Iodineminho Study: Iodine Supplementation and Prevalence of Iodine Deficiency In Pregnant Women

Abstract

Context

Iodine is necessary for proper brain development. The prevalence of iodine deficiency in Portuguese pregnant women led the health authorities, in 2013, to recommend iodine supplementation for women in preconception, throughout pregnancy, and during lactation.

Objective

This work aimed to assess the effect of iodine supplementation initiated in the preconception or the first trimester of pregnancy on the prevalence of iodine deficiency and maternal thyroid status.

Methods

An observational prospective cohort study was conducted that followed the thyroid function and iodine status of women recruited during preconception or in the first trimester of pregnancy.

Results

Median urinary iodine concentration (UIC) was significantly higher among women taking iodine supplements (no-supplement group UIC = 63 µg/L; supplement group UIC = 100 µg/L; P = .002) but still below the levels recommended by the World Health Organization. Only 15% of pregnant women had adequate iodine status and 17% showed a UIC of less than 50 µg/L. There was no influence whether iodine supplementation was started during preconception or during the first trimester of gestation (UIC preconception group: 112 µg/L vs UIC pregnancy group: 91 µg/L; P = .569). In the first trimester of pregnancy, total thyroxine levels were lower and free triiodothyronine levels were higher in nonsupplemented women. Thyroglobulin levels were lower in women who started iodine supplementation during preconception compared to nonsupplemented women and women who started iodine supplementation during gestation.

Conclusion

In the Minho region of Portugal, fertile women have insufficient iodine intake. Additional public health measures are needed since the current recommendations for iodine supplementation for pregnancy are unsatisfactory to achieve adequate iodine status.

Iodine is an essential micronutrient to produce thyroid hormones, which regulate metabolism, growth, and development. Maternal iodine status is particularly important during the first trimester of pregnancy when the fetal thyroid status relies entirely on maternal iodine and thyroid function. The fetal thyroid gland starts producing thyroid hormone in the second trimester, but the maternal thyroxine (T4) contribution remains relevant throughout gestation (1).

Iodine deficiency is considered the most important preventable cause of brain impairment. Severe gestational iodine deficiency has been associated with detrimental and irreversible consequences for the offspring, including hypothyroidism and impaired brain development (2). Iodine deficiency is also associated with a greater risk of stillbirth, delayed physical development, and impaired mental function (3). The consequences of mild-to-moderate iodine deficiency in pregnancy for fetal development are less clear (4, 5).

The World Health Organization (WHO) guidelines advise that pregnant and lactating women have an iodine intake of 250 µg/day. During pregnancy, a median urinary iodine concentration (UIC) between 150 and 249 µg/L defines a population with adequate iodine intake (3).

Despite worldwide improvements during recent decades, iodine deficiency still prevails in various countries, particularly in pregnant women. Studies in Portugal have shown that in the Minho region the median UIC indicated iodine deficiency in women and their progeny (6). It was also observed that free T4 (FT4) levels in the first trimester were lower than expected (7) and that this parameter predicted the psychomotor development of the progeny (8). In a subsequent study in the whole country, adequate UIC (>150 µg/L) was observed in only 17% of women in mainland Portugal, 8.2% in the Madeira islands, and 2.3% in the Azores islands (9).

Strategies to overcome iodine insufficiency in the population include recommendations on the fortification of commonly used food sources, such as the iodization of salt, or supplementation, specifically through pills, for pregnant women (3). In 2013, the Portuguese health authorities recommended iodine supplementation in the preconception period, throughout pregnancy, and during lactation in the form of potassium iodide, 150 to 200 µg/day (10). In the present report we assessed the effectiveness of this health policy recommendation on the prevalence of iodine deficiency and on maternal thyroid status.

