Low-Glycemic Load Diets and Thyroid Function: A Narrative Review and Future Perspectives

1. Introduction

Overt thyroid dysfunction, such as hyperthyroidism and hypothyroidism, results in significant alterations of thyroid hormones that unequivocally affect body weight. However, individuals with an intact hypothalamic-pituitary-thyroid axis (HPT) may experience fluctuations in thyroid function tests that persist within the normal reference range under specific circumstances with uncertain clinical significance. It appears that nutritional factors and extreme fluctuations in calorie intake, such as those encountered during periods of starvation or overfeeding, could influence thyroid function tests. Furthermore, in euthyroid subjects, obesity can lead to subtle yet definite alterations in thyroid function, with an unclear impact on body weight. These observations suggest that dietary patterns may influence thyroid function in individuals without HPT axis pathology. Recently, Low-Glycemic Load (LGL) diets have gained popularity and shown promise in the treatment and/or prevention of various medical conditions, including diabetes, obesity, cardiovascular disease, and epilepsy. In this review, we aim to explore the challenges associated with nutritional management using LGL diets and their impact on thyroid function, both in health and disease.

2. Thyroid Hormones Physiology

Thyroid hormones (THs) encompass 3,3′,5-triiodo-L-thyronine (T3) and thyroxine (T4) [1]. While circulating T4 is exclusively synthesized and released by the thyroid gland, approximately 20% of circulating T3 originates from the thyroid gland, with the remaining 80% resulting from deiodination of T4 in the peripheral tissues, via deiodinase (DIO) enzymes [2]. These enzymes, namely DIO1, DIO2, and DIO3 play a crucial role in modulating circulating and intracellular THs concentrations. Possessing a selenocysteine group, they catalyze the conversion of T4 to either T3 or reverse T3 (by removing an iodine atom), generating active and inactive forms, respectively [3]. More than 99% of circulating THs are bound to specific binding proteins; the thyroxine-binding globulin (TBG), transthyretin (TTR), and albumin. Unbound THs enter their target cells through specific membrane transporters such as the monocarboxylate (MCT) family and organic anion transporters (OATPs) [4]. T3 exerting its effects by binding to nuclear receptors, specifically either Trα or TRβ. These receptors are distributed widely throughout the body's tissues, enabling T3 to regulate numerous gene expressions in a genomic manner. In infants and children, THs play a crucial role in brain development and overall growth. In adults, THs significantly influence skeletal integrity, cardiovascular function, and metabolism. Notably, THs increase basal metabolic rate and thermogenesis [5]. and regulate carbohydrate [6] and lipid homeostasis [7]. Additionally, non-genomic actions of THs, mediated by T4 binding to cell membrane receptors, have been elucidated [4]. The production of THs is regulated by the HPT axis, as illustrated in Figure 1. Hypothalamic thyrotropin-releasing hormone (TRH) stimulates the synthesis and release of pituitary thyrotropin (thyroid-stimulating hormone, TSH), which, acting through its receptor (TSHR) on follicular thyroid cells, initiates and facilitates all stages of THs biosynthesis and secretion. Subsequently, THs enter the brain and inhibit the synthesis and secretion of TRH and TSH to maintain their concentrations within the normal range. Additionally, various neural, humoral, and local factors modulate the hypothalamus-pituitary-thyroid (HPT) axis and, under specific circumstances, contribute to modifications in the axis's physiological function [8]. Excessive or deficient production of THs by the thyroid gland results in hyperthyroidism or hypothyroidism, respectively. The main causes of hyperthyroidism include Graves' disease or toxic adenomas, while hypothyroidism can result from Hashimoto's thyroiditis or thyroidectomy. Both conditions can be categorized into subclinical forms, characterized by changes in TSH and thyroid hormones (THs) within the normal range, or overt forms, which involve alterations in both TSH and THs. It is estimated that in the general population the prevalence of thyroid function alterations is ~0.5-4% in areas with sufficient dietary iodine intake [9].

3. Methodology

In this narrative review, a comprehensive search of the PubMed®, Web of Science®, and Google Scholar® databases was conducted from July to December 2023. Additionally, a manual search was performed to identify articles not available in these databases. The search terms employed were: "Glycemic Index," "Glycemic Load," "Glycemic Load Diet and Thyroid," and "Glycemic Index and Thyroid." The scope of our investigation encompassed studies, review papers, systematic reviews, and meta-analyses published within the past 10 years leading up to the search date. Only articles published in English were considered for inclusion in our study. Considering limited findings, the search was expanded to include relevant literature from preceding years.

