Is there a link between obesity phenotype and thyroid diseases? A mini-review of current concepts

Obesity is a chronic, progressive and relapsing disease, characterized by excessive and dysfunctional adiposity. In the adult population it is defined as body mass index (BMI) >30 kg/m² (in Caucasian population) or increased percentage of adipose tissue [1].

Although there is an agreement for the BMI values defining obesity, there is no clear consensus for the cut-off values for normal and excessive body fat percentage. Most commonly these are held to be >25% for men and >30–35% for women [2,3]. The global prevalence of obese people has tripled over the last 3 decades [4], being an emerging health problem in most developed countries. Recently BMI has been considered a screening tool, but to establish full diagnosis, a more detailed examination should be performed, including assessment of obesity-related complications, to adequately stratify the health risk [5,6]. Thus, obese individuals are not a homogenous group and may represent different obesity phenotypes, according to metabolic status. The most common obesity phenotypes presented in the literature are: metabolically healthy obese (MHO) and metabolically unhealthy obese (MUO), which can be further categorized into those with developed metabolic syndrome (MetS) and non-fully developed metabolic syndrome (pre-MetS) [7].

The most recent criteria of MetS are based on the measurement of blood pressure, triglycerides (TG), HDL-cholesterol, fasting glucose (FG), and waist circumference (WC). To diagnose MetS, 3 out of 5 abnormal findings should be confirmed [8]. On the opposite pole, there are metabolically healthy obese individuals (MHO phenotype). These patients, despite being classified as obese, do not present components of MetS. Moreover, MHO should have no antihypertensive, glucose-lowering, or hypolipemic treatment [9]. Although this phenotype does not present any metabolic complications of obesity, it cannot be considered as healthy as the non-obese population. In long-term observational studies the individuals classified as MHO eventually develop metabolic alterations, progressing to the category of MUO [10,11,12], and may have other obesity-related complications.

Pathogenesis of obesity is complex and involves psychological, behavioral, neuroendocrine, and environmental factors. However, the fundamental element leading to obesity is imbalance between energy uptake and energy expenditure. From this perspective the pivotal role in metabolism regulation and obesity development is played by the thyroid gland. Indeed, the interaction between thyroid hormones and obesity is bidirectional. Excessive adiposity and dysfunction of adipose tissue may be related to the development of thyroid functional and structural impairment such as hypothyroidism [13,14,15], autoimmunity [14,16], thyroid nodules [17,18,19], and thyroid cancer [20,21]. In most cases the hormonal imbalance can be reversed by weight loss, no matter the method (bariatric surgery, dietary intervention, or pharmacotherapy) [22]. The prevalence of thyroid disorders in the obese population is higher than in normal weight individuals [23], and the European guidelines recommend the assessment of thyroid function in all obese patients [24].

There is much evidence in the literature on the relation between obesity and thyroid diseases, both at functional and morphological levels; however, still there is scarce data on the prevalence of particular disorders according to obesity phenotypes. The aim of this mini-review is to present the current knowledge on the interaction between thyroid gland disorders and obesity, with special focus on obesity phenotypes.

The thyroid gland plays a fundamental role in metabolism regulation and energy expenditure. In the physiological mechanism, triiodothyronine (T3) serum concentration exerts negative feedback on thyrotropin releasing hormone (TRH) and thyroid stimulating hormone (TSH) production, regulating the activity of tissue-specific deiodinases (D1, D2, D3) and therefore conversion of thyroxine (T4) to triiodothyronine (T3). Thyroid hormones (TH) act in genomic and non-genomic mechanisms, either binding to nuclear thyroid receptors (TR, tissue-specific isoforms alpha 1, 2 and beta 1,2,3) or non-nuclear receptors located in mitochondria, cytoplasm, and cell membrane. They control thermogenesis, set the basic metabolic rate, regulate hepatic lipid metabolism and lipids turnover [25]. Thyroid hormones, specifically T3, also regulate food intake at the central level, by modulating the release of orexigenic peptides in hypothalamus—agouti-related peptide (AgRP) and neuropeptide Y (NPY) [26,27].

Recently, it has also been suggested that TH may stimulate white adipose tissue (WAT) browning and indeed augment brown adipose tissue (BAT)—the main site of facultative thermogenesis [28].

Several mechanisms of hypothalamus-pituitary-thyroid (HPT) axis alterations in obesity have been proposed, including low-grade inflammation, insulin resistance (IR), disturbed adipokines production, altered deiodinases activity, and TR expression as well as relative resistance to TH.

