (<50 μg/ L) Moderate (50− 99 μg/L) Mild (100− 149 μg/ L) Iodine deficiency (Total) 1.Istanbul 320 110 (62− 171) 55 (17 %) 89 (28 %) 77 (24 %) 221 (69 %)a,b _{99 (31 %)}B,C 2.West Marmara 64 122 (50− 197) 16 (25 %) 13 (20 %) 8 (13 %) 37 (58 %)a _{27 (42 %)}C 3.Agean 252 83 (49− 134) 64 (25 %) 84 (33 %) 52 (21 %) 200 (79 %)b,c _{52 (21 %)}A,B 4.East Marmara 176 94 (54− 134) 41 (23 %) 59 (34 %) 47 (27 %) 147 (84 %)c _{29 (16 %)}A 5. West Anatolia 113 93 (55− 146) 24 (21 %) 38 (34 %) 26 (23 %) 88 (78 %)b,c _{25 (22 %)}A,B 6.Mediterranean 138 92 (43− 155) 38 (28 %) 39 (28 %) 26 (19 %) 103 (75 %)b,c _{35 (25 %)}A,B 7.Central Anatolia 25 63 (47− 144) 7 (28 %) 9 (36 %) 3 (12 %) 19 (76 %)b,c _{6 (24 %)}A,B 8.West Black Sea 48 117 (63− 234) 8 (17 %) 13 (27 %) 9 (19 %) 30 (63 %)a _{18 (38 %)}C 9.East Black Sea 46 89 (38− 144) 14 (30 %) 11 (24 %) 10 (22 %) 35 (76 %)b,c _{11 (24 %)}A,B 10. Northeast Anatolia 79 80 (38− 140) 24 (30 %) 23 (29 %) 16 (20 %) 63 (80 %) b,c _{16 (20 %)}A,B 11.Centraleast Anatolia 100 95 (49− 193) 25 (25 %) 27 (27 %) 18 (18 %) 70 (70 %) a,b _{30 (30 %)}B,C 12.Southeast Anatolia 83 84 (49− 159) 21 (25 %) 25 (30 %) 14 (17 %) 60 (72 %) a,b,c _{23 (28 %)}A,B,C All regions 1444 94 (52¡153) 337 (23 %) 430 (30 %) 306 (21 %) 1073 (74 %) 371 (26 %)

Percentages within each row (regions) that do not share same letters (superscripts) are significantly different from each other (p<0,05). *_{Data are given as n (%) except UIC, given as median (25}th_–75th _{percentile). Chi-square test was performed.}

their offspring in our country. Although some national surveys from Europe have been published concerning the iodine status of pregnant women or school girls aged 14–15 years, no similar study pairing the pregnant women and their offspring exist [28,29]. This study showed moderate and mild iodine deficiency in the pregnant women and new-borns respectively despite household salt iodization program. These results indicate important rates of iodine deficiency, in the most critical period of human life when iodine is required for optimal foetal brain development.

4.1. UIC values in pregnant women and their offspring

Regarding the different cut-off values of urine iodine levels in pregnant women and their off-spring, this study showed that the inci-dence of iodine deficiency was higher in mothers than in their new- borns (74 % vs. 51 %), although the median UIC of the mothers (94 μg/L) and their off-spring (96 μg/L) were very close. These results

were not in accordance with previous studies from Iran, showing lower

UIC in mothers compared with their new-borns [23]. There are no specific data concerning UIC for the definition of iodine deficiency, in new-borns. In this study, new-born iodine deficiency (UIC less than 100 μg/L) was defined according to the normal values accepted for

children under the age of two years [24]. Using two-year-old children’s normal range for evaluating the data of one or two-day-old new-born may not be appropriate. A study on healthy, term Swiss new-borns concluded that the current WHO median UIC cut-off (100 μg/L) for

iodine sufficiency in infancy may be too high for the first week after birth [30]. In this Swiss study, the median UIC in the total sample (n = 1224) was 77 μg/L and a gradual increase in median UIC within the

range of 70− 100 μg/L was found from days one through four. If we

considered our cut-off value as 77 μg/L, we would have found 618/1444

(43 %) iodine deficiency in our group of new-borns. This prevalence would be still too high.

4.2. Iodine supplementation in pregnancy and UIC outcome

A national table salt iodization program in Turkey has been estab-lished since 1998, and 89 % of pregnant women reported using iodized salt in our survey. The rate of Turkish pregnant women using iodized salt was very near to the objectives determined by the WHO [11]. In a previous study from the Northern part of Turkey, 99 % of mothers declared having used iodized salt, but 57 % of them were still iodine deficient [19]. We demonstrated that there were significant differences between regions concerning household iodized salt use. Pregnant women from Eastern Anatolian regions (no. 10,11, and 12) used less iodized salt and those who did not use iodized salt had a higher per-centage of new-borns with iodine deficiency.