Material and Methods

Study Design and Participants

The IodineMinho Study is an observational, prospective cohort study that monitors women from preconception/beginning of pregnancy to delivery and then mother/child pairs until age 12 months. The overall study design and aims, as well as the sample size required, have been registered and published (ClinicalTrials.gov, NCT04288531) (11). The study was authorized by national (National Data Protection Commission authorization No. 11200/2017), regional (ACES Cávado I, Regional Health Administration of the North Committee on Health Ethics, opinion No. 65/2018), and local (Braga's Hospital Ethics Committee reference 16/2018) authorities, and the participants signed informed consent. The study complies with the Declaration of Helsinki. The study was performed in compliance with the STROBE (Strengthening the Reporting of Observational Studies in Epidemiology) guidelines.

Women in the present study were enrolled from 10 representative Family Health Units from the urban and suburban areas of the Minho region in the North of Portugal, between January 2020 and August 2022. Women in preconception or in the first trimester of pregnancy (confirmed according to amenorrhea) and intending to deliver at Braga Hospital were enrolled on providing informed consent. Women were excluded if they moved out of the Family Health Units where they were recruited or if gestational age was further than 13 weeks at the time of blood and urine collection. Women with thyroid disease were excluded only if they were treated with thyroid-related medication.

Demographic and clinical information were collected: age and self-reported current and previous pathologies, including thyroid diseases; parity and previous pregnancies; smoking and alcohol consumption; and folic acid and iodine supplementation. The Graffar socioeconomic and cultural scale was applied; it provides a scale from I to V: class I corresponds to families with high education, well-remunerated jobs, high-standard houses, and affluent neighborhood; class V corresponds to the lower end of the spectrum in these indicators. Height, preconception weight, and body mass index (BMI) (calculated as weight in kg/height in m²) were recorded. Dietary habits were evaluated by a food frequency questionnaire at recruitment. To obtain food consumption, the frequency reported for each item was multiplied by the respective standard average portion in grams and by a seasonal variation factor for foods consumed in specific seasons (0.25 was considered for the average seasonality of 3 months). Foods were converted into nutrients using the Food Processor Plus computer program (ESHA Research). The nutritional information was taken from the US Department of Agriculture's food composition tables and adapted to typical Portuguese foods (12, 13). Thyroid ultrasonography was performed at recruitment by the same clinician using a real-time sonography (General Electrics Vivid E95), with a 4.5- to 12-MHz linear array transducer. Thyroid volume was calculated as the sum of maximal length × width × depth × 0523 of each lobe (isthmus not included). The presence of goiter was defined as thyroid volume greater than 18 cm³, according to Gutekunst et al (14).

Women were considered iodine supplemented if they reported intake of a daily tablet containing 150 to 200 µg of iodine, as recommended in national guidelines (10), through one of several commercially available tablets containing potassium iodide (isolated or in multivitamin formula).

Fig. 1 depicts the study flowchart. Study sample 1 includes women at recruitment, characterized as in preconception or already pregnant. Study sample 2 includes only pregnant women during the first trimester, considering all women who were pregnant at recruitment plus those who became pregnant thereafter. In study sample 1, we considered as iodine supplemented women who started iodine supplementation before recruitment; pregnant women in study sample 2 were considered iodine supplemented if they had started iodine supplementation before urine sample collection during the first trimester.

Reported iodine supplementation was the exposure variable. UIC was the dependent variable. The following independent variables were considered as potential confounders: maternal age (15), smoking habits (smoking is considered a goitrogenic substance since it inhibits the absorption of iodine during gestation and lactation) (16), preconception BMI (15), dietary iodine intake, and use of iodized salt.

We further compared the present data with that of a former study of 140 pregnant women in the first trimester, performed before the current iodine supplementation guidelines (recruited 2003-2005), and therefore not taking iodine supplements (6, 10).

Subclinical hyperthyroidism was defined as a condition with normal FT4 and thyrotropin (TSH) below normal reference levels (0.55-4.78 mIU/L) and subclinical hypothyroidism as a condition with normal FT4 and TSH above normal reference levels (0.55-4.78 mIU/L).

