Current Iodine Status in Japan: A Cross-sectional Nationwide Survey of Schoolchildren, 2014-2019

Abstract

Context

Japan has been regarded as a long-standing iodine-sufficient country without iodine fortification; however, data on nationwide iodine status are lacking.

Objective

This study aimed to characterize the iodine status in Japan.

Methods

From 2014 through 2019, a nationwide school-based survey was conducted across all districts in Japan. Urinary iodine concentration (UIC), creatinine (Cr) concentration, and anthropometry were assessed in healthy school-aged children (SAC) aged 6 to 12 years. Their iodine status is regarded as generally representative of the nation’s iodine status.

Results

A total of 32 025 children participated. The overall median UIC was 269 μg/L, which was within the World Health Organization’s adequacy range. There was a regional difference in UIC values within 14 regions, and the lowest and highest median UICs were found in Tanegashima Island (209 μg/L) and Nakashibetsu, Hokkaido (1071 μg/L), respectively. The median UIC ≥ 300 μg/L was observed in 12 of 46 regions. By using estimated 24-hour urinary iodine excretion (UIE), the prevalence of SAC exceeding the upper tolerable limit of iodine for Japanese children was from 5.2% to 13.7%. The UIC values did not change with age, body surface area and body mass index percentile, whereas the Cr concentration simultaneously increased suggesting the effect of urinary creatinine on UI/Cr and estimated 24-hour UIE values.

Conclusions

The iodine intake of Japanese people is adequate, but in some areas it is excessive. The incidence and prevalence of thyroid disorders associated with iodine intake should be obtained, especially in the areas where high amounts of iodine are consumed.

Iodine is a fundamental component of thyroid hormones, and adequate intake is essential for maintaining thyroid function. Iodine deficiency has many adverse effects throughout the human life cycle, remains in all regions worldwide, and affects populations at all stages of economic development (1, 2). It is well known that the association of thyroid disorders with iodine intakes appears to be U-shaped at the population level (3). Deficient as well as excessive exposure to iodine may cause thyroid dysfunction. Recommended human biomarkers of iodine status are urinary iodine concentrations (UIC), thyroglobulin, TSH, thyroglobulin autoantibody (TgAb), thyroid peroxidase autoantibody (TPOAb), goiter rate, or thyroid volume; however, the thresholds for iodine sufficiency or excess are established only for a population’s median UIC and thyroglobulin in school-aged children (SAC) (4, 5). To define the national iodine status, the World Health Organization (WHO) recommends using the median of spot UIC in SAC aged 6 to 12 years as a nationally representative sample. Adequate iodine intake corresponds to the median UIC values in the range of 100 to 199 μg/L, and 200 to 299 μg/L has been referred to as “more than adequate.” Recently, the Iodine Global Network (IGN), the successor of the International Council for the Control of Iodine Deficiency Disorders, redefined an adequate iodine intake as in the range of 100 to 299 μg/L of median UIC values including categories previously referred to as “adequate” and “more than adequate” (2). The median UIC ≥ 300 μg/L indicates excessive iodine intake in children and adults, although the term “excessive” means in excess of the amount required to prevent and control iodine deficiency (1). According to the IGN Global Scorecard 2020 on iodine nutrition in the general population, the data of cross-sectional UIC studies have been available from 152 of 194 WHO member states, and, globally, 21 countries still have insufficient iodine and 13 countries have documented excessive iodine in their diets (6). Excessive iodine intakes in populations can result from diets that are naturally high in iodine and/or groundwater (2, 5).

Japan has been regarded as a long-standing iodine-sufficient country or even an iodine-excessive country without iodine fortification because of the regular intake of iodine-rich food when compared with other countries. The issue of iodine nutrition has not been at the forefront of the public health agenda and an official nationwide iodine survey has been never performed. In addition, there is no national surveillance system to monitor iodine intakes; therefore, recent data regarding the population’s iodine status are lacking. The purpose of this study is, for the first time, to assess the current status of iodine nutrition in Japanese by a nationwide school-based survey.

Materials and Methods

This work was a part of the research project “National survey of iodine intake and its relation to thyroid disorders in Japan” conducted by the Japan Thyroid Association starting in 2013. The objective of this project is to evaluate the current national iodine status and to provide information on iodine-related thyroid diseases in Japan.

Japan is an island country situated off the eastern seaboard of the Eurasian continent in the northern hemisphere. It consists of the main islands of Hokkaido, Honshu, Shikoku, Kyushu, Okinawa, and more than 6800 smaller islands of various sizes (Fig. 1). Forestland and fields, agricultural land, and developed land such as residential or industrial land account for 67%, 12%, and 5% of the nation’s surface area (377 975 km²), respectively. More than 50% of the total population (126.17 million in 2019) was concentrated in the 3 major metropolitan areas: the Kanto, Chukyo (Nagoya city), and Kinki areas. There are 47 prefectures, within which there are 1718 municipalities and the 23 cities in metropolitan Tokyo (7).