4. Effects of Nutrition, Starvation/Overfeeding, and Obesity on Thyroid Function in Euthyroid Subjects

4.1. Nutrition

Nutrition and calorie intake is inextricably linked with thyroid function. Adequate thyroid hormone production, in subjects with intact HPA axis, depends on sufficient dietary iodine intake. THs are iodinated products of the amino acid tyrosine, so iodine is an essential element for their synthesis [1]. Iodine deficiency has major health consequences. During pregnancy iodine requirements increase significantly and according to the Systematic Review and Meta-Analysis of Candido et al. positive effects were seen from daily iodine supplementation (200 μg/d) on iodine status and maternal thyroid concentrations, especially prior or in early pregnancy stages [10]. In infants and children iodine deficiency causes irreversible mental retardation and growth disorders while in adults causes goiter and other symptoms of hypothyroidism [11]. Despite a global application of iodine supplementation programs, such as salt iodination over the last decades, iodine deficiency remains a major public health problem in few areas [11]. On the other hand, excessive iodine intake could lead to thyroid dysfunction manifesting as either hyperthyroidism or hypothyroidism, particularly in individuals with preexisting subclinical thyroid disorders. This underscores the importance of maintaining appropriate iodine intake [12]. Additionally, among various nutritional factors, certain trace elements such as selenium (Se), zinc (Zn), and iron (Fe) play a significant role in influencing thyroid function, and their deficiency can critically impact thyroid hormone (TH) homeostasis. Notably, obese individuals have been observed to exhibit lower circulatory levels of Zn and Se, along with abnormal thyroid function.In a recent systematic review by Zavros et al, encompassing 13 randomized controlled trials, it was found that Zn supplementation affects thyroid function by increasing free triiodothyronine (FT3) levels. However, the effects of selenium were deemed controversial [13]. On the contrary, in patients with HT, Se supplementation reduced serum TPOAb and TgAb levels after 6 months of treatment [14]. Concerning iron (Fe) deficiency, recognized as the most prevalent nutritional deficiency worldwide, data indicates that it may adversely affect thyroid function and autoimmunity. This impact is particularly noteworthy in pregnant women and women of childbearing age [15].

An important factor which has raised concerns the last decades is that of the consumption of goitrogenic foods, which are those foods that contain substances such as goitrin, thiocyanates, and isothiocyanates, and that seem to interfere with iodine metabolism leading to hypothyroidism, mostly in animal and in vitro studies [16]. The randomized trial reported by Chartoumpekis et al. found no impact of the consumption of cruciferous vegetables on THs levels or on thyroid autoimmunity [17]. Also, one small study conducted by Kim et al. on five healthy participants showed transient decreases by 25% of radioiodine uptake from baseline after 6-hour of 15.2 oz intake of kale juice, twice per day for 7 days and no effect on TSH and FT4, despite increases in serum and urinary thiocyanate concentrations ( [18]. This evidence suggests that the ordinary consumption of cruciferous vegetables is unlikely to cause hypothyroidism or autoimmune thyroid disorders in humans [19]. Other toxic nutritional elements such as lead, cadmium, chromium, nickel, aluminum, and even certain food additives, i.e., nitrates used for fish and meat preservation, may provoke or support hypothyroidism when their levels in the circulation are in excess [20].

Data regarding the effects of macronutrient composition in normocaloric diets on thyroid function in healthy individuals are scarce. In a big cross-sectional study by Brdar et al, it was found that the dietary pattern is related with subtle differences in thyroid function tests in euthyroid subjects. They examined 18 different food consumption patterns and plasma FT3, FT4, and TSH levels in 4585 lean or overweight self-reported euthyroid individuals, by using a food frequency questionnaire. They found that the frequent consumption of foods with high GI, such as fruit juices and refined bread, pasta and rice, had a negative association with TSH levels, and a positive association with FT3 and FT4 levels, while a high consumption of protein or foods rich in saturated fatty acids had a negative association with FT3 and FT4 levels. However, there are several limitations in this study. First, although they selected individuals without thyroid disease or taking any medication affecting thyroid function, up to 719 subjects had either THs or TSH levels outside the reference range and were excluded from analysis. Also, they did not measure thyroid autoantibodies and they did not perform thyroid ultrasonography [21].