Sample Collection and Measurements

Spot urine samples (∼30 mL) for iodine and creatinine were collected at recruitment and during the first trimester of pregnancy. Creatinine levels were determined to account for the urine dilution. UIC was measured using a spectrophotometric assay based on the Sandell-Kolthoff reaction (17). The results were expressed as microgram of iodine per liter of urine (µg/L). Urinary creatinine was measured by the kinetic Jaffé reaction (18) with direct colorimetric detection. According to the WHO, UIC less than 150 μg/L was considered an indication of iodine insufficiency, UIC less than 50 µg/L as moderate to severe iodine insufficiency, and UIC greater than or equal to 500 μg/L as excessive iodine intake (3).

Maternal blood samples for TSH, thyroid hormones (total T4 [TT4] and FT4 and total and free triiodothyronine [TT3 and FT3]), thyroglobulin (Tg), thyroxine-binding globulin (TBG), human chorionic gonadotropin, antithyroid peroxidase (anti-TPO), and antithyroglobulin (anti-Tg) antibodies were collected at recruitment (preconception or pregnancy) and during the first trimester of pregnancy. Venous blood was collected using anticoagulant-free tubes and, after centrifugation, serum was stored at –80 °C until further analyses. Serum levels of TT4, FT4, TT3, FT3, TSH, Tg, anti-TPO, and anti-Tg antibodies were measured by qualitative chemiluminescence immunoassay, using a Siemens Atellica IM. TBG and human chorionic gonadotropin were measured by qualitative chemiluminescence immunoassay, using a Siemens Immulite 2000 Analyzer. Normal reference levels are TSH = 0.55 to 4.78 mIU/L; TT4 = 4.5 to 10.9 µg/dL; FT4 = 0.70 to 1.58 ng/dL; TT3 = 0.60 to 1.81 ng/mL; FT3 = 2.3 to 4.2 pg/mL; Tg less than 55 µg/L; TBG = 13 to 39 µg/mL; anti-TPO less than or equal to 60 IU/mL; and anti-Tg less than or equal to 4.5 IU/mL.

Statistical Analysis

The normality of the distribution assumption was assessed visually through histograms, using the Kolmogorov-Smirnov and Shapiro-Wilk tests and the skewness and kurtosis measures. Accordingly, data with normal distribution are presented as mean ± SD or median with interquartile range (IQR) if the normality assumption was not accomplished. To compare the mean of the 2 groups, a 2-tailed independent-samples t test was applied, when data was normally distributed, and the Mann-Whitney U test, when data failed this assumption. The Fisher exact test was applied to compare the distributions of 2 nonstratified categorical features and the independent chi-square test for comparing categorical variables. A 2-way analysis of variance was used to assess the interaction effects of iodine supplementation and the preconception/pregnancy moment at recruitment on UIC levels. To check for selection bias, recruited women were compared with all pregnant women in 2021, and with a subsample of pregnant women from 2020 and 2022, from the recruitment sites. A Kruskal-Wallis test was used to compare the educational level (ordinal variable–categorized as primary, secondary, and higher education), and a one-way analysis of variance to assess the mean age difference between groups.

A binary logistic regression analysis was performed to test the association between adequate or insufficient UIC levels and intake of iodine supplements. There was complete information for 89 pregnant women (65 with UIC <150 µg/L and 24 with UIC >150 µg/L); this population was identical to the original regarding demographic and clinical variables. Statistical significance was defined by a P value less than .05. Statistical analyses were performed using IBM SPSS Statistics (version 27) (IBM Corp).

Results

In the 10 Family Health Units, 3027 pregnancies were registered during the 3 years of recruitment (1102 in 2020, 984 in 2021, and 941 in 2022). From all women that were eligible for the study (N = 3027), a subsample of 1715 was randomly selected (stratified by Family Health Unit and by year of inclusion) to assess for selection bias. No differences were found with respect to age or education level between these women and those enrolled in the study, which supports generalization. A total of 168 women were invited to participate in the study, of whom the following were lost/excluded: 21 due to dropout or communication failure, 3 for not having a biological fluid collection at recruitment, and 4 for receiving levothyroxine treatment; and 14 (10%) due to spontaneous miscarriage in the first trimester. Remained in the study 126 women: a total of 30 (24%) were recruited during preconception and 96 were pregnant (76%) before 13 weeks of gestation (study sample 1) (see Fig. 1).