Subjects

Healthy elementary school children without previous or present history of thyroid disease were recruited in all 47 prefectures of Japan. Because our research project has not been authorized or financially supported by any government-related agency, we directly contacted the boards of education of municipalities in all 47 prefectures across Japan individually, explaining the details of the survey and requesting their cooperation. The local school boards in the cities and towns selected several elementary schools to cooperate with the survey. In some areas, all the elementary schools within the area were included in the survey. Subsequently, a meeting was arranged with the school board members and the directors of the elementary schools. After approval, a letter was sent to the guardians with the informed consent forms attached.

Methods

The guardians were provided with iodine-free polypropylene cups and test tubes. To avoid the effect of iodine in school lunches on urinary iodine concentration, the casual urine samples were obtained in each child’s home the morning after a holiday, before breakfast, and then stored in the refrigerator. The collected urine samples were frozen at –30°C until they were analyzed for iodine and creatinine concentrations.

Body weight and height were measured in school using the standard anthropometric method according to WHO (8), and the data were obtained from the most recent routine health checkup record. The body surface area was calculated according to the formula: body surface area (BSA) (m²) = weight (kg)^0.425 × height (cm)^0.725 × 71.84 × 10^-4 (9), and body mass index (BMI), expressed as kg/m², was calculated with the formula: BMI = weight (kg)/height² (m). The growth charts for Japanese children and teens aged through 17 years by the Japanese Society of Pediatric Endocrinology were used to calculate the SD scores (Z-scores) of weight for age (weight SDS) and BMI for age (BMI SDS) in boys and girls (10). The SD scores of height for ages (height SDS) were formulated by growth charts based on the criteria of the national medical aid program for specific pediatric chronic diseases (11). The SDS is used to evaluate the anthropometric measurements of the recruited children of various ages and both sexes. The BMI for age and sex percentile (BMI percentile) was formulated based on the growth charts for Japanese (12). The BMI-for-age weight status categories suggested by the US Centers for Disease Control and Prevention (CDC); the corresponding percentiles are as follows: underweight, less than the 5th percentile; healthy weight (HW), 5th percentile to less than the 85th percentile; overweight (OW), 85th to less than the 95th percentile; and obesity (OB), 95th percentile or greater (13).

Assays

Urine samples were centrifuged at 3000 rpm for 5 minutes, then 100 μL of the supernatant and 50 μL of internal standard solution were diluted into 5 mL of 0.05% tetramethyl ammonium hydroxide (25%; Wako Pure Chemical Industries, Ltd., Japan). To avoid signal drifting, tellurium and indium (SCP Science, Canada) were applied as an internal standard, and the final concentration of tellurium and indium was 20 and 4 μ/L, respectively. Potassium iodide of JIS special grade, ≧ 99.5% (Wako Pure Chemical Industries) was used as a reference material for iodine. The 1000 ppm of iodide solution was prepared by dissolving 0.131 g of iodine in 100 mL of ultrapure water. The iodine content in urine diluted at a ratio of 1:51.5 with tetramethyl ammonium hydroxide solution was directly analyzed by inductively coupled plasma mass spectrometry, iCAP Q with CASX260 (Thermo Fisher Scientific K.K., Tokyo) at the Central Research Laboratory Tsukuba, Kotobiken Medical Laboratories, Inc., Tsukuba, Japan. For analytical quality control, 2 reference materials, Seronorm Trace Elements Urine L-1 and L-2 (SERO AS, Billingstrad, Norway), were prepared like the samples and measured after every 30th sample. The sensitivity of this assay was 3 μg/L. The intra-assay and inter-assay coefficients of variation were 1.5% to 6.9% and 1.8% to 6.3%, respectively. To ensure the quality of urinary iodine (UI) analyses in the iodine laboratory, Kotobiken Medical Laboratories, Inc., has been participating in the Ensuring the Quality of Urinary Iodine Procedures (EQUIP) program in the CDC since 2015 (14).

The urinary creatinine (Cr) concentration was measured by colorimetric enzymatic assay and the UI was expressed as a concentration in μg iodine per 1000 mL of urine (UIC, μg/L) or relative to Cr excretion (UI/Cr ratio, UI/Cr, μg/gCr). Because the 24-hour urinary iodine excretion (UIE) is regarded as representative of 24-hour iodine intake, the estimated 24-hour UIE was calculated on measurement of iodine and creatinine concentrations in spot urine and using predicted anthropometry-based 24-hour creatinine reference values for healthy well-nourished children, as suggested previously (15). The equation used for Japanese is as follows: for children aged between 2 and 18 years (16), predicted creatinine excretion (mg/24 hours) for boys = 0.94 × height (cm) + 35.01 × weight (kg) – 221 or 116.1 × age (year) – 356; predicted creatinine excretion (mg/24 hours) for girls = 2.74 × height (cm) + 19.57 × weight (kg) – 330 or 77.6 × age (year) – 129. WHO guidelines for iodine deficiency disorder status use “μg/L” for cutoff points of UIC in iodine deficiency disorder surveillance, and 1 μg/L = 0.00788 nmol/L for conversion to SI units.