4.2. Starvation/overfeeding

Besides, it seems that thyroid function is modulated by starvation, or overfeeding. Calorie deprivation has profound effects on thyroid function in individuals with intact HPA axis. Vagenakis et al, in 1975 first described that caloric deprivation for 4 weeks in 9 obese subjects resulted in a striking decrease in serum T3 and increase of rT3, which returned to baseline values during a 5-day period of refeeding [22]. These findings were confirmed later in numerous studies [23,24] and are considered as a type of non-thyroidal illness syndrome. It seems that carbohydrate deprivation instead of starvation per se is responsible for the T3 drop [25]. Furthermore, long-term calorie restriction induces similar alterations in THs in euthyroid individuals. In a study, 28 healthy subjects adhering to a calorie restriction diet (comprising a balanced array of foods providing over 100% of the recommended daily intake for essential nutrients and an energy intake of 1779±355 kcal/day) for an average duration of 6 years were examined. TH levels were measured and compared with those of 28 sedentary individuals matched for age and sex, as well as 28 exercising individuals matched for body fat, all of whom followed Western diets. The findings revealed that subjects undergoing long-term calorie restriction exhibited lower serum triiodothyronine (T3) values [26]. Moreover, short-term overfeeding leads to THs alterations. A recent study by Basolo et al, has examined the effects of overfeeding (200% of energy requirements) in 58 euthyroid non-obese subjects with normal glucose tolerance test(Basolo et al). In this study, participants underwent 24-hour interventions featuring five overfeeding diets with varying macronutrient compositions in a crossover design. The diets included a low-protein overfeeding diet (LPF: 3% protein, 51% carbohydrate, and 46% fat), a high-protein overfeeding diet (HPF: 30% protein, 44% fat, and 26% carbohydrate, n = 51), and three overfeeding diets with normal protein content (20%): standard (50% carbohydrate and 30% fat), high-carbohydrate (75% carbohydrate and 5% fat), and high-fat (60% fat and 20% carbohydrate). The changes in thyroid hormones (THs) observed after the overfeeding diets were dependent on the macronutrient composition. Following LPF or HPF, plasma TSH decreased by 13-16%, reverse triiodothyronine (rT3) remained unchanged, while mean plasma free thyroxine (fT4) was either unchanged or slightly decreased. Plasma free triiodothyronine (fT3) slightly increased or decreased following LPF or HPF diets. Notably, overfeeding with normal protein diets exhibited no discernible effects on THs [24].

4.3. Obesity

Obesity is a major health problem, and its prevalence has been on the rise globally over the last few decades [27]. Obesity carries an increased risk of various health conditions such as type 2 diabetes, cardiovascular diseases, dyslipidemia, certain cancers, and musculoskeletal disorders. Subtle changes in thyroid function have been also described in obesity. Indeed, obese children and adults had either serum TSH, free T3 (FT3) and/or free T4 (FT4) to the upper normal range [28, 29], or only a slight increase in TSH concentration without alterations in FT3 and FT4 levels [30, 31]. It remains unclear if obesity per se causes these alterations or if a type of subclinical hypothyroidism predisposes to obesity. The first hypothesis is more likely since weight loss restores thyroid function [32]. High serum leptin levels in obese individuals, produced by adipose cells [33], could be a candidate mechanism explaining the elevated TSH levels in obesity, as it is known that leptin stimulates TSH release [34]. Other mechanisms have also been proposed, such as the presence of resistance in TH action [35], insulin resistance and chronic low-grade inflammation [36].

5. Low Glycemic Load (LGL) Diets

In 1981, Jenkins et al. demonstrated the different effects of foods on the blood glucose level and introduced the term of Glycemic Index (GI), which represents a physiological assessment of a food’s carbohydrate content through its effect on postprandial blood glucose concentrations [37]. Today there is consensus that the glycemic response to a food or a diet is better predicted if not only the quality but also the quantity of carbohydrates is considered. The relevant index that combines the GI of a food with the amount of its carbohydrate content is the Glycemic Load (GL=GI multiplied by the amount of carbohydrate in a typical serving), which allows the glycemic effect of a food, a meal, or a whole diet to be compared as realistically consumed [38]. However, there is still confusion over the meaning of the GI and GL and the appropriate way they should be utilized. To facilitate their widespread use, revised versions of international tables of GI and GL values have been published [39].