Obesity is often characterized by a low-grade inflammatory status. The impairment of oxygen and blood supply in hypertrophic adipocytes leads to hypoxia and ischemia that in turn stimulate infiltration of macrophages [29]. Dysfunctional tissue overproduces immunomodulating adipose tissue–derived compounds (adipokines) such as leptin, adiponectin, visfatin, and resistin, which have immunomodulating properties. In combination with increased oxidative stress, overproduction of reactive oxygen species (ROS), and overexpression of cyclooxygenase 2, it provokes endothelial damage and further changes in immunity [30,31,32,33]. Overproduction of cytokines may also have inhibitory effect on sodium/iodide transporter [34,35] and deiodinases activity [36]. Moreover, cytokine-induced increased permeability of blood vessels contributes to the altered morphology of thyroid gland [34].

Insulin resistance can be defined as resistance to the insulin stimulation of the target organs: mainly adipose tissue, skeletal muscles, and liver. This condition is usually acquired, and related to excessive body fat [37]. Dysfunctional adipose tissue with hypertrophic adipocytes account for the release of proinflammatory cytokines that, together with accompanying lipolysis and overproduction of free fatty acids (FFA), aggravates inflammation, facilitates ectopic deposition of fat, and impairs insulin signaling in WAT [38,39,40]. On the systemic level, the lipotoxic effect of accumulation of FFA on skeletal muscles and liver also impairs insulin signaling and induces such proinflammatory responses as macrophage infiltration, ROS production, steatosis, and fibrosis with all its negative metabolic consequences [41,42,43,44]. A correlation between insulin and TH concentration as well as the stimulatory effect of insulin on deiodinases activity has been observed [45,46,47]. In addition, an association between high fT3/fT4 ratio and IR has also been documented [48]. Moreover, IR and hyperinsulinemia may contribute to thyroid morphological changes and carcinogenesis [49], acting synergistically with TSH and growth factors [50,51] and having an impact on angiogenesis [52].

In most obese patients we can commonly observe normal or increased TSH concentration, normal or slightly decreased fT4 concentration, and fT3 within reference range. These findings can be linked with leptin, which is an important effector in the postulated “hypothalamus-pituitary-adipose tissue-axis” [26, 53,54,55]. In the study of Betry et al. [54] a strong correlation between leptin level and TSH concentration has been demonstrated, independently from BMI, in the group of 800 obese patients. In a recent study by Karpuzoglu et al. it was revealed that maternal serum leptin was positively correlated with TSH in newborns, being a risk indicator of hypothyroidism [56]. Leptin is exclusively secreted by adipose tissue and it regulates energy homeostasis and food intake. It crosses the blood-brain barrier and plays an important role in synthesis and regulation of TH production. Leptin stimulates D1 and conversion of T4 to T3 in liver and kidneys [57,58] as well as in white adipose tissue [59].

Through a central mechanism it regulates TRH secretion and in turn TSH production. It is observed that after weight loss and reduction of amount of circulating leptin, there is a reduced stimulation of TRH and TSH [53]. Physiologically, leptin induces TRH expression in the paraventricular nucleus (PVN) and production of anorexigenic peptides in the arcuate nucleus (AN). In AN leptin activates proopiomelanocortin (POMC) neurons and downregulates the synthesis of AgRP and NPY, which subsequently results in a decreased food intake [60]. However, due to leptin resistance, in obesity the impact of circulating leptin on synthesis of anorexigenic peptides is minimal.

As mentioned before, obese patients, even if in euthyroid state, have relatively higher TSH, fT3 and fT3/fT4 ratio in comparison to non-obese individuals [45]. It can be interpreted as a compensatory mechanism preventing further weight gain. However, despite the rise of fT3 and TSH, in individuals with obesity no spontaneous weight loss is observed, probably due to reduced expression of thyroid and beta-adrenergic receptors and altered function of deiodinase in adipose tissue [61,62]. It may cause relative resistance to TH at both peripherical and central levels. It was demonstrated that one of the reasons for increased TSH production in obesity may be pituitary resistance to fT3 [63,64,65]. D1 is overexpressed in white adipose tissue of obese individuals and the expression of TSH receptors and TR alfa 1 is decreased. This state is reversible: after weight loss (surgical or non-surgical) the expression of D1 deiodinase, expression of TSH receptor, and TR alfa1 is restored to normal. In these patients TSH and fT3/fT4 ratio also declines [22,61,66]. In euthyroid obese patients TH treatment is not advised for weight loss, as there is no evidence for its efficacy or safety [67,68]. American and European guidelines strongly recommended against such a therapy [24,69].