At recruitment, 94 (74%) women were taking iodine supplements: 17 (57%) during preconception and 77 (80%) were already pregnant.

Table 1 presents the demographic and clinical data at recruitment. The mean (±SD) age was 32 (±5) years, 10 (8%) women had thyroid disease (6 with thyroid dysfunction and 2 with thyroid nodules and no iodine supplementation and 2 with nodular pathology and taking iodine supplementation), and 11 (9%) were anti-TPO antibodies positive. Socioeconomic status (Graffar scale), parity and previous pregnancies, previous miscarriages, smoking and alcoholic habits, and preconception BMI did not differ among women taking or not taking iodine supplementation.

Food frequency questionnaires (n = 105) revealed a median (IQR) estimated dietary iodine intake of 59 (24-87) µg/day and a median (IQR) estimated dietary selenium intake of 100 (79-132) µg/day.

Table 2 presents information on thyroid function. Nonpregnant women taking iodine supplements had a higher median UIC than nonpregnant women without iodine supplementation (UIC = 84 µg/L vs UIC = 65 µg/L, respectively; P = .021). Similarly, pregnant women on supplementation had higher UIC than nonsupplemented pregnant women (UIC = 90 µg/L vs UIC = 66 µg/L, respectively; P = .015), but still below that recommended for iodine sufficiency. Nonpregnant supplemented women had a higher FT4 and lower Tg than nonsupplemented nonpregnant women. Other biochemical parameters and thyroid volume did not differ between supplemented and nonsupplemented women. Ten percent of women at recruitment presented with goiter.

Table 3 shows the demographic and clinical data of all pregnant women during the first trimester (study sample 2) (see Fig. 1). No differences concerning socioeconomic status, parity and previous pregnancies, previous miscarriages, smoking and alcoholic habits, and preconception BMI were observed according to iodine supplement intake, but the presence of thyroid disease was higher in the nonsupplemented group. Among supplemented women, only 6 lacked information of the timing of initiation of iodine supplementation.

During the first trimester of pregnancy, the median UIC was significantly higher in supplemented women (UIC = 100 µg/L vs UIC = 63 µg/L; P = .002). Still, there was no difference between iodine supplementation started during preconception or the first trimester of gestation (UIC = 112 µg/L vs UIC = 91 µg/L, respectively; P = .569) (Table 4). TT4 levels were lower and FT3 were higher in nonsupplemented women. Tg levels were lower in women who started iodine supplementation during preconception compared to nonsupplemented women and to women who started iodine supplementation during gestation. Five women (5%) presented with subclinical hyperthyroidism, all but one taking iodine supplements and all with UIC less than 150 µg/L.

Considering iodine status categories based on median UIC (3), only 15% of pregnant women had a UIC greater than 150 µg/L, whereas 17% had a UIC lower than 50 µg/L (Table 5).

As shown in Table 6, the use of iodine supplements was the only factor associated with UIC adequacy. Iodine supplementation was associated with an increased likelihood (4.4 times) of having adequate UIC levels. Age, preconception BMI, smoking habits, or dietary iodine intake were not associated with UIC levels.

Regarding the formula of iodine, 46% reported the intake of isolated potassium iodide and 54% multivitamins with iodine. Median UIC and other thyroid parameters were identical irrespective of the type of supplement used (data not shown).When these study data were compared with that reported in 2009, before the recommendation of iodine supplementation, no differences between noniodine-supplemented women could be observed (2022 UIC = 63 µg/L vs 2009 UIC = 65 µg/L). Those on iodine supplementation in the present study had a higher median UIC (100 µg/L). Noticeably, altogether, the percentage of women with iodine sufficiency was not significantly different in 2009 and 2022 (11% vs 15%, respectively).