Statistics

The results were presented as mean, SD, median, range, or interquartile range (IQR). The UIC was distributed asymmetrically and skewed toward high values. Their logarithmically transformed values were therefore used to normalize the distribution. The normality of the transformed data was tested using the Anderson-Darling test, the D’Agostino and Pearson test, and the Kolmogorov-Smirnov test. The differences between paired data or groups were examined using 1-way ANOVA with Tukey multiple comparison test, the Kruskal-Wallis test, or Dunn multiple comparison test. The differences between 2 unmatched groups for normally and nonnormally distributed data were tested using the unpaired t test and Mann-Whitney test, respectively. The significance of differences in prevalence between the groups was determined using the Fisher exact test. Simple linear regression analysis was used to test for correlations between UIC, BSA, and BMI SDS. A P value < 0.05 was considered significant. Data processing and statistical analysis were performed using GraphPad Prism 8.0 from GraphPad Software Inc. (San Diego, CA, USA).

This study was conducted according to the guidelines laid down in the Declaration of Helsinki and the research plan was approved by the Ethics Committee of the Japanese Red Cross Hokkaido College of Nursing (approval numbers and date: No. 170 on December 19, 2013, and No. 29-291 on March 15, 2018, respectively). Written informed consent was obtained from each of the participants’ guardians.

Results

Between February 2014 and December 2019, 162 elementary schools in 49 cities or towns within 32 prefectures participated, in which 51 839 children were included and 32 603 guardians gave their consent to the survey. An additional 39 849 guardians agreed to their own nutrition survey by a food frequency questionnaire (17). A total of 32 025 of 32 603 children aged from 5.9 to 14.9 years provided spot urine samples. After excluding 13 children aged younger than 6 years or older than 12 years, and an additional 42 children with an unknown sex, the total number of subjects aged between 6 and 12 years was 31 970, and the number of males and females were 16 446 and 15 524, respectively. The male-to-female ratio was 1:0.94. There were no differences in the mean age, height, height SDS, and the median urinary Cr concentration between males and females. The mean weight, weight SDS, BMI, BMI SDS, and the median estimated 24-hour UIE and predicted 24-hour urinary Cr excretion in males were significantly higher than those in the females (P < 0.0001) The median UIC and UI/Cr were significantly lower for boys than for girls, and vice versa for estimated 24-hour UIE and predicted 24-hour urinary Cr excretion (Table 1).

Urinary Iodine Concentration in Schoolchildren

The urinary iodine concentrations in spot urine samples obtained at the time of study enrollment were highly variable and the overall median (IQR) UIC was 269 (164-511) μg/L, which was within the WHO’s adequacy range (100-299 μg/L) for SAC. The 2.5 and 97.5 percentiles were 70 and 3820 μg/L, respectively. The median UI/Cr value was 255 μg/gCr and the median value of estimated 24-hour UIE was 190.3 μg/d (Table 1). The proportion of the children for whom the UIC was < 100 μg/L was 8.3% (2436 of 32 025 children). High UIC values exceeding 1 mg/L were found in 3960 of 32 025 children (12.4%).

Distributions of UIC by Region

The numbers of subjects included in each of the 14 regional divisions were 59 to 3696 (mean: 2288) and the male-to-female ratio was 0.88 to 1.46 (mean: 0.99). The mean (SD) values of age, height, weight, BMI, or BSA of the subjects ranged from 9.4 (1.7) to 9.9 (1.7) years, 129.4 (11.8) to 135.8 (12.4) cm, 28.7 (8.6) to 33.3 (10.8) kg, 16.4 (2.4) to 17.9 (3.2) kg/m², or 0.64 (0.16) to 0.73 (0.19) m², respectively. The median UI value in each regional division was calculated by combining data from all the elementary schools in the region together and depicted on the map according to the UIC value (Fig. 1). The regional divisions where the median UIC exceeded 300 μg/L regarded as “excess iodine intake” were Hokkaido including the main island (342 μg/L), Rebun Island (633 μg/L), and Rishiri Island (699 μg/L), and the Hokuriku region (312 μg/L) (Fig. 2, Table 2). The median UIC ≥ 300 μg/L was observed in 12 of 46 cities or towns. The distribution was as follows: the median UIC < 250 μg/L, 12 sites; 250 to 299 μg/L, 22 sites; 300 to 499 μg/L, 7 sites; 500 to 999 μg/L, 4 sites; ≥ 1000 μg/L, 1 site. The lowest and highest median UICs were found in Tanegashima Island (209 μg/L) and Nakashibetsu, Hokkaido (1071 μg/L), respectively (Fig. 1). The distribution of the median UI/Cr and estimated 24-hour UIE were similar throughout all 14 regions (Table 2).

Age-specific Changes of UI and Cr

Age-specific changes of the median UIC, UI/Cr, estimated 24-hour UIE, urinary Cr concentration, and predicted 24-hour Cr secretion in the children between 6 and 12 years are presented in Table 3. There were no significant changes in the median UIC values from children between 7 and 12 years (265-291 μg/L); however, the median UIC in children aged 6 years was 291 μg/L and significantly higher than those in children aged 7 to 11 years. In contrast, the median UI/Cr decreased from 320 to 205 μg/gCr, although the difference in UI/Cr values was not significant among children aged 8 to 10 years. The median values of estimating 24-hour UIE increased in parallel with the increase of urinary Cr concentration and predicted 24-hour Cr excretion with age.