Although national salt iodization program has significantly reduced incidence and severity of iodine deficiency, it is still not optimal. High household iodized salt consumption rate among pregnant women was not reflected in their UIC and iodine deficiency rates, in our study (Table 2). This might indicate improper use of iodized salt, inappro-priately iodized salt, or iodine lost during the packaging process or on the market shelves. However, it was not surprising to find that new- borns had higher UIC and lower iodine deficiency rates when their mothers consumed iodized salt (Table 4). During the pregnancy there is an influx of iodine as all nutrients/elements from mother to foetus at expense to deplete mother stores [31]. Additionally, iodide transport and uptake by the mammary gland are active mechanisms mediated by the sodium-iodide symporter (NIS) which are enhanced during lactation [32]. Considering an adequate iodine supply, breast milk contains iodine in a magnitude of 100− 200 μg/L [33]. In this German study,

there was a drastic increase of the breast milk iodine level in the time after 2000 compared with the 1980s and the 90 s as periods with iodine deficiency. Nearly 80 % of total iodine in mature human milk is in the form of iodide. Especially iodine content is highest in colostrum and may reach to the levels of 200− 400 μg/L [34]. Increased iodine flow into the

mammary gland compensates for insufficient maternal iodine intake at the expense of maternal iodine reserves [35]. All these factors can explain this discord between the UIC of the pregnant women and their offspring.

Pregnant women with UIC < 150 μg/L increased the risk of having an

iodine-deficient off-spring 1.53 times (by 53 %) in the regression anal-ysis. This finding also emphasised the importance of iodine supple-mentation in pregnant and lactating women in order to have lower rates of iodine-deficient new-borns. A randomized, double-blind, placebo- controlled supplementation trial was conducted in lactating women who received placebo, 75 μg/d, or 150 μg/d of iodide after their infants’

birth, until 24 weeks postpartum, in New Zealand [36]. It was shown that the breast milk iodine concentration decreased in the first six months in the placebo receiving lactating women. Supplementation with 75 or 150 μg/d increased the breast milk iodine concentration but

was insufficient to ensure adequate iodine status in the women or their infants.

Table 2

Characteristics of the pregnant women according to UIC levels.* Deficiency (<150 μg/ L) Optimal (150− 249 μg/ L) More than adequate (≥250 μg/ L) p valuea Age median (25th_–75th percentile) 30 (25− 34) 30 (26− 34) 30 (27− 35) 0.135 Parity n (%) 0.539 Parity=1 314 (72 %) 75 (17 %) 50 (11 %) Parity=2 351 (75 %) 73 (16 %) 46 (10 %) Parity>2 408 (76 %) 74 (14 %) 53 (10 %) Thyroid disease n (%) 0.14 Yes 84 (67 %) 26 (21 %) 15 (12 %) No 989 (75 %) 196 (15 %) 134 (10 %) Socioeconomic level n (%) 0.463 Income equal to or less than minimum wage

373 (73 %) 82 (16 %) 59 (12 %) Income more than

minimum wage 700 (75 %) 140 (15 %) 90 (10 %)

Education level n (%) 0.92

Less than secondary

school 409 (75 %) 82 (15 %) 58 (11 %) More than secondary

school 664 (74 %) 140 (16 %) 91 (10 %)

Household Salt type n

(%) 0.948 Iodized salt 954 (74 %) 199 (16 %) 133 (10 %) Non-iodized salt 119 (75 %) 23 (15 %) 16 (10 %) Iodine supplementation during pregnancy n (%) 0.754 Yes 58 (73 %) 11 (14 %) 10 (13 %) No 1014 (74 %) 211 (16 %) 139 (10 %) Mode of delivery n (%) 0.139 Vaginal delivery 274 (78 %) 49 (14 %) 28 (8 %) Caesarean section 799 (73 %) 173 (16 %) 121 (11 %) Povidone-iodine exposure with skin disinfectant before birth n (%)

0.013

Yes 659 (72 %) 147 (16 %) 109 (12 %) No 414 (78 %) 75 (14 %) 40 (8 %)

*_{For each criterion with two or three subgroups numbers of women from a} total of 1444 are distributed into the three UIC ranges. In brackets, percentages are given for women in the subgroups distributed into the three UIC ranges from totally.100 %.

In 7/12 regions, the median UIC for the new-borns was below the cut-off level of 100 μg/L (Table 3). Although more studies are needed to

demonstrate the harmful effects of mild iodine deficiency (UIC in the range of 50− 99 μg/day), values of UIC < 50 μg/L and especially

UIC < 20 μg/L may result in important neurologic impairments [37]. In

our national survey, 13 % of new-borns presented severe iodine defi-ciency. In regions 6, 10, 11, and 12, severe iodine deficiency rates reached 25 %, 32 %, 25 %, and 24 %, respectively. This percentage of severe iodine deficiency, in these regions, should be considered as very disturbing and alarming.

4.3. Characteristics of the pregnant women and their effects in UIC

Although increased parity was not a risk factor for iodine deficiency in pregnant women, it was found to be a significant risk in children. The regression analysis showed that the risk of iodine deficiency in the offspring increased 1.42 times (by 42 %) when mother’s parity was more than two compared with parity of one. We could propose that increased number of parities would result in exhaustion of thyroid iodine stores. A study conducted in Iran showed that median UIC in mothers who had three or more children was lower compared with mothers with only one child [23]. In our survey, new-borns with lower birthweight had a higher risk of iodine deficiency. Risk of iodine deficiency was increased 1.47 times (by 47 %) in new-borns with birth weight <3000 g compared with those with a birthweight >3600 g. We can speculate that lower birthweight in term new-borns and increased parity may indicate diminished placental perfusion and lower placental iodine transfer during pregnancy.

