After having reminded the benefits of the omega-3 polyunsaturated fatty acids (PUFA) on several autoimmune diseases, we aim to review their benefits in the thyroid setting both at experimental and clinical levels. We start our work by providing a snapshot of fish consumption since fishes represent a major dietary source of omega-3 PUFA. The other side of the coin (or the other face of Janus, upon comparing fish to this god of the Ancient Roman mythology) is that eating fish means also “eating pollutants”. One relevant pollutant is mercury (Hg), a trigger of autoimmunity that is measurable inside the human thyroid. However, in fish the two faces have unequal size. Indeed, in small oily fish, such as anchovy and sardine, the good face represented by the omega-3 PUFA is much larger than the bad face represented by the pollutants. Just the opposite is true for the large, top predator fish, such as swordfish and tuna. Thus, we also provide a review of studies showing thyroid consequences from eating fish. Finally, after having reminded the interactions between the omega-3 PUFA and thyroid hormones at several levels, and having reminded that other natural compounds have shown benefits in the setting of thyroid autoimmunity, we talk about the commercial side of the benefits of all such substances, namely the expanding market of nutraceuticals. Overall, we think that the use of supplements containing omega-3 in the clinical thyroid setting has enough scientific rationale.

The European Market Observatory for Fisheries and Aquaculture Products (EUMOFA) considers fish and seafood as “the aggregation of finfish, crustaceans and mollusks and cephalopods”, regardless of being fresh or chilled and frozen (1). Examples of crustaceans are lobsters, shrimps and crabs; examples of mollusks are oysters and clams; examples of cephalopods are octopus, squid, and cuttlefish (1). Household expenditure on fishery and aquaculture products grew by 17% from 2019 to 2020, much greater than the 2.1% inflation of prices for such products (1). This increase was most likely due to the increased at-home consumption due to the closing of the hotels and restaurants because of the COVID-19 pandemic (1). In 2017, the expenditure on fish in the European Union (EU) was equal to euro 54262 million, which was the highest in the world. However, in terms of per capita expenditure, with euro 106 EU ranked 8th after Iceland, Japan, Korea, Norway, Australia, Israel, and Switzerland (1). Excluding out-of-home consumption, household nominal expenditure on fishery and aquaculture products in 2020 (with % variation over 2019 given in parentheses) was led by Spain (euro 13608 million, +39%) and Italy (euro 12277 million, +3%), with Slovenia (euro 90 million, +1%) and Malta (euro 58 million, +4%) at the bottom of the ranking.

Consolidated data for consumption and other items are available up to 2019 (1). Consumption of fishery and aquaculture products in the EU dropped to 12.30 million tonnes of live weight equivalent (LWE) in 2019, continuing a declining trend that started in 2017. Wild products accounted for 9.41 million tonnes LWE (76% of 12.30), and farm products for 2.89 million tonnes LWE (the remaining 24%). Per capita apparent consumption, estimated at 23.97 kg LWE of mostly wild-caught products, was almost stable in 2019 compared with 2018. Portugal remains the major EU consumer, with 59.91 Kg per capita (-2% vs 2018). Portugal is followed by Spain (46.02 Kg, unchanged), Denmark (42.56 Kg, +6%), France (33.26 Kg, -0.5%), Luxembourg (32.84 Kg, -3%), and Italy (31.21 Kg, +1%). At the other extreme, Hungary and the Czech Republic consume 6.28 and 6.0 Kg (+3% and +7% vs 2018, respectively) (1).

Concerning the most consumed fish, 16 products were considered, and they were listed in decreasing order of consumption: tuna, salmon, cod, Alaska pollock, shrimp, mussel, hake, herring, squid, surimi, sardine mackerel, trout, sprat (brisling), saithe (coalfish) and other products. Illustrative per capita consumption in 2019 (with percentage change vs 2018 given in parentheses) were 3.10 Kg (+ 2%) for tuna, 2.11 Kg (-1%) for cod, 1.02 Kg (+ 2%) for hake, 0.98 Kg (-17%) for herring, 0.58 Kg (+1%) for sardine, 0.53 Kg (-12%) for mackerel, and 6.58 Kg (-2%) for other products (1). In 2020, over 80% of the total volume of fresh fishery and aquaculture products consumed by households in 11 EU countries analyzed was accounted for by Spain, Italy, and France. In decreasing order, the top five fresh fish species consumed by households in 2020 were: hake, salmon, sardine, European seabass, and gilthead seabream in Spain; gilthead seabream, mussels, salmon, European seabass, and anchovy in Italy; salmon, cod, saithe (coalfish), trout and gilthead seabream in France. Data for swordfish, a seafood species that will be mentioned subsequently in our review, were not provided.