Gender- and Age-specific Differences of UI and Cr

There was no significant change of median UIC values by age in both males and females except for in 6-year-old males whose UIC was significantly higher than that in 9-year-old males (282 vs 254 μg/L, P = 0.0308). The median UI/Cr decreased in both males and females; on the contrary, the median estimated 24-hour UIE, urinary Cr concentration, and predicted 24-hour Cr excretion increased gradually with age (Table 4). When the median values were compared between males and females, in males, the UIC values in those 7 to 12 years old and the UI/Cr values from those 7 to 10 years old were lower than in females. The estimated 24-hour UIE and predicted 24-hour Cr excretion in males was lower than females throughout all ages. The urinary Cr concentration did not significantly differ between males and females except that the females’ levels were higher than the males’ levels between the ages of 10 and 11, although the difference was only 1.8 to 4.1 mg/dL (Table 4).

Relationship of BSA With UI and Cr Values

There was no significant correlation between BSA values and UIC (R²= 0.000005), UI/Cr (R² = 0.001048), or estimated 24-hour UIE (R²= 0.005637) in 31 983 urine samples; the same result was observed when examining the results separately for males and females. The BSA-specific median values of UIC and UI/Cr decreased with an increasing BSA level, whereas the median values of 24-hour UIE, urinary Cr concentration, and predicted 24-hour Cr excretion increased, although there was no significant difference in the median UIC values except between 0.4 and 0.6 (Table 5). As an overall trend, the median UIC, UI/Cr, and estimated 24-hour UIE values did not differ between males and females, or if it did, the males’ values were lower than the females’ values. The median urinary Cr concentration and predicted Cr excretion values were higher in males than in females.

Correlation Between UIC, UI/Cr, and Estimated 24-hour UIE Values

There was a significant positive correlation between UIC and UI/Cr (Spearman r = 0.8320, P < 0.0001) (Fig. 3A), UI/C, and estimated 24-hour UIE (Spearman r = 0.7809, P < 0.0001) (Fig. 3B), or UI/Cr and estimated 24-hour UIE (Spearman r = 0.8839, P < 0.0001) in a total of 32 025 urine samples (Fig. 3C). The equation and R² values for linear regression were as follows: UI/Cr = 0.9031 × UIC + 55.61, R² = 0.7887; UIC = 1.135 × estimated 24-hour UIE + 106.7, R²= 0.7151; UI/Cr = 1.276 × estimated 24-hour UIE + 27.18, R²= 0.8741.

Relationship of BMI and Childhood Weight Status with UI and Cr Values

There was no significant linear correlation between BMI values and UIC (R²= 0.000104), UI/Cr (R² = 0.000852), or estimated 24-hour UIE (R²= 0.003627) in the total number of 31 948 urine samples; the same results were obtained when examined separately for males and females. The subjects were divided into 4 categories of childhood weight status according to the BMI percentile (ie, underweight, HW, OW, and OB). The median values of UIC and UI/Cr of the children with OW or OB were slightly but significantly lower than those of children with HW (258 or 251 vs 271 μg/L) and (227 or 230 vs 259 μg/gCr), respectively, whereas the median urinary Cr concentration, predicted 24-hour Cr excretion and estimated 24-hour UIE values significantly increased with increasing BMI percentile (Table 6). However, there were no significant differences among the 4 categorized groups if the median UIC values were compared separately in males and females.

Discussion

This is the first nationwide iodine status survey conducted in Japan. The overall median UIC value in schoolchildren aged 6 to 12 years was 269 μg/L, and the children with UIC < 100 μg/L was 8.3%, suggesting that current iodine intake in Japanese was adequate, not excessive nor deficient. According to the latest IGN Global Scorecard 2021, 46 of 135 countries with adequate iodine intake have median UIC values between 200 and 299 μg/L. Japan’s median UIC was 265 μg/L and ranked 10th among these 46 countries (6). This reference value was derived from our interim report of the present study (18).

The Japanese population has been regarded as free from goiter; however, there has been a debate on the existence of endemic goiter. The earliest report published in 1899 suggested that there were some cases of local residents with endemic goiter in the coastal area of Hokkaido (18). From 1960 to 1964, the first large epidemiological study was conducted by Suzuki et al. in children residing in 3 goitrous regions on the coast of Hokkaido (ie, Rebun Island, Rishiri Island, and the Hidaka District), suggesting a close relationship between seaweed consumption and prevalence of visible goiter (19). Besides Hokkaido, a large epidemiological study of SAC was conducted in the Hokuriku region between 1972 and 1975, which was the last reported goiter survey in Japan (20). There was no significant difference in the prevalence of goiter between the urban (Kanazawa City) and the coastal (Wajima City) areas (2.7% vs 3.0%).