Considering the education level, pregnant women with less than the secondary school created an increased risk of iodine deficiency in their offspring, 1.27 times (by 27 %). Mothers with higher degrees of edu-cation had less parity, a higher income, and greater access to household iodized salt. All these factors may explain the higher iodine levels in infants of mothers with a higher level of education. Our results were in accordance with the results of the Danish study where intake of iodine containing supplements was less common among women with a low level of education [12]. In Iran, low knowledge of iodine deficiency of women at child bearing age was linked to lower iodine status [38]. In a cross-sectional survey of 1026 UK mothers, 55 % were unable to identify sources of iodine and only 9% of women could recognise milk as a source of iodine [39]. There is no study from Turkey that reports the awareness of the Turkish women on iodine deficiency nor knowledge on iodine sources.

4.4. Changes in UIC according to different regions

In our survey, we also found some statistical differences between regions. The prevalence of iodine deficiency in pregnant women was different among the regions. The parameters we evaluated in this study did not show any differences in these regions (no. 2 and no. 8) to explain higher iodine status, and in region no 4 to explain lower iodine status of the pregnant women. We assumed that parameters which were not investigated in this survey had also affected and increased the maternal iodine intake. On the other hand, iodine deficiency was significantly higher in new-borns in region 10 compared with the other regions. This region is located in the most Eastern part of the country, with a higher proportion of families receiving less than guaranteed minimum wage, more women with a parity of more than two, lower education levels, and lower rates of household iodized salt access. All these factors, which were significantly different compared to other regions, most probably influenced the higher iodine deficiency rates detected in new-borns. On the other hand, new-born iodine deficiency was detected with signifi-cantly lower rates in regions 4, 5, and 9 compared with the other re-gions. Regions 4 and 5 are located in the Western part of Turkey, whereas region 9 is located in the Eastern part of Turkey, all along the Black Sea, possibly with easier access to seafood rich in iodine. Region 9 mothers declared higher rates of more than guaranteed minimum wage and higher rates of education level more than secondary school, which may result in a more balanced and diversified nutrition and access to iodized salt.

4.5. Confounders and limitations of the study

This survey had some limitations. Although our sample size was 2812 at the beginning, we managed to have 51 % of the targeted number and ended up with 1444 pregnant women-offspring pairs. Similar dis-crepancies between the target and studied numbers were reported from studies on iodine deficiency [23]. In regions 9 and 12 the percentages were lower, 29 % and 31 %, respectively. The prevalence of iodine deficiency in Turkish new-borns differed from 28 to 100 % in different previous regional studies [17,27]. Pregnant women in Turkey had an iodine deficiency rate which were between 57–90% in local studies [19,

27]. Although the sample size in our national survey was calculated based on a prevalence of 30 %, we found it to be 51 % for new-borns and 74 % for pregnant women. Post-hoc sample size calculation indicated that present sample was sufficiently large to determine the prevalence in both populations. In this survey, we only had an urine sample at the end of the pregnancy from the pregnant women and no sampling was per-formed during the pregnancy. Although this can be considered as a

Table 3

Prevalence and severity of iodine deficiency in new-borns in 12 NUTS regions in Turkey.*

Regions n UIC

Median

(25th_-75th _percentile)

Iodine deficiency (UIC < 100 μg/L) Optimal (UIC ≥ 100 μg/L) Severe

(<20 μg/L) Moderate (20− 49 μg/L) Mild (55− 99 μg/L) Iodine deficiency (Total)

1.Istanbul 320 107 (58− 201) 26 (8 %) 41 (13 %) 83 (26 %) 150 (47 %)b _{170 (53 %)}B 2.West Marmara 64 84 (44− 149) 4 (6 %) 15 (24 %) 18 (28 %) 37 (58 %)b _{27 (42 %)}B 3.Agean 252 93 (34− 172) 36 (14 %) 39 (16 %) 57 (23 %) 132 (52 %)b _{120 (48 %)}B 4.East Marmara 176 120 (65− 189) 12 (7 %) 17 (10 %) 46 (26 %) 75 (43 %)a _{101 (57 %)}C 5. West Anatolia 113 157 (60− 263) 5 (4 %) 16 (14 %) 19 (17 %) 40 (35 %)a _{73 (65 %)}C 6.Mediterranean 138 72 (20− 229) 34 (25 %) 19 (14 %) 24 (17 %) 77 (56 %)b _{61 (44 %)}B 7.Central Anatolia 25 97 (53− 190) 1 (4 %) 3 (12 %) 9 (36 %) 13 (52 %)b _{12 (48 %)}B 8.West Black sea 48 101 (53− 194) 4 (8 %) 6 (13 %) 14 (29 %) 24 (50 %)b _{24 (50 %)}B 9.East Black sea 46 166 (88− 240) 0 (0 %) 3 (7 %) 12 (26 %) 15 (33 %)a _{31 (67 %)}C 10. Northeast Anatolia 79 35 (16− 81) 25 (32 %) 28 (35 %) 11 (14 %) 64 (81 %)c _{15 (19 %)}A 11.Centraleast Anatolia 100 63 (20− 147) 25 (25 %) 18 (18 %) 19 (19 %) 62 (62 %)b _{38 (38 %)}B 12.Southeast Anatolia 83 75 (23− 166) 20 (24 %) 17 (21 %) 11 (13 %) 48 (58 %)b _{35 (42 %)}B