BAT is involved in the process of non-shivering thermogenesis [70]. Physiologically, BAT activity depends on the sympathetic nervous system; this process can be regulated centrally (via the hypothalamus), but also may be regulated by TH [71]. Moreover, BAT is an important site of T4 to T3 conversion. Sympathetic stimulation enhances local T3 production via increased activity of BAT-specific D2 [72,73]. Obesity is characterized by catecholamine resistance [74]. Downregulation of adrenergic receptors in BAT may affect the activity of D2, that contributes to lowering thermogenesis and energy consumption [75,76,77,78,79]. The possible mechanism can be linked with inflammatory state and macrophage infiltration of BAT and hyperleptinemia [80]. It has been shown that leptin may influence the activity of deiodinases: it may increase activity of D1 and decrease activity of D2 in BAT, contributing to lowering the metabolic rate and thermogenesis [13, 81]. Importantly, in obese individuals also the quantity of BAT is reduced in comparison to the lean population [74,82,83]. Impaired differentiation of preadipocytes into BAT is plausibly the consequence of the catecholamine resistance mentioned before [74, 79] and concomitant oxidative stress and low grade inflammation [84, 85].

Most euthyroid obese individuals present a typical constellation of TSH and free thyroid hormones; however, the prevalence of hypothyroidism is 10 times higher than in the normal-weight population [23,86,87]. The relationship between autoimmune thyroid disease (AITD) and obesity is not well established and the data from different studies are conflicting. The prevalence of anti-thyroid peroxidase antibodies (TPOAb) positivity in the adult population with severe obesity ranges between 11–17% [84,88], which is similar to the general, non-obese population. Interestingly, the presence of anti-thyroid antibodies in individuals with elevated TSH is lower than in the non-obese subjects. This phenomenon may be explained by dominant non-immune causes of TSH increase. In the meta-analysis from 2019 [14], obesity was correlated with positive TPOAb (RR 1.93, 95% CI 1.31–2.85, p = 0.001), but not with positive anti-thyroglobulin antibodies (TGAb). Obese patients had increased risk of hypothyroidism (RR 1.86, 95% CI 1.63–2.11, p < 0.001) and slightly (at the border of statistical significance) elevated risk of Hashimoto's thyroiditis (p = 0.077). Contrarily, in the novel study by Guo et al. [89] hyperthyretropinemia has not been associated with the BMI category; however, in the presence of positive anti-thyroid antibodies, the risk of developing hypothyroidism in obese individuals was significantly increased and correlated with BMI. The obesity may facilitate the development of autoimmunity including AITD, but in obesity autoimmune etiology of hyperthyreotropinemia or hypothyroidism does not play the dominant role.

Obesity is also associated with thyroid hypoechogenicity on ultrasound examination, but contrarily to the general population, it is not related to AITD, but rather to the accumulation of fat in the neck area and increased permeability of small thyroid vessels associated with obesity [90,91].

The morphological changes of the thyroid gland in obese individuals are often found. The proposed mechanisms are linked to the higher concentration of TSH, insulin, and growth factors that stimulate development of goiter and thyroid nodules [17].

Insulin exerts both metabolic and mitogenic effects on the thyroid cells, while insulin lowering agents exhibit anti-goitrogenic effects [92]. The insulin receptor (IRec) belongs to the tyrosine kinase family and is represented in two isoforms—A (IRec-A) and B (IRec-B)—which differ in their tissue distribution, affinity to other than insulin ligands (proinsulin, insulin-growth factors), and therefore activation of the specific molecular pathways [93]. IRec-A is predominantly present in less-differentiated cells, has a higher affinity to IGF, and its activation triggers mitogenic pathways. IRec-B is expressed mostly in well-differentiated cells and is responsible for metabolic activity of insulin [94, 95]. Moreover, the needed insulin concentration for metabolic pathway activation is much lower than for mitogenic pathway activation. However, under persistent insulin stimulation, the proportions of IRec-A to IRec-B in different tissues may change, causing upregulation of IRec-A [96]. Such a situation is observed in obese individuals [20, 97] and can partially explain why obesity is also a well-established risk factor of carcinogenesis [98,99]. Obese individuals exhibit 55% increased risk of TC development in comparison to lean individuals. Interestingly, this increased incidence is specific for papillary (PTC), follicular (FC), and anaplastic (ATC) variants, while the risk for medullary thyroid cancer (MTC) is reduced. The differences in the risk of tumor type may be associated with their different histological origin (follicular cells for PTC, FTC, and ATC vs. parafollicular cells for MTC) [21].