Since the 1990s, some epidemiological studies on iodine status have been reported in local areas of Japan by using UIC. Interestingly, the UIC values reported in the studies on adults using an iodide-selective electrode were 0.8 to 8.5 mg/L and much higher than the subsequent epidemiological studies using the method based on the Sandell-Kolthoff reaction, although the exact reason is not clear (18). In 2002, we reported the iodine status of SAC aged 6 to 12 years in Hokkaido and Tokyo as part of a WHO project to establish the international reference values for thyroid volume by ultrasound (21). The median UICs in Asahikawa (21), Nakashibetsu of Hokkaido (22), and Tokyo (23) were 288, 728, and 281 μg/L, respectively. Although Tokyo and Asahikawa are both urban areas and the distance between the 2 cities is ~580 miles, their median UICs were very similar. In addition to SAC, we conducted a survey between 2005 and 2012 on the iodine status of children ≥ 13 years, adults (24), and pregnant and lactating women (25, 26) in the metropolitan area including Kanagawa and Chiba Prefectures. The median UIC values were 242 μg/L, 213 μg/L, 219 μg/L, and 135 μg/L, respectively, suggesting adequate iodine intake sustained in Japanese for the past 2 decades by taking into account the results of the present study.

There was a regional difference in the population’s iodine intake across all divisions of Japan. Twelve cities or towns with the median UIC ≥ 300 μg/L were distributed mainly in the Hokkaido and Hokuriku regions (Fig. 1 and Table 2). Although there has been a misconception that most residents in the coastal areas of Hokkaido traditionally consume too much iodine, the iodine intake was excessive in 4 of the 7 sites of the main island of Hokkaido, and 2 of these 4 areas were situated in noncoastal areas. In Nakashibetsu, a dairy farming town in an inland area of Hokkaido, a higher UIC (728 μg/L) associated with high thyroid volume in SAC, which was approximately twice that reported in other countries, was observed in a 2002 survey (22). In the present 2014 survey, the median UIC was still the highest (1071 μg/L). The iodine intake of a population is closely related to traditional local food and lifestyle habits. The major source of iodine in Japanese is seaweed (27) served in a large variety of ways under several different names (eg, Kombu [kelp], Hijiki, and Wakame). Almost all of the Kombu is produced in Hokkaido and transported mainly from the Hakodate area where the UIC value in the present study was 358 μg/L and that is 1 of the 5 major kelp production areas, to various parts of Japan through the so-called “Kombu Road” since the 17th century (28). The Hokuriku region has been one of the main kelp accumulation areas and its annual consumption of Kombu is the largest in Japan as of today. In Rishiri and Rebun Island, Hokkaido, Kombu is the main product, and the custom of eating kelp has long been established. The reason for the high iodine intake observed in certain areas of Japan is unknown and further investigation is needed.

The divergent effects of excess iodine on thyroid function have been reported. Thyroid disorders related to excess iodine intake are hypothyroidism, increased thyroid volume or goiter, thyroid autoimmunity, postpartum thyroiditis, iodine-induced hyperthyroidism, Graves’ disease, and thyroid cancer (4, 5). In rat models, the thyroid gland handles excessive quantities of iodine by inhibiting the organification of iodide by the generation of several inhibitory substances on thyroid peroxidase activity (acute Wolff-Chaikoff effect), although the precise mechanism is not fully understood. This inhibitory effect is often of short duration and once plasma iodide levels drop below a critical level, an “escape” from or adaptation to this block occurs. This adaptation is associated with a marked decrease in expression of the sodium-iodide symporter on thyroid follicular cells. It is believed that if the so-called escape phenomenon does not occur, clinical or subclinical hypothyroidism may develop even in humans, especially in certain susceptible individuals (ie, those with underlying thyroid disease or predisposing risk factors) (4, 5). Several studies (3, 29, 30) and other studies using a meta-analysis (31, 32) have reported a higher prevalence of subclinical and overt hypothyroidism in areas with excessive iodine intake compared with an adequate or deficient intake. Both deficient and excessive iodine intake increase thyroid volume by different etiology and there is a time lag between changes in the goiter prevalence and changes in iodine status (22, 33). A recent review suggests that the risk of goiter increases when the UIC ≥ 300 μg/L or the water iodine concentration ≥ 100 μg/L in mainland China (34). Epidemiological data on the association between thyroid autoimmunity and excessive iodine intake are conflicting and still under debate (5). Iodine-induced hyperthyroidism by excess iodine is frequently observed in euthyroid persons with previous thyroid disease residing in iodine-insufficient areas, usually in later phases of the iodine supplementation program (3).