Regarding diabetes mellitus pathophysiology, it has been suggested that postprandial hyperglycemia may be a contributing factor in driving hyperinsulinemia and insulin resistance [40]. Randomized controlled trials have shown that low-GI/GL diets attenuate postprandial glucose concentrations, thereby improving long term glycemic control, other established cardiometabolic risk factors and adiposity [41]; thus, low GI/GL diets may reduce the risk of certain chronic diseases [42,43]. In line to these observations, is a recent umbrella review of meta-analyses of prospective cohort studies, in which it has been shown that there was a positive association between dietary GI and the risk of type 2 diabetes, coronary heart disease and colorectal, breast and bladder cancers, as well as between GL and the risk of type 2 diabetes, coronary heart disease, and stroke. Regarding other types of cancer there was no significant association [44].

A few years ago, LGL diets appeared to be a promising alternative to conventional diets in the treatment and prevention of obesity [45]. However, the latest meta-analyses both in children and adults suggest no clear benefit of the LGI/LGL diet over other high GI or GL diets. According to a 2023 systematic review and meta-analysis, LGI and LGL diets do not seem to be associated with changes in adiposity, cardiometabolic or glucometabolic markers in children with overweight or obesity [46]. In line with these results, a very recent systematic review of 10 RCTs indicated little to no difference in change of body weight with low GI/GL diets versus higher GI/GL diets or any other diet in adults, suggesting the need for more well-designed and adequately powered studies [47].

6. LGL Diets Lead to Subtle Changes of Thyroid Function Tests in Euthyroid Subjects

Thyroid function parameters in euthyroid individuals undergoing weight loss through dietary interventions have been explored in only a limited number of studies. Moreover, these investigations have primarily focused on weight loss in obese subjects with or without diabetes mellitus and other co-morbidities. Several types of diets have been used, but the effect of each type of diet on thyroid function is challenging due to the simultaneous impact of the underlying disease, calorie restriction and weight loss on the results (Table 1). Recently LGL diets have gained popularity due to their effectiveness in promoting weight loss and improving glycemic regulation, in individuals with obesity or diabetes [48]. In one study, 7 healthy non-obese euthyroid young adults were randomly assigned to follow either a high-fat, low-carbohydrate diet for 7 days, followed by an isocaloric high-protein, low-carbohydrate diet for another 7 days, or the reverse sequence. Thyroid hormone levels were measured after each dietary phase. Despite the diets being designed to maintain body weight, both groups experienced a small but significant weight loss after each diet. Serum levels of thyroid-stimulating hormone (TSH) and triiodothyronine (T3) decreased after each diet, while thyroxine (T4) and reverse triiodothyronine (rT3) remained stable. The decline in T3 was more pronounced after the high-fat diet [49]. There is a recently published pilot study with excellent design [50]. It is a randomized-controlled-crossover trial in 11 young healthy lean adults. The participants were randomized to follow two isocaloric diets for a minimum of three weeks each with one week washout period in between; either a ketogenic diet (KD:15% carbohydrate, 60% fat, 25% protein) and then the high-carbohydrate low-fat, HCLF: (55% carbohydrate, 20% fat, 25% protein) diet or vice versa. KD is a very low-carbohydrate diet and consequently it could consider as a LGL diet, though it causes also metabolic acidosis. Both interventions led to substantial weight loss, with KD exhibiting a more pronounced effect, despite the diets being isocaloric and comparable in resting metabolic rate and physical activity levels. Their findings suggest that the influence of dietary macronutrient composition on body weight defies a straightforward explanation through the principles of thermodynamic energy balance. Notably, thyroid hormones (THs) remained unchanged on a high-carbohydrate, low-fat (HCLF) diet, while on a ketogenic diet (KD), serum T3 levels decreased and T4 levels increased, with TSH levels remaining constant. The observed lower serum T3 levels on the KD might be anticipated to yield effects contrary to those observed. The authors propose that unmeasured parameters regulating T3 action could be implicated. For instance, during KD, alterations in intracellular T3 concentration, affinity with thyroid hormone receptors (TR), or the involvement of nuclear co-repressors may play a significant role [50].

