The median score of the final HEI-2015 in our population was lower than that reported previously in different studies [15, 22, 23]. Another point that should be mentioned is the high frequency (> 80%) of women with worse scores for some components, like “whole grains”, total vegetables” and “dairy” which reflects the poor alimentary habitus from the studied population. It is not surprising, since the studied population is composed from women attended in public basic health care units, where most patients have limited social conditions in the studied region. We believe that it has impact in diet quality since processed food, in detrimental of fresh food, are cheaper and of more assessable for this population. How it impacts the iodine status during the pregnancy lead the authors to propose the present study.
The median UIC in the studied group was within the WHO’s recommended range for pregnant women [11], conferring to the population a status of iodine sufficiency. However, a large variation in UIC, with a high prevalence of samples compatible with inadequate iodine status (either insufficient or excessive/more than sufficient), was detected. Elucidation of the determinants of such variations would improve clinical and socioeconomic strategies to minimize the impact of iodine insufficiency on pregnancy outcomes. Previously, we demonstrated, in this same population, that lower age, higher BMI and multiparity were associated with a higher risk for excessive or more than sufficient iodine status [5]. It was speculated that those women probably had poorer diet quality and higher intake of salt or processed food. Corroborating this hypothesis, it has been reported, by other authors, that women with previous pregnancies were more likely to have a poorer diet quality compared to those with first-time pregnancies [24]. Additionally, previous findings show better diet quality in nulliparous women than multiparous women [24]. In contrast, older women in their first pregnancy and with an adequate BMI should represent women with a higher risk for iodine insufficiency according to previous study [5]. We postulate that this last group would have better self-care and probably diet quality, since they do not have multiparity at a young age and are normal weight women.
The findings in the present study support this hypothesis by showing that insufficient UIC was associated with better diet quality and higher scores of HEI-2015. The components of the HEI-2015 that showed better scores (in relation to adequate group) were “total fruit”, “whole fruit”, “total vegetables” and “refined grains “. Additionally, a significant negative correlation was found between the scores in those three parameters of HEI and UIC. In contrast, excessive UIC was related to worse intake of “whole fruits” and “total vegetables”. As previously noted, data in literature regarding the relationship diet quality and iodine status are scarce [15]. A study evaluating pregnant women in the Navajo Birth Cohort Study (NBCS) demonstrated that they had lower UIC than pregnant from the National Health and Nutrition Examination Survey (NHANES), with higher risk for iodine insufficiency [15]. Curiously they also had better scores in the components of total fruits and whole fruits of the HEI-2015, like the present study [15]. Other components that had better scores among NBCS, in comparison to NHANES, were whole grains and refined grains. This last component was also related to iodine status in our sample [15].
We need to emphasize that not only iodine insufficiency, but also excessive iodine status may negatively impact thyroid function and pregnancy outcomes. In a previous study by our group [10], an increased risk for GDM and HDP was also demonstrated when UIC was above the recommended range. We cannot affirm whether this association is related to diet quality, amounts of salt intake or even thyroid dysfunction related to excessive iodine intake.
In this study, we could not demonstrate a direct association between the component sodium and the UIC; however, we can assume that the low scores in whole fruits and “total vegetables” denote worse diet quality in pregnant women with excessive iodine status, which may be related to the high consumption of processed foods, and consequently salt and sodium. In contrast, better scores in “total fruit”, “whole fruit” and “total vegetables” detected in pregnant women with iodine insufficiency may be related to a better diet quality, which is commonly associated with a low intake of salt and processed food. The HEI-2015 is not an instrument to directly assess the consumption of processed or ultra-processed food, but it is well known that quality of diet is inversely related to the consumption of these kinds of food [25–27]. Similarly to us, a limitation about the application of the HEI-2015, intending to assess sodium intake, was already noticed, since this instrument was not designed to this aim [15]. In pregnant from the NBCS cohort and with lower insufficient iodine status it was also not demonstrated different scores in the component sodium of HEI-2015 in comparison to pregnant women from HNANES and with adequate iodine status [15].
Since 1982, the entire Brazilian population has received a minimum amount of iodine in table salt because of advances in public health policies. On the other hand, it is known that the WHO recommends that the maximum daily consumption of salt be less than five grams per person. However, according to the Brazilian Institute of Geography and Statistics (IBGE), the average salt consumption of Brazilians is 12 grams daily, a value that exceeds twice the recommended [23, 25–27]. Because of this, a determination was published by the National Health Surveillance Agency (ANVISA) in the Official Gazette of the Union (DOU) of April 25, 2013, changing the iodation range of salt used in Brazil. According to the new rule, the addition of iodine in table salt ranged from 20–60 to 15–45 milligrams per kilo (mg/kg) of salt [12]. The same is equivalent to 150–450 µg per 10 grams of salt. It is important to note that a previous study revealed that 98.5% of the table salt analyzed from that cohort had iodine concentration compatible with governmental recommendations (20–60 mg/kg) [5].
It is important to study other pathways to explain diet quality impact on thyroid function, not solely related to iodine consumption. Limited consumption of fruit and vegetables was associated with an increased risk of ‘high total lipid peroxide levels in serum’, which could be related to autoimmune diseases, such as Hashimoto’s thyroiditis (HT) [24–27].
It must be acknowledged that our study has some limitations. First, as is true with all cross-sectional observational studies, causal inference is not possible. Diet quality may vary according to socioeconomic status, and as such, dietary patterns may vary substantially between high and low socioeconomic countries. Additionally, this study did not assess the amount of iodine intake based on the ingested food. In Brazil, the food composition tables do not include iodine amount evaluation. Based on the impossibility to adequately assess the amount of iodine intake, according to the ingested food, this study focused on diet quality. The main objective was to observe if there is any relationship between diet quality and iodine status.
Another important limitation of the present study is related to fact that the results were gathered 5–7 years ago. However, supposing that the major source of iodine intake in our country is related to the salt intake and that there was no change in the governmental requirement to iodine-fortification since that time, we believe that the data present in the manuscript are still actual. Furthermore, the present study reinforces the necessity of additional investigations regarding salt intake and maybe a sodium-focused food frequency questionary. Lastly, this study was conducted before the COVID-19 pandemic, that is a well-known factor that led to changes in nutritional behaviors [28–30]. However, the major changes in nutritional behaviors during the COVID-19 pandemic were related to the lockdown [28–30], what occurred only in the earliest moments of pandemic in Brazil.