Except for some case reports and small studies, there were only a few epidemiological studies linking thyroid dysfunction with dietary iodine intake in Japan (19, 20, 35-37). Suzuki et al reported a high visible goiter prevalence rate (2.6%-6.6%) as high as 24% in a seaweed fishing village on Rishiri Island without clinical hyper- or hypothyroidism, and a very high UIE (mean: 23.3 mg/d) combined with a high daily consumption of dried seaweed (10-20 g of Kombu/d) which contains ~2.4 mg iodine/g. In addition, an iodine-restricted diet markedly decreased the size of the goiter in 4 of 50 subjects with goiter. The authors concluded that excessive and long-standing intake of dietary iodine was the major cause of the goiter called “endemic coastal goiter” (19). Although the exact reason was not clear, endemic goiter in the coastal area of Hokkaido seems to have disappeared in the late 1970s, whereas high dietary iodine intake as expressed by high UIE continued during the subsequent 20 years. Inoue et al reported the incidence of chronic lymphocytic thyroiditis in children was higher in the coastal area than in the urban area of the Hokuriku area (0.53 vs 0.14%) without significant difference in goiter rate (20). Nagata et al measured UIC, serum TgAb, and TPOAb in healthy adults residing in 4 geographically separated areas (ie, Yamagata, Okinawa, Nagano, and Hyogo) in 1988. The mean UIC values were from 810 to 1620 μg/L and there was no difference in the rate of positive TgAb or TPOAb among the 4 regions concluding no effect of iodine intake on the TgAb and/or TPOAb positivity (35). Konno et al measured serum TSH and UIC of adults residing in 5 coastal areas of Hokkaido in 1992. The frequency of high UI value (defined as 591 μg/L or greater) correlated with that of hypothyroidism (TSH > 5.0 mU/L) with negative thyroid autoantibody suggesting excessive iodine intake as a possible cause of hypothyroidism in addition to chronic thyroiditis in the iodine sufficient area (36). The UIC in both studies was measured by an electrode method. In 2000, Ishigaki et al reported that the median UIC values by using the Sandell-Kolthoff reaction in 250 SAC of Nagasaki and in 100 adults of Hamamatsu and South Kayabe, Hokkaido were 363 μg/L, 208 μg/L, and 1016 μg/L, respectively. In Nagasaki, only 4 cases of goiter (1.6%) and 2 cases (0.8%) of cystic degeneration without thyroid nodule were detected by ultrasound (37). The relationship between iodine intake and thyroid function or thyroid diseases in Japan cannot be evaluated from the previous studies. In addition, Japan lacks current data on the incidence and prevalence of thyroid disease in the entire or regional population. An epidemiological surveillance together with monitoring iodine intake is required to elucidate whether the existence of relatively higher iodine intake than other iodine-sufficient countries can be regarded as an adverse health consequence.

The WHO and United Nations suggested that the maximal tolerable upper intake level (UL) from all iodine sources of 1 mg/d would be safe for most of the population except those with predisposing risk factors or underlying thyroid disorders (5). The UL is defined as the highest level of nutrient intake (ie, likely to pose no risk of adverse health effects for almost everyone in the general population). Because there was no evidence of increased susceptibility to iodine in children, the ULs in children were derived by adjustment of the adult UL. In children aged 6 to 12 years, the ULs established by the US Institute of Medicine (38) and the European Commission Scientific Committee on Food (39) are 300 to 600 and 250 to 450 μg/d, respectively, whereas in Japan the ULs were set higher than in other countries (ie, 500 μg/d for those 6-7 years of age, 1200 μg/d for those 12-14 years in the 2015 version (40) or 550 μg/d for those 6-7 years, 700 μg/d for those 8-9 years, 900 μg/d for 10-11 years, and 2000 μg/d for those 12-14 years in the 2020 version) (41). In the present survey, the prevalence of SAC with the estimated 24-hour UIE values (UI/Cr × predicted 24-hour Cr excretion) exceeding the UL range were 18.4% in those 6 to 11 years of age and 9.0% in those age 12 years. By using the ULs newly revised in 2020, the prevalence of SAC exceeding the UL decreased to 13.7%, 12.4%, and 11.3% in those age 6 to 11 years and 5.2% in those age 12 years. The prevalence for males was significantly higher than for females across all ages (20.3% vs 14.2% in the 2015 version, 13.6% vs 9.4% in the 2020 version). In other words, 86.3% to 94.8% of subjects consumed iodine below the UL; however, it is not possible to declare the remaining subjects unsafe because the dose-response curve for the iodine is not established (42). To the best of our knowledge, there are few reports on the clinical significance of the prevalence of persons consuming iodine exceeding the UL, and its association with thyroid disorders in a population level is uncertain. The consequences of prolonged exposure to high dietary iodine especially in children are not well understood, and further research is necessary.

A higher median UIC and UI/Cr values in females than in males were observed in SAC aged 6 to 12 years. In females, the lower UIC with higher UI/Cr values has been reported and our results were contrary to previous reports from the United States (43) and Israel (44). There was little difference in urinary Cr concentrations between males and females, and an exact reason for the gender differences in UIC and UI/Cr is unknown. We also observed that UIC values did not change with age, BSA, and BMI percentile, whereas UI/Cr values decreased, and the estimated 24-hour UIE, urinary Cr concentration as well as the predicted 24-hour Cr excretion simultaneously increased suggesting the effect of urinary creatinine on UI/Cr and estimated 24-hour UIE values.