7. LGL Diets and Weight Reduction in Hypothyroidism

It is well established that T3, the active form of THs, regulates basal metabolic rate, thermogenesis and food intake [51]. Inadequate production of THs in overt hypothyroidism (either autoimmune or due to other causes) is frequently associated with weight gain whereas thyrotoxicosis of any cause leads to weight loss [52]. The meta-analysis of Song et al., including 22 studies, showed that obesity was significantly associated with an increased risk of hypothyroidism (RR = 1.86, 95% CI 1.63-2.11, p < 0.001). Further analysis also revealed that obesity was clearly associated with Hashimoto Thyroiditis (HT) (RR=1.91, 95% CI 1.10-3.32, p=0.022) and with high levels of anti-TPO antibodies (RR = 1.93, 95% CI 1.31-2.85, p = 0.001), but not with Graves’ disease. This data also indicates the bidirectional relationship between obesity and the risk of hypothyroidism. In patients with HT, treatment with L-thyroxine has been shown to induce significant decreases in body weight and BMI after euthyroidism is restored [53]. However, other studies have found that even after restoration of euthyroidism, hypothyroid patients could experience difficulties in achieving weight loss, implying that prevention of obesity is crucial for thyroid disorders [54]. Besides, it is known that in patients receiving adequate substitution therapy with L-thyroxine, T3 levels may frequently be in the low-normal range, despite normal FT4 and TSH levels; this might explain why these patients might struggle with weight loss, as decreased T3 levels have been linked to difficulty in maintaining or further decreasing weight [55].

Several studies have suggested that eating habits could affect the risk of manifesting inflammatory and immune diseases, including autoimmune thyroiditis (HT) [56]. Furthermore, different nutritional interventions have been tested in patients with HT. Lactose-restricted, gluten-free, Paleolithic-style, and “autoimmune protocol” diets, implemented in HT patients for periods ranging from a few weeks to several months, have been shown to produce significant decreases in thyroid autoantibodies and, in some instances, in TSH as well [56]

We found only one report from Esposito et al, in which the effects οf a LGL diet on 180 patients with HT was studied (Table 1) [57]. Among these patients, 108 adhered a LGL diet (12-15% of the total daily energy intake) devoid of bread, pasta, fruit, and rice and free of goitrogenic food (experimental Group, n=108) and 72 were on a low caloric diet without food restrictions (control Group, n=72). The intervention period spanned three weeks, during which baseline and concluding assessments were made for thyroid hormones and thyroid autoantibodies. Notably, the authors did not specify whether patients with HT were undergoing levothyroxine treatment, but their thyroid function tests consistently indicated values within the normal range. Patients in both groups achieved similar body weight loss, but body fat reduction was greater in patients on the LGL diet. Thyroid hormones did not change in any diet. Intriguingly, thyroid autoantibody levels demonstrated a marked decrease ranging from -40% to -57% in the experimental group, while conversely increasing in the control group. These findings underscore the potential efficacy of the LGL diet in mitigating thyroid autoantibody levels among patients with HT [57].

8. Impact of Low-Glycemic Load diets on the Risk of Thyroid Cancer

Papillary thyroid cancer constitutes the predominant form of thyroid cancer, accounting for 84% of all cases [58]. Over the past three decades, there has been a consistent rise in the incidence of papillary thyroid cancer, potentially linked to the overdiagnosis of indolent microcarcinomas, a result of the widespread utilization of thyroid ultrasonography [59]. However, the possibility of a genuine rise in thyroid cancer cannot be dismissed, as larger tumors continue to be diagnosed [60]. Radiation exposure during childhood is a well-established risk factor, as evidenced by a comprehensive pooled analysis of 12 studies [61]. Additionally, emerging evidence suggests a potential association between obesity and thyroid cancer development [62]. Healthy diet, avoiding red meats and sugar protects from cancer development [63]. Regarding thyroid cancer in a case-control study by Randi et al. the researchers explored the influence of Glycemic Load (GL) and Glycemic Index (GI) in the diet [64] (Table 1). The study included 399 subjects with histologically confirmed thyroid carcinoma (of which 274 had papillary carcinomas) within the two years preceding their enrollment. Dietary habits were assessed using a food-frequency questionnaire and compared with those of 616 control subjects who did not have neoplastic disease. In this study, GI and GL were determined mainly by high dietary intakes of bread, rice and pasta, and both were associated with increased thyroid cancer risk. In detail, the OR for the highest tertile of GI compared with the lowest one was 1.70 for papillary and 1.57 for follicular thyroid cancer. The respective ORs for GL were 2.17 for papillary and 3.33 for follicular thyroid cancer [64]. Also, the European Prospective Investigation into Cancer and Nutrition (EPIC) study, a large cohort with a high heterogeneity in dietary intake across the 10 participating countries, investigated the associations between differentiated thyroid carcinoma (TC) and the intake of energy, macronutrients, GI and GL in almost half a million participants. This study revealed significant associations between TC and high total energy and low polyunsaturated fats (PUFA) intakes. The associations with starch and sugar intakes and GI were significantly heterogeneous across Body Mass Index (BMI) groups: there was a positive association of TC with starch and GI in participants with a BMI ≥ 25kg/m², and with sugar intake in those with BMI < 25kg/m² [65].