All regions 1444 96 (41¡191) 192 (13 %) 222 (15 %) 323 (22 %) 737 (51 %) 707 (49 %)

Percentages within each row (regions) that do not share same letters (superscripts) are significantly different from each other (p<0,05). *_{Data are given as n (%) except UIC, given as median (25th–75th percentile). Chi-square test was performed.}

limitation, a study from Japan tried to characterise the gestational change of UIC during pregnancy and found no differences in UIC values of three trimesters [40].

The results in Tables 1 and 3, which show the pregnant women’s and newborns’ prevalence of iodine deficiency by regions, can be thought as contradictory. The situation that the prevalence of iodine deficiency was higher in mothers in the regions number 4, 5 and 9 but lower in their offspring could create confusion. However, in the present study maternal IUC was only one of the factors affecting iodine status of new- borns, beside parity, birth weight, maternal education level and the povidone-iodine exposure with skin disinfectant. Also, there might be other parameters not investigated in this survey which might have an

effect on iodine status of new-borns. The results of regression analysis showed that the risk of iodine deficiency increased 1.53 times in new- borns whose mother had iodine deficiency.

The knowledge/awareness of iodine supplementation of Turkish pregnant women was not included in the questionnaire. We have not questioned their obstetricians on their awareness about iodine supple-mentation, either. Healthcare professionals have low awareness of the importance of iodine in pregnant women worldwide. In the USA 75 % of obstetricians and midwives reported to rarely or never prescribe iodine- containing supplements [41]. In a study from Turkey, 67.1 % of physi-cians thought it was unnecessary to offer iodine supplementation to pregnant women [42]. In our study only 5% of the pregnant women declared to have received iodine supplementation during their preg-nancy. We assume that heightening the awareness of Turkish physicians will contribute to ameliorate the very high iodine deficiency rates of our study.

Our study was not designed to investigate the neonatal thyroid- stimulating hormone (TSH) levels. A recent study from Turkey, showing the results of the national Neonatal TSH screening program, including 1 270 311 new-borns, observed 7.2 % elevated TSH in 2014 [43]. The highest elevated TSH rate (15.9 %) was observed in region 12 (southeast Anatolia). In our study, pregnant women in region 12 had one of the lowest median UIC (84 μg/L) and 25 % severe iodine deficiency.

The new-borns of the region 12 had one of the lowest median UIC (75 μg/L) and one of the highest iodine deficiency rates (58 %). The

national screening results showed that the lowest rate of elevated TSH was 4% and was observed in region 2 (west Marmara) [43]. In our study the region 2 had the highest median UIC for the pregnant women and the lowest iodine deficiency rate (58 %).

Although we have investigated the iodized salt consumption of the pregnant women, no question was asked neither about the dietary habitudes nor about the quantity of products consumed, rich in iodine (dairy products and sea products). The quality of household iodization is not investigated, either, in this study. A study from Turkey showed that only 40 % of household salt samples were sufficiently iodized, con-taining >15 ppm iodine, which is the lower limit recommended by WHO [44]. Together with this insufficient salt iodization rates, salt use for cooking, consumption of non-iodized industrial salt via processed bak-ery and meat products and dietary habits may all explain the low median UIC in our population, reporting to have an access rate of 89 % to household iodized salt. The main dietary sources are milk, eggs and fish.

Table 4

Characteristics of the new-borns and the pregnant women according to neonatal UIC levels.*

New-born Characteristics Deficiency

(<100 μg/L) Optimal (≥100 μg/L) p valuea Sex n (%) 0.341 Girl 309 (53 %) 279 (47 %) Boy 428 (50 %) 428 (50 %) Gestational week n (%) 0.236 37th-39th weeks 602 (52 %) 560 (48 %) ≥40 weeks 135 (48 %) 147 (52 %) Birth Weight n (%) <3000 g 175 (55 %) 145 (45 %) 0.045 3000− 3600 grams 405 (52 %) 374 (48 %) >3600 g 157 (46 %) 188 (55 %) Height median (25th-75th percentile) 50 (49− 51) 50 (49− 51) 0.101

Head circumference median (25th-

75th percentile) 35 (34− 36) 35 (34− 36) 0.813

Mode of delivery n (%) 0.317

Vaginal delivery 171 (49 %) 180 (51 %) Caesarean section 566 (52 %) 527 (48 %)

Umbilical cord care with disinfectants containing povidone-iodine n (%) 0.644 Yes 93 (50 %) 95 (51 %) No 644 (51 %) 612 (49 %) Maternal Characteristics

Age median (25th-75th percentile) 30 (26− 35) 30 (26− 34) 0.572

Socioeconomic level n (%) 0.005

Equal or less than minimum wage

income 288 (56 %) 226 (44 %) More than minimum wage income 449 (48 %) 481 (52 %)