Discussion

The present study shows that 74% of women report compliance with the recommendation issued by the national health authorities on iodine supplementation during preconception and pregnancy. Only 31% of first trimester pregnant women started iodine supplementation during preconception. However, even iodine supplementation during preconception failed to increase the median UIC to the recommended levels for pregnancy. Only 17% of supplemented women displayed adequate UIC, which remained unchanged between 2009 and 2022, despite the health policy recommendations for iodine supplementation issued in 2013. Our results are supported by data published in recent Portuguese studies, with worrisome low UIC in iodine-supplemented pregnant women (19, 20). This suggests that, in the Minho region and probably in the whole country, fertile women have inadequate iodine intake. Current recommendations for iodine supplementation during pregnancy may still be insufficient for this particularly vulnerable population group to achieve an adequate iodine status.

The lack of effect of iodine supplements can be related to many factors, mainly to the time of administration of iodine supplementation and the physiologic response to iodine repletion, which may be different depending on the degree of iodine deficiency before treatment (21). A few studies investigated the benefits of starting iodine supplementation before pregnancy, showing that higher iodine levels are achieved with longer treatment prior to pregnancy. In a Spanish study, women who started supplementation with iodized salt 1 year or earlier before pregnancy had a higher UIC and smaller thyroid gland volume in the third trimester than women who received 200 or 300 µg/day iodine supplements during pregnancy (22). In Australia, women who started a 150 µg/day iodine supplementation before conception had UIC within the adequate range, unlike those who started iodine supplementation during pregnancy (23). Two smaller studies in Italy also showed that iodized salt consumption 2 years before conception and iodine supplementation with 150 µg/day before pregnancy were associated with lower TSH levels and lower prevalence of isolated hypothyroxinemia when compared with iodine supplementation started during pregnancy (24, 25). With the current evidence, the American Thyroid Association guidelines recommend low-dose iodine supplements starting 3 months before pregnancy in women living in mild to moderate iodine-deficiency regions (26). Other possible explanations for the observed iodine insufficiency are overreporting of compliance with supplement intake and the amount of iodine actually available in the pills may be lower than that presented on the label, as has been shown in several multivitamin formulas (27).

In the present study, the number of women recruited at preconception was small, and most women started supplementation 12 weeks before pregnancy. Median UIC and thyroid hormone parameters did not differ depending on the time of iodine supplementation initiation, except for lower Tg levels in pregnant women who started iodine supplementation during preconception. This interesting finding adds to the discussion on which parameter better reflects individual iodine status. While UIC is highly affected by recent iodine intake and, as such, cannot be used as an indicator of overall individual iodine status, Tg is rather an indicator of a long-lasting (chronic) homeostatic status and/or thyroid function. As such, Tg has been suggested as an alternative marker for chronic iodine deficiency, but a reference interval related to iodine status in pregnancy has not yet been established (28).

The WHO, the United Nations International Children's Emergency Fund, and the Iodine Global Network (formerly the International Council for Control of Iodine Deficiency Disorders) currently recommend a daily dietary iodine intake of 250 µg, which, in iodine-deficient regions, can be provided through universal salt iodization or administration of iodine supplements (3). The widespread evidence of iodine deficiency in developed countries has forced medical societies in Europe and North America and health authorities in several countries, including Portugal, to recommend that all women during preconception, gestation, and lactation take a daily tablet containing at least 150 µg of iodine (10, 26, 29-31). In Portugal, the recommended daily dose of iodine supplementation in pregnant women is 150 to 200 µg. The data of the present study suggest that iodine supplementation during preconception and pregnancy failed to ensure iodine sufficiency. Other measures may be necessary, such as universal policies of iodized salt, which is not mandatory in Portugal. We have recently showed that sales of iodized salt in Portugal are scarce and, as such, so is the estimated iodine intake through salt (32, 33). The women in the present study, as inferred from reports in food questionnaires filled out at recruitment, had a median (interquartile range) estimated dietary iodine intake of 59 (24-87) µg/day, which is low.

As for selenium, reported information suggests adequate intake in our population. This is of relevance given the role of selenium in thyroid hormone homeostasis; its deficiency can exacerbate the effects of iodine deficiency (34).