The measurement of urinary iodine in 24-hour collections is the most precise method of estimating daily iodine intake and often considered as the reference standard for a population’s iodine intake and used for validating other methods (45), although no cutoffs of 24-hour UIE values currently exist to interpret a population’s iodine status (46). As an alternative to 24-hour urine samples, we calculated the estimated 24-hour UIE by the validated equations (16). The use of such an equation introduces some uncertainty in the precision and accuracy of the values of daily UI excretion (39, 47) and the amount of daily Cr excretion varies by age, sex, racial/ethnic background, skeletal muscle mass, and protein intake (48, 49). In addition, the urinary Cr concentration may fluctuate according to the storage condition of urine samples (ie, temperature and pH) (48). However, there are several reports on a significant relationship between Cr-adjusted 24-hour iodine excretion in single spot urine samples and observed 24-hour UIE by 24-hour urine collection in children (15) and adults (46, 50, 51); therefore, the estimated 24-hour UIE from spot urine samples may reflect a reasonable iodine status in a population. We observed a highly positive correlation between the estimated 24-hour UIE, UI/Cr, and UIC values, suggesting that the estimated 24-hour UIE and UI/Cr might be of value to assess a population’s iodine status in well-nourished and ethnically homogeneous children even taking into account the variability of age-, BSA-, and BMI-dependent creatinine excretion. However, it is still under debate which parameter is more reliable to assess a population`s iodine status (5, 47).

The effect of BMI or obesity on UIC in children has been reported in iodine-sufficient regions and results are controversial (52-55). In 2 studies from Mexico, a positive correlation of UIC with BMI SDS, height SDS, and prevalence of OW and OB was observed (52), and the OB SAC had higher iodine intake with normal thyroid volume than healthy weight children (415.5 vs 269 μg/d) (53). The authors assume that this association of excessive iodine intake with unhealthy weight could be explained by the availability and consumption of snack foods rich in energy, carbohydrates, fat, and iodized salt in children. In contrast, the study of SAC in Brazil (median UIC: 221.6 μg/L) reported that the risk of excess iodine intake was found to be 36% lower in OW/OB children, although no significant correlation was found between UIC and BMI by age (54). A recent report from Italy suggests lower UIC values in OB children than in adequate weight children in an iodine-sufficient group and BMI as a confounding factor in monitoring the iodine status of SAC (55). In the present study, the median UIC and UI/Cr values in OW and OB children were slightly lower than in HW children, whereas the median estimated 24-hour UIE was positively correlated with childhood weight status. However, this difference of UIC disappeared when the subjects were divided into males and females. The observed changes of UIC, UI/Cr, and estimated 24-hour UIE might be related to urinary Cr concentration that increased with BMI percentile. These conflicting results, including ours, are difficult to explain the impact of BMI on UIC or iodine intake, and further studies are needed.

This study has several limitations. First, we were unable to conduct the survey fully following the WHO-recommended method (56); therefore, we cannot rule out selection bias because of the sample selection method; however, the survey was conducted in almost two-thirds of all prefectures of Japan including major remote islands. The average sample number and male-to-female ratio in 14 regional divisions were 2288 and 0.99, respectively. In Japan, the National Health and Nutrition Survey based on a census has been conducted annually by the Ministry of Health, Labour and Welfare; however, iodine is not included in the micronutrient survey items of the survey. Second, we collected urine samples over a 5-year period; therefore, the data may not capture potential variations in iodine intake (eg, seasonal variations). Third, the estimated 24-hour UIE that we calculated from spot urine samples was not validated with the iodine excretion by 24-hour urine collection. The strength of our research was a large sample size (ie, > 32 000 children participated). In addition, the UIC in all the urine samples was measured continuously by a single skilled laboratory technician using the same instrument of inductively coupled plasma mass spectrometry that is generally considered the gold standard for UI analysis. The quality of UI analyses in the laboratory has been monitored by the EQUIP program of the CDC.

In conclusion, the iodine intake of Japanese people is generally adequate; however, Japan’s median UIC was ranked higher among the countries with the median UIC level between 200 and 299 μg/L. In addition, there were some regions with high iodine intake ≥ 300 μg/L. The incidence and prevalence of thyroid disorders associated with iodine intake must be obtained especially in the areas where high amounts of iodine are consumed.