Education level n (%) 0.001

Less than secondary school 312 (57 %) 237 (43 %) Secondary school and above 425 (48 %) 470 (53 %)

Household Salt type n (%) 0.01

Iodized salt 641 (50 %) 645 (50 %) Non-iodized salt 96 (61 %) 62 (39 %)

Iodine supplementation during

pregnancy n (%) 0.765

Yes 39 (49 %) 40 (51 %)

No 697 (51 %) 667 (49 %)

Povidone-iodine exposure with skin disinfectant before birth n

(%) 0.001 Yes 436 (48 %) 479 (52 %) No 301 (57 %) 228 (43 %) Maternal UIC n (%) 0.002 Deficiency (<150 μg/L) 577 (54 %) 496 (46 %) Optimal (150− 249 μg/L) 93 (42 %) 129 (58 %) More than adequate (>250 μg/L) 67 (45 %) 82 (55 %)

Parity n (%) 0.004

Parity=1 199 (45 %) 240 (55 %) Parity=2 238 (51 %) 232 (49 %) Parity >2 300 (56 %) 235 (44 %)

*_{For each criterion with two or three subgroups numbers of new-born from a} total of 1444 are distributed into the two UIC ranges. In brackets, percentages are given for new-born in the subgroups distributed into the three UIC ranges from totally.100 %.

a_{Chi-square and Mann Whitney U tests were performed.}

Table 5

Logistic regression results of factors affecting iodine status of new-borns.

Predictors OR (95 % CI) p

value

Maternal UIC (Ref. 150− 249 μg/L)

<150 μg/L 1.53

(1.14− 2.07) 0.005

250− 499 μg/L 1.23

(0.73− 1.72) 0.593

Parity (Ref. Parity = 1)

Parity = 2 1.22

(0.93− 1.59) 0.14

Parity > 2 1.42

(1.08− 1.87) 0.01

Birth weight (Ref. > 3600 g)

<3000 g 1.47

(1.07− 2.02) 0.015

3000− 3600 grams 1.26

(0.97− 1.64) 0.072

Maternal education level (Ref. secondary school and

above)

Less than secondary school 1.27

(1.01− 1.60) 0.037

Povidone-iodine exposure with skin disinfectant

(Ref. Yes)

No 1.37

It is estimated that milk contributes almost one half of the total iodine intake in the UK and in many industrialised countries the dairy products have become major sources of nutritional iodine with an estimated contribution between 25–70 % to the daily iodine intake [45–47]. Iodine in milk comes from indirect fortification through animal feeds and iodine containing antiseptic use [48–50]. The consumption of household salt may represent only a low part of total salt consumption. In industrial countries, the daily per capita consumption ranges from 1− 2 g house-hold salt representing 10–20 % of total estimated daily intake of 10 g/day [51]. The majority of (unintentional) salt intake comes from bakery goods, meat products (ham and sausage) or from canned vege-tables. As a rule, salt for these preservation purposes in the food manufacturing is not iodized (except countries with Universal Salt Iod-ization prescribed by legislation). Although a national table salt iod-ization program has been established in Turkey, since 1998, industrial salt is not iodized. USI (universal salt iodization) is effective, but only when it’s truly universal when a) sufficient iodine (20–40 mg/kg) is added to all salt for human consumption, including table salt and salt used in food production by manufacturers of processed foods and con-diments; b) more than 90 % of households in a country have access to and consume adequately iodized salt (≥15 mg/kg). It is generally assumed that the iodine requirements of all population groups are covered in settings where USI has been successfully implemented for ≥2 years, and iodine intakes in school-age children are adequate [52]. In order to prevent the iodine deficiency in pregnant women, and most importantly, severe iodine deficiency in Turkish new-borns, Turkey needs more drastic measures such as iodization of industrial salt, bread and/or iodine addition to mineral vitamin supplements given during pregnancy, changing the diet habitudes to increase the seafood and fish consumption and specially to increase the milk consumption together with optimising its concentration in iodine [14]. A research investi-gating food choices of Turkish mothers and more knowledge of food iodine concentration is also needed.

A confounder factor in this survey may be the utilisation of skin disinfectants containing povidone-iodine which is still very frequent in Turkey [19]. The risk of iodine deficiency was increased 1.37 times (by 37 %) in new-borns whose mothers were not exposed to disinfectants containing povidone-iodine. This was an expected effect of iodine, absorbed by the skin and transported through the placenta to the foetus. Procedures such as Caesarean section, vaginal douching, epidural anaesthesia, and umbilical cord antisepsis are the causes of iodine exposure in pregnant women and the new-born [19]. It had been shown that these procedures could result in a 12-fold increase in human milk iodine levels [53]. The results of our survey showed that the povidone-iodine exposure decreased the rate of iodine deficiency in new-borns. The neonatal iodine deficiency prevalence was probably masked and underestimated due to this iodine exposure.