Concomitant thyroid diseases are common findings in individuals with obesity. The relationship between obesity and thyroid is complex and bidirectional. Hypothyroidism may promote development of obesity, but also obesity may be the underlying cause of thyroid dysfunction and morphological changes. Although thyroid dysfunction is not mentioned among obesity-related disease, taking into consideration its high prevalence in comparison to non-obese population, in our opinion thyroid evaluation should be recommended for all individuals with obesity. Although a recommendation to measure TSH concentration in every obese patient is mentioned in the latest European consensus, other guidelines suggest performing thyroid diagnostics only in case of clinical suspicion of thyroid disorder [5,6,24].

Regulation of TH production is complex, influenced by environmental, physiological, and genetic factors. In the context of obesity, thyroid dysfunction, most commonly hypothyroidism, may contribute to decreased thermogenesis and metabolic rate. This in turn facilitates weight gain [15, 124] and may result in unfavorable body composition [125,126,127]. Moreover, lower fT4 levels are associated with adverse changes in lipid profile, muscular and liver accumulation of fatty acids, IR and, in consequence, the development of MetS [48,124,128,129]. It was observed that even small TSH rise (even within reference ranges) is associated with higher BMI [130,131]. However, hypothyroidism rarely causes the development of severe obesity. Higher fT3 and fT3/fT4 ratio that are typically found in obese individuals can be interpreted as a compensatory mechanism preventing even more weight gain [45]. In many studies it was confirmed that normal thyroid function can be restored after weight loss [22,61,66,132,133,134]. The alterations in thyroid function can rather be the consequence of obesity than its only cause and may reflect an adaptational change in HPT axis, playing its role as a metabolic regulator.

Recently the concept of Adiposity Based Chronic Disease (ABCD) has emerged [135]. Indeed, the spectrum of metabolic consequences expressed by the different obesity phenotypes is a manifestation of different grades of adipose tissue sickness. From this point of view it is more likely that unhealthy metabolic status puts the obese individuals at higher risk of thyroid related diseases, rather than obesity per se. As fat tissue can be considered a separate endocrine organ, importantly not only the amount of adipose tissue but its altered distribution and function may lead to health deterioration. As highlighted before, dysfunctional adipose tissue with hypertrophic adipocytes is characterized by inflammation and an altered profile of produced cytokines and adipokines that contribute to metabolic dysregulation and putatively to autoimmunity [136].The study of Marzullo et al. showed that higher leptin levels in obese subjects were associated with the presence of thyroid antibodies and thyroid autoimmunity [16]. Furthermore, recent research revealed that another adipokine—visfatin—correlates with TPOAbs in hypothyroid patients both in adults and children [137, 138]. In the report by Belligoli et al. the omental and subcutaneous adipose tissue of the patients with obesity and impaired glucose metabolism was characterized by impaired vascularization, increased adipocyte size, and significant reduction in adipocyte stem cells, contributing to hypoxia, inflammation, and reduced capacity for TG storage [139]. Thus, persistent low-grade inflammatory status provokes endothelial damage, changes in immunity, development of IR that in the long term may lead to the alteration of thyroid morphology and function [31,32,33,129]. In conclusion, although patients with hypothyroidism are at higher risk of gain weight, also individuals with unhealthy obesity phenotypes are at higher risk of hypothyroidism, development of thyroid autoimmunity, thyroid nodules, and thyroid cancer. The higher risk may be particularly attributed to altered glucose metabolism and IR. Indeed, not excessive adiposity per se measured by BMI, but particularly dysfunction and central distribution of adipose tissue with concomitant metabolic complications may contribute most to the risk of autoimmunity, morphological changes, and malignant transformation, which incidence might be dependent on obesity phenotype. Importantly, there are also gender differences indicating other genetic, environmental, and hormonal factors. Typically, in men excessive adiposity per se (regardless of metabolic status) was an independent risk factor for thyroid cancer and thyroid autoimmunity, while in women the risk was higher only in those presenting an unhealthy obesity phenotype. Gender differences regarding the consequences of adiposity and metabolic complications may be explained with different hormonal status and gender-specific fat distribution. These conclusions drawn from the review are in agreement with the current approach to obesity diagnosis and treatment; they should also direct the attention of clinicians to the assessment of body composition, body fat distribution, and obesity complications, rather than focussing on BMI itself.