Abbreviations

Abbreviations
 
• BMI

  body mass index

 
• BSA

  body surface area

 
• CDC

  US Centers for Disease Control and Prevention

 
• Cr

  creatinine

 
• EQUIP

  Ensuring the Quality of Urinary Iodine Procedures

 
• HW

  healthy weight

 
• IGN

  Iodine Global Network

 
• IQR

  interquartile range

 
• OB

  obesity

 
• OW

  overweight

 
• SAC

  school-aged children

 
• SDS

  SD score

 
• TgAb

  thyroglobulin autoantibody

 
• TPOAb

  thyroid peroxidase autoantibody

 
• UI

  urinary iodine

 
• UIC

  urinary iodine concentration

 
• UIE

  urinary iodine excretion

 
• UL

  upper intake level

 
• WHO

  World Health Organization

Acknowledgments

The authors thank the participating children; their caregivers, elementary school teachers, and staff; local medical associations; and boards of education who cooperated in the national survey. We also thank the following people for their efforts in the planning, preparation, and implementation of this survey: Kiyoko Miyamoto (Okayama University); Hideko Ishida (Ishida Clinic, Hiroshima); Yoshiaki Matsuda (Kagoshima University); Kiyoshi Hanamura (Matsumoto City Education Committee, Hanamura Clinic); Shunichi Yamashita (Nagasaki University); Takao Kobayashi (Hamamatsu Medical Center); Kanshi Minamitani (Teikyo University Chiba Medical Center); Sueshi Itou (Tsuruoka Municipal Shonai Hospital); Hiroshi Noguchi and Yasushi Noguchi (Noguchi Hospital, Beppu); Yoshiko Ohe (Nishinomiya Medical Association); Ken Hoki (Hoki Dermatology Clinic, Nishinomiya); Masanobu Yamada and Masatomo Mori (Gunma University); Eiji Takeda (Tokushima University); Tsuyomu Ikenoue (University of Miyazaki); Kunio Higa (Tomigusku Central Hospital); Shigeo Sato (Sato Pediatric Clinic, Iwate); Ikuo Takahashi (Akita Pediatric Association); Kaoru Oikawa (Shimane Pediatric Association); Kazumichi Onigata (Shimane University); Yoshiki Takagi (Kaneryou Seaweed Corporation, Uto); Yuji Takemoto (Osaka Medical Association); Nobutaka Sasaki (Onomichi Medical Association); Muneaki Kiso (Kiso Hospital, Onomichi); Naofumi Akamizu (Wakayama Medical University); Hitoo Nakano (Kyushu University); Tatsuhiko Kawarabayashi (Fukuoka Sukoyaka Health Foundation); Hideaki Kido (Director, Fukuoka Prefectural Board of Education); Ryozo Tadami (Mayor, Maizuru City); Minoru Okazaki (Sado General Hospital); Jinya Ito (Nemuro Municipal Hospital); Takashi Wada and Toshinari Takamura (Kanazawa University, Graduate, School of Medical Sciences and College of Medical); Akio Yamamoto (Saga Medical Association, Yamamoto Memorial Hospital, Imari); Haruo Jyaana (Dean, Hokkaido University of Education); Fujio Kakuya (Furano Hospital, Hokkaido); Eiichi Suzuki (Niigata University Medical and Dental Hospital); Takanori Ueda and Mitsufumi Mayumi (University of Fukui); Shinsuke Kameda and Masahiro Aoshima (Kameda General Hospital).

We thank all members of the committee who aided with collecting data. The members of the study committee “National survey of iodine intake and its relation to thyroid disorders in Japan”, Committee of the Japan Thyroid Association were: Yozen Fuse (Chair, Foundation for Growth Science), Yoshimasa Shishiba (Vice Chair, Foundation for Growth Science), Nobuyuki Amino (Kuma Hospital), Naoko Arata (National Center for Child Health and Development), Shohei Harada (Seitoku University), Tomonobu Hasegawa (Keio University School of Medicine), Mitsuru Ito (Kuma Hospital, Kobe), Yoshiya Ito (Japanese Red Cross Hokkaido College of Nursing), Keiichi Kamijo (Kamijo Thyroid Clinic, Sapporo), Masahiko Kawai (Kyoto University Hospital), Seigo Kinuya (Kanazawa University Graduate, School of Medical Sciences and College of Medical), Naomi Kitatani (Kansai Electric Power Hospital), Yo Kunii (Showa University Northern Yokohama Hospital), Kanshi Minamitani (Teikyo University), Takashi Misaki (Tenri Hospital), Keisuke Nagasaki (Niigata University), Kunihiro Nakada and Mika Tamura (Hokko Memorial Hospital, Sapporo), Hitoshi Noguchi and Yasushi Noguchi (Noguchi Hospital, Beppu), Kazumichi Onigata (Shimane University), Hiromi Shimura and Shinichi Suzuki (Fukushima Medical University), Yoshiyuki Shirai (National Hospital Organization Okayama Medical Center), Asuka Suzuki (Teikyo University), Noboru Takamura (Nagasaki University), Nobu Tsukada (Kagawa Education Institute of Nutrition), Yumiko Urakawa and Mayu Yamaguchi (Kamakura Women’s University), Shinichi Yamashita (Nagasaki University), Kunihiko Yokoyama (Matsuto Ishikawa Central Hospital), Hiroshi Yoshimura and Natsuko Watanabe (Itoh Hospital, Tokyo), and Minoru Irie (Senior adviser, Foundation for Growth Science).

We express our deepest gratitude to the Hokkaido Federation of Fisheries Cooperative Associations for their cooperation in the Hokkaido survey. We also express special thanks to Norihiro Takahashi, Central Research Laboratory, Tsukuba, Kotobiken Medical Laboratories, Inc., who measured the iodine concentration in all the urine samples. We also are grateful to Naomi Iwasaki for drawing figures and Sheryn Mason for assistance in the preparation of the manuscript.

Author Contributions

Y.F., Y.S. and M.I. designed the study, Y.F. acquired the data, performed data analyses, and drafted/approved all versions of the manuscript. Y.I. acquired the data, performed data analyses, and aided with interpretation of the data, and all authors approved the manuscript.

Clinical trial registration

The UMIN (University Hospital Medical Information Network) CTR (Clinical Trials Registry) ID was UMIN000039138.

Disclosures

The authors declare that no competing financial interests exist.

Data Availability

None of the datasets generated during and/or analyzed during the current study are publicly available to preserve the confidentiality of the participants.

References