5. Conclusions

This study was the first national cross-sectional survey that pairs pregnant women with their offspring, permitting the investigation of the iodine status of new-borns in relation to their mothers. It showed high levels of iodine deficiency rates in the pregnant women-offspring-pairs which should impose the re-evaluation of household iodized salt pro-gram (intensive monitoring of iodine levels of salt at the factory, household, and retail level). Active public health and policy strategies like iodization of the industrial salt, appropriate dietary recommenda-tions and educative programmes to increase the knowledge/awareness on iodine are very important challenges for Turkey, in the coming years.

Declaration of Competing Interest

The authors report no declarations of interest.

Acknowledgements

The Turkish Pediatric Association provided funding for the cost of the materials and assays used in the study. Electronic case report form system funding was supported by the Turkish Neonatal Society. The Turkish Pediatric Association and Turkish Neonatology Society contributed to and supported the study through their members. The corresponding author had full access to all data in the study and had final responsibility for submitting it for publication. We are grateful to Canan Uremis, secretary of the Turkish Pediatric Association, for the hard work accomplished and the time devoted.

Appendix A. Supplementary data

Supplementary material related to this article can be found, in the online version, at doi:https://doi.org/10.1016/j.jtemb.2020.126664.

References

[1] N.K. Moog, S. Entringer, C. Heim, P.D. Wadhwa, N. Kathmann, C. Buss, Influence of maternal thyroid hormones during gestation on fetal brain development, Neuroscience 342 (2017) 68–100.

[2] B. de Benoist, E. McLean, M. Andersson, L. Rogers, Iodine deficiency in 2007: global progress since 2003, Food Nutr. Bull. 29 (2008) 195–202.

[3] World Health Organization, Assessment of Iodine Deficiency Disorders and Monitoring Their Elimination: a Guide for Programme Managers, 2nd ed., World Health Organization, 2001 (Access date: 04.04.2020), https://apps.who.int/iris /handle/10665/61278.

[4] E.N. Pearce, M. Andersson, M.B. Zimmerman, Global iodine nutrition: Where do we stand in 2013? Thyroid 23 (5) (2013) 523–528.

[5] Iodine Global Network 2018 Annual Report, 2020 (Access date: 04.04.2020),

https://www.ign.org/cm_data/IGN_2018_Annual_Report_5_web.pdf. [6] G.M. de Escobar, M.J. Obreg´on, F.E. del Rey, Iodine deficiency and brain

development in the first half of pregnancy, Public Health Nutr. 10 (2007) 1554–1570.

[7] S.C. Bath, C.D. Steer, J. Golding, P. Emmett, M.P. Rayman, Effect of inadequate iodine status in UK pregnant women on cognitive outcomes in their children: results from the Avon Longitudinal Study of Parents and Children (ALSPAC), Lancet 382 (Jul. (27)) (2013) 331–337.

[8] K.L. Hynes, P. Otahal, I. Hay, J.R. Burgess, Mild iodine deficiency during pregnancy is associated with reduced educational outcomes in the offspring: 9-year follow-up of the gestational iodine cohort, J. Clin. Endocrinol. Metab. 98 (May (5)) (2013) 1954–1962.

[9] M.B. Zimmermann, S.Y. Hess, P. Adou, T. Toresanni, R. Wegmüller, R.F. Hurrell, Thyroid size and goiter prevalence after introduction of iodized salt: a 5-y prospective study in schoolchildren in Cˆote d’Ivoire, Am. J. Clin. Nutr. 77 (Mar. (3)) (2003) 663–667.

[10] P. Laurberg, S. Andersen, R.I. Bjarnad´ottir, A. Carl´e, A. Hreidarsson, N. Knudsen, L. Ovesen, I. Pedersen, L. Rasmussen, Evaluating iodine deficiency in pregnant women and young infants-complex physiology with a risk of misinterpretation, Public Health Nutr. (Dec. (12A)) (2007) 1547–1552.

[11] World Health Organization, Assessment of Iodine Deficiency Disorders and Monitoring Their Elimination: a Guide for Programme Managers, 3rd ed., World Health Organization, 2007 (Access date: 04.04.2020), https://apps.who.int/iris /handle/10665/43781.

[12] S.L. Andersen, P. Laurberg, Iodine supplementation in pregnancy and the dilemma of ambiguous recommendations, Eur. Thyroid J. 5 (Mar. (1)) (2016) 35–43. [13] S.C. Bath, S. Button, M.P. Rayman, Availability of iodised table salt in the UK - is it

likely to influence population iodine intake? Public Health Nutr. 17 (Feb. (2)) (2014) 450–454.

[14] M. Bouga, M.E.J. Lean, E. Combet, Contemporary challenges to iodine status and nutrition: the role of foods, dietary recommendations, fortification and supplementation, Proc. Nutr. Soc. 77 (Aug. (3)) (2018) 302–313.

[15] D. Mackerras, J. Powers, J. Boorman, D. Loxton, G.G. Giles, Estimating the impact of mandatory fortification of bread with iodine on pregnant and post-partum women, J. Epidemiol. Community Health 65 (Dec. (12)) (2011) 1118–1122. [16] https://www.worldbank.org/en/country/turkey/overview (Access date:

04.04.2020).

[17] F. Kis¸lali, S. Cetinkaya, U. Dilmen, H. Yas¸ar, T. Teziç, Cord blood thyroid- stimulating hormone and free T4 levels in Turkish neonates: is iodine deficiency still a continuing problem? Pediatr. Int. 52 (2010) 762–768.