The results of this study should raise awareness among health authorities about the potential deleterious effects of iodine insufficiency on the health of the offspring (35, 36). Follow-up of the IodineMinho cohort will provide additional data on thyroid hormone parameters throughout gestation, and its effect on child iodine levels and psychomotor development (11). Studies on the effects of iodine supplementation during pregnancy in mild to moderate iodine-insufficient areas, as is the case of Portugal, on early childhood neurodevelopment have reported inconsistent findings (4, 37-39). Furthermore, few studies have examined the role of preconception iodine supplementation on birth outcomes and child neurodevelopment (40). The IodineMinho study will contribute to the elucidation of these questions.

The main limitation of this study pertains to the self-reported nature of the information on the use of iodine supplements since compliance is not guaranteed. This poses the risk of misclassification; if present, it is likely nondifferential. Another limitation is the interpretation of the median UIC of small subgroups, since it is more affected by outliers and provides larger confidence intervals. Furthermore, UIC is not validated for classifying an individual's iodine status, as a single sample may not be representative of general iodine status.

In summary, the present study shows that women intending to become pregnant and pregnant women in the Minho region of Portugal are mildly iodine deficient and that current iodine supplements are not sufficient to meet the increased iodine needs in pregnancy. These data call for awareness on the part of public health authorities of the need for additional measures to ensure adequate iodine intake for pregnant women in Portugal.

Acknowledgments

The authors are thankful to Leonor Pereira and Rui Macedo for their help in retrieving information from the nonrecruited women. The authors are thankful to Unilabs, Germano de Sousa, Dr Hilário de Lima, Dr Joaquim Chaves, Trofa Saúde Braga, Synlab, and S. Lázaro Laboratories, which participated in biological specimen collections. The authors are thankful to the clinicians and nurses at the Family Health Units and Braga's Hospital and to the participants.

Funding

This work has been funded by Portuguese National funds, through the Foundation for Science and Technology (FCT) - project UIDB/50026/2020 (DOI: 10.54499/UIDB/50026/2020), UIDP/50026/2020 (DOI: 10.54499/UIDP/50026/2020) and LA/P/0050/2020 (DOI: 10.54499/LA/P/0050/2020) and by the project NORTE-01-0145-FEDER-000039, supported by Norte Portugal Regional Operational Programme (NORTE 2020), under the PORTUGAL 2020 Partnership Agreement, through the European Regional Development Fund (ERDF). S.M. is the recipient of a fellowship (2022.12249.BD) from FCT. S.R. has a junior researcher contract under DL 57/2016/CP1377/CT0128 (DOI: 10.54499/DL57/2016/CP1377/CT0128).

Author Contributions

M.L.P.: methodology, investigation, formal analysis, writing—original draft; S.R.: methodology, resources, investigation, data curation, visualization, writing—review and editing; S.M.: investigation, writing—review and editing; T.I.M.K.:—investigation, writing—review and editing; A.Q.: investigation, writing—review and editing; A.M.: investigation, writing—review and editing; L.V.: investigation, writing—review and editing; M.C.N.: investigation, writing—review and editing; M.R.G.: investigation, writing—review and editing; A.B.: investigation, writing—review and editing; P.C.: formal analysis, writing—review and editing; and J.P.: conceptualization, methodology, supervision, project administration, funding acquisition, and writing—review and editing. All have contributed to the final approval of the version to be submitted.

Disclosures

The authors declare no conflict of interest.

Data Availability

The data generated in the current study will be available from the corresponding author on reasonable request.

Clinical Trial Information

ClinicalTrials.gov ID NCT04288531 (registered at February 28, 2020).

References

Abbreviations

 
• BMI

  body mass index

 
• FT3

  free triiodothyronine

 
• FT4

  free thyroxine

 
• IQR

  interquartile range

 
• T4

  thyroxine

 
• TBG

  thyroxine-binding globulin

 
• Tg

  thyroglobulin

 
• TPO

  thyroid peroxidase

 
• TSH

  thyrotropin

 
• TT3

  total triiodothyronine

 
• TT4

  total thyroxine

 
• UIC

  urinary iodine concentration

 
• WHO

  World Health Organization