Based on a document by CBI (Centre for the Promotion of Imports from developing countries) the swordfish fishery is very important for Southern Europe, especially Spain and Italy, the two European countries with the highest consumption of swordfish (2). With 22,676 tonnes in 2017, Spain is the leading European producer of swordfish, followed by Italy and Portugal. The top importers from non-European countries of frozen swordfish are Portugal, Italy and Spain (6,316, 3,762 and 2,130 tonnes, respectively). Italy’s imports of swordfish have increased by 18% since 2014 with most deliveries coming from Spain. Based on a document by Oceana (3), which is the no-profit largest international ocean conservation organization, Greece, Italy, and Spain are the European countries where swordfish is consumed the most, but numbers were not provided. According to a Spanish paper (4), which in turn reports data published in 1995, the consumption of swordfish by Spanish adults averaged 0.35 g/day, with marked differences between communities, ranging from areas of no swordfish consumption to two areas of highest consumption (1.06 g/day and 1.17 g/day). On an annual basis, the said 0.35, 1.06 and 1.17 g/day correspond to 0.13, 0.39 and 0.40 Kg/year. In Italy, swordfish accounts for 4.9% of the national consumption of fresh fish (5), and in the year 2017 it was the fifth species most consumed (5.7%), following European seabass (17%) (6). Thus, of the 29.80 Kg of fish consumed by Italians in 2017, 1.46 Kg were accounted for by swordfish. A recent Italian survey of 560 consumers (7), led to the identification of 24 seafood species that were commonly purchased. The highest preference was for gilthead bream (Sparus auratus), European seabass (Dicentrarchus labrax), swordfish (Xiphias gladius), and European pilchard (Sardina pilchardus), with approximately 13% of consumers who purchased swordfish.

There is abundant literature, which is accruing over the years, about the benefits of the omega-3 fatty acids as such or omega-3-rich-foods on some disorders (Table 1), including autoimmune diseases (8–18). An example of the clinical studies that show such benefits for illustrative autoimmune diseases is summarized in Table 2 (19–36), with more details provided for type-1 diabetes (T1D) due to the endocrine nature of this disorder. Table 1 also illustrates the magnitude of the literature on other topics of this review, such as fish consumption.

As well known, fish (particularly seafood) is a good source of high-quality proteins, vitamins, and minerals, including minerals relevant to thyroid physiology (iodine and selenium). As also well known, fish is a major source of the long-chain omega-3 polyunsaturated fatty acids (abbreviated as either LC n-3 PUFA or omega-3 PUFA), particularly docosahexaenoic acid (DHA) and eicosapentaenoic acid (EPA). Figure 1 illustrates the content of EPA and DHA in representative seafood (17) while Figure 2 illustrates the oily fish (with their scientific, English and Italian names), which will be referred to in some sections of this review. Italian names are reported because in Italy we refer to the oily fish as “pesce azzurro” (“azure fish”) due to the blue-green color of the green in the dorsal and later scales (37). In the absence of a scientific definition, their definition is culinary (37). In contrast, the n-6 (omega-6 PUFA), such as arachidonic acid and linoleic acid, are derived largely from plant sources. Plants (and therefore plant oils and seeds) are also a source of another LC n-3 PUFA: α-linolenic acid (ALA). Tissues can convert ALA into EPA and DHA, but because the corresponding conversions are only 1 to 10% and 0.5-5% in human tissues, this conversion process is inefficient in humans (38). Thus, the main dietary source of EPA and DHA is cold-water oily fish such as mackerel, salmon, herring, sardines, anchovies, etc…. Canned fish contains omega-3 PUFA, but some amounts may be removed during processing (39). A similar decline in content of the omega-3 PUFA occurs with frying (39, 40).

Fish and seafood contain many contaminants, since anthropic activities (e.g., agriculture, industry, mining) increase their concentration in the aquatic ecosystem (17, 41–43). Clearly, because of biomagnification, pollutants are mostly concentrated in the longer-living and larger predatory fish species such as swordfish, tuna, and shark. Classical fish contaminants are the heavy metals, with Hg being worthy of mention in this review because it acts as an environmental trigger of autoimmunity (17, 44–51), a mechanism that has also been observed for cadmium (Cd) (51–55). In predatory marine fish, approximately 90% of Hg is methylated (methylmercury), while the remainder consists of minimal amounts of inorganic mercury, ethylmercury, and phenylmercury (4).

A recent Italian study on the muscle tissue of 30 Mediterranean swordfish determined that the rank order of toxic metals was Hg, Cd and lead (Pb), while the rank order of essential metals was zinc (Zn), copper (Cu), nickel (Ni) and chrome (Cr) (56). Particularly, Hg, Cd, and Pb levels exceeded the respective critical values 1.02, 0.30 and 0.25 μg/wet weight (that is, the legal safety limits established by the European Community) in eight, three, and two of the swordfish specimens examined, respectively. In a Canadian study, Hg was detected in all samples of swordfish, tuna, marlin, and shark purchased from major supermarket outlets and fish retailers, with swordfish containing the greatest concentration (57). In a Food and Drug Administration document on “Mercury Levels in Commercial Fish and Shellfish”, where 68 types of fish are ranked based on their mean content of Hg, the 68th, 67th and 66th positions are occupied by tilefish (1.123 parts per million [ppm]), swordfish (0.995 ppm) and shark (0.979 ppm]) (58). In contrast, sardine and anchovy occupy the 5th and 8th positions with a mere 0.013 and 0.016 ppm concentration of Hg, respectively (58). Thus, it is no surprise that the blood Hg concentrations of 285 adult seafood consumers in Long Island (NY, USA) were positively associated with weekly tuna steak or sushi intake and monthly or weekly swordfish, shark, or marlin intake (59).

With particular reference to thyroid autoimmunity, associations between total blood Hg and positive thyroid autoantibodies (thyroglobulin autoantibodies [TgAb] and thyroperoxidase autoantibodies [TPOAb]) were evaluated using the National Health and Nutrition Examination Survey (NHANES), 2007-2008 (60). Such associations were searched in 2,047 non-pregnant, non-lactating women. Compared to women with the lowest Hg levels (≤0.40 μg/L), those with Hg levels >1.81 μg/L (upper quintile) had 2.24 greater odds (95% CI