[18] A. Gokcel, M. Serteser, A. Coskun, I. Unsal, Iodine deficiency has recurred in Istanbul, Turkey, Endocrinol. Metab. Int. J. 5 (1) (2017) 172–175.

[19] A.K. Yaman, F. Demirel, B. Ermis¸, I.E. Pis¸kin, Maternal and neonatal urinary iodine status and its effect on neonatal TSH levels in a mildly iodine-deficient area, J. Clin. Res. Pediatr. Endocrinol. 5 (2) (2013) 90–94.

[20] https://biruni.tuik.gov.tr/bolgeselistatistik/anaSayfa.do?dil=en (Access date: 04.04.2020).

[21] https://ec.europa.eu/eurostat/web/nuts/statistical-regions-outside-eu (Access date: 04.04.2020).

[22] D. Levie, S.C. Bath, M. Guxens, T.I. Korevaar, M. Dineva, E. Fano, J.M. Ibarluzea, S. Llop, M. Murcia, M.P. Rayman, J. Sunyer, R.P. Peeters, H. Tiemeier, Maternal iodine status during pregnancy is not consistently associated with attention-deficit hyperactivity disorder or autistic traits in children, J. Nutr. (March) (2020) [Epub ahead of print].

[23] P. Nazeri, P. Mirmiran, M. Hedayati, Y. Mehrabi, H. Delshad, F. Azizi, Can postpartum maternal urinary iodine be used to estimate iodine nutrition status of newborns? Br. J. Nutr. 115 (7) (2016) 1226–1231.

[24] WHO Secretariat, M. Andersson, B. de Benoist, F. Delange, J. Zupan, Prevention and control of iodine deficiency in pregnant and lactating women and in children less than 2-years-old: conclusions and recommendations of the Technical Consultation, Public Health Nutr. 10 (12A) (2007) 1606–1611.

[25] K.T. Tang, F.F. Wang, S.S. Fu, L.E. Braverman, J.D. Lin, G.S. Won, A simple microplate method with improved low iodine concentration sensitivity in urinary iodine measurement, Thyroid 25 (2015) 1173–1174.

[26] N. Yordam, A. Oz¨on, A. Alikas¸ifo˘glu, T. Gunes, N. Budak, M.E. Atabek, I. Karakucuk, F. Delange, Iodine deficiency in Turkey, Eur. J. Pediatr. 158 (6) (1999) 501–505.

[27] S. Kurtoglu, M. Akcakus, C. Kocaoglu, T. Gunes, N. Budak, M.E. Atabek, I. Karakucuk, F. Delange, Iodine status remains critical in mother and infant in Central Anatolia (Kayseri) of Turkey, Eur. J. Nutr. 43 (5) (2004) 297–303. [28] S. Vandevijvere, S. Amsalkhir, A.B. Mourri, H. Van Oyen, R. Moreno-Reyes, Iodine

deficiency among Belgian pregnant women not fully corrected by iodine- containing multivitamins: a national cross-sectional survey, Br. J. Nutr. 109 (12) (2013) 2276–2284.

[29] M.P. Vanderpump, J.H. Lazarus, P.P. Smyth, P. Laurberg, R.L. Holder, K. Boelaert, J.A. Franklyn, British Thyroid Association UK Iodine Survey Group, Iodine status of UK schoolgirls: a cross-sectional survey, Lancet 377 (2011) 2007–2012. [30] C.M. Dorey, M.B. Zimmermann, Reference values for spot urinary iodine

concentrations in iodine-sufficient newborns using a new pad collection method, Thyroid 18 (2008) 347–352.

[31] F.D. de Carvalho, M. Quick, Surprising substrate versatility in SLC5A6: Na+- coupled I- transport by the human Na+/multivitamin transporter (hSMVT), J. Biol. Chem. 286 (1) (2011) 31–37.

[32] C. Spitzweg, W. Joba, W. Eisenmenger, A.E. Heufelder, Analysis of human sodium iodide symporter gene expression in extrathyroidal tissues and cloning of its complementary deoxyribonucleic acids from salivary gland, mammary gland, and gastric mucosa, J. Clin. Endocrinol. Metab. 83 (1998) 1746–1751.

[33] M. K¨ohler, A. Fechner, M. Leiterer, K. Sp¨orl, T. Remer, U. Sch¨afer, G. Jahreis, Iodine content in milk from German cows and in human milk: new monitoring study, Trace Elem. Electrolytes 29 (2012) 119–126.

[34] N. Etling, E. Padovani, F. Fouque, L. Tato, First-month variations in total iodine content of human breast milks, Early Hum. Dev. 13 (1986) 81–85.

[35] R.M. Parr, E.M. De Maeyer, V.G. Iyengar, A.R. Byrne, G.F. Kirkbright, G. Schoch, L. Niinisto, O. Pineda, H.L. Vis, Y. Hofvander, A. Omololu, Minor and trace elements in human milk from Guatemala, Hungary, Nigeria, Philippines, Sweden, and Zaire: results from a WHO/IAEA joint project, Biol. Trace Elem. Res. 29 (1991) 51–75.

[36] H.M. Mulrine, S.A. Skeaff, E.L. Ferguson, A.R. Gray, P. Valeix, Breast-milk iodine concentration declines over the first 6 mo postpartum in iodine-deficient women, Am. J. Clin. Nutr. 92 (4) (2010) 849–856.

[37] B.G. Biban, C. Lichiardopol, Iodine deficiency, still a global problem? Curr. Health Sci. J. 43 (2017) 103–111.

[38] P. Nazeri, P. Mirmiran, G. Asghari, N. Shiva, Y. Mehrabi, F. Azizi, Mothers’ behaviour contributes to suboptimal iodine status of family members: findings from an iodine-sufficient area, Public Health Nutr. 18 (Mar. (4)) (2015) 686–694. [39] E. Combet, M. Bouga, B. Pan, M.E. Lean, C.O. Christopher, Iodine and pregnancy - a UK cross-sectional survey of dietary intake, knowledge and awareness, Br. J. Nutr. 114 (1) (2015) 108–117.

[40] Y. Fuse, T. Ohashi, S. Yamaguchi, M. Yamaguchi, Y. Shishiba, M. Irie, Iodine status of pregnant and postpartum Japanese women: effect of iodine intake on maternal and neonatal thyroid function in an iodine-sufficient area, J. Clin. Endocrinol. Metab. 96 (12) (2011) 3846–3854.

[41] S. De Leo, E.N. Pearce, L.E. Braverman, Iodine supplementation in women during preconception, pregnancy, and lactation: current clinical practice by U.S. obstetricians and midwives, Thyroid 27 (3) (2017) 434–439.

[42] A. Kut, H. Kalli, C. Anil, U. Mousa, A. Gursoy, Knowledge, attitudes and behaviors of physicians towards thyroid disorders and iodine requirements in pregnancy, J. Endocrinol. Invest. 38 (10) (2015) 1057–1064.

[43] N. Çaylan, B. Tezel, S. ¨Ozbas¸, N. S¸ahin, S¸ Aydın, D. Acıcan, B. Keskinkılıç, Neonatal thyroid-stimulating hormone screening as a monitoring tool for iodine deficiency in Turkey, J. Clin. Res. Pediatr. Endocrinol. 8 (2) (2016) 187–191.

[44] M.F. Erdo˘gan, K. A˘gbaht, T. Altunsu, S. Ozbas¸, F. Yücesan, B. Tezel, C. Sargin, I. Ilbe˘g, N. Artik, R. K¨ose, G. Erdo˘gan, Current iodine status in Turkey, J. Endocrinol. Invest. 32 (7) (2009) 617–622.

[45] M.B. Zimmermann, Iodine deficiency, Endocr. Rev. 30 (4) (2009) 376–408. [46] Food Standards Agency, Retail Survey of Iodine in UK Produced Dairy Foods, FSIS

02/08: 16 June 2008. https://webarchive.nationalarchives.gov.uk/201204 03220603/http:/www.food.gov.uk/multimedia/pdfs/fsis0208.pdf (Access date 04.04.2020), 2020.

[47] O.L. van der Reijden, V. Galetti, I. Herter-Aeberli, M.B. Zimmermann, C. Zeder, A. Krzystek, M. Haldimann, A. Barmaz, M. Kreuzer, J. Berard, P. Schlegel, Effects of feed iodine concentrations and milk processing on iodine concentrations of cows’ milk and dairy products, and potential impact on iodine intake in Swiss adults, Br. J. Nutr. 122 (2) (2019) 172–185.

[48] F. Sch¨one, K. Sp¨orl, M. Leiterer, Iodine in the feed of cows and in the milk with a view to the consumer’s iodine supply, J. Trace Elem. Med. Biol. 39 (2017) 202–209.

[49] L.J. Coneyworth, L.C.H.A. Coulthard, E.H. Bailey, S.D. Young, J. Stubberfield, L. Parsons, N. Saunders, E. Watson, E.M. Homer, S.J.M. Welham, Geographical and seasonal variation in iodine content of cow’s milk in the UK and consequences for the consumer’s supply, J. Trace Elem. Med. Biol. 59 (2020), 126453.

[50] O.L. van der Reijden, V. Galetti, M. Hulmann, A. Krzystek, M. Haldimann, P. Schlegel, E. Manzocchi, J. Berard, M. Kreuzer, M.B. Zimmermann, I. Herter- Aeberli, The main determinants of iodine in cows’ milk in Switzerland are farm type, season and teat dipping, Br. J. Nutr. 119 (5) (2018) 559–569.

[51] L. Kloss, J.D. Meyer, L. Graeve, W. Vetter, Sodium intake and its reduction by food reformulation in the European Union-A review, NFS J. 1 (2015) 9–19. [52] https://www.ign.org/blog-simplify-2018.htm (Access date: 25.05.2020). [53] J.P. Chanoine, M. Boulvain, P. Bourdoux, A. Pardou, H.V. Van Thi, A.M. Ermans,

F. Delange, Increased recall rate at screening for congenital hypothyroidism in breast fed infants born to iodine overloaded mothers, Arch. Dis. Child. 63 (1988) 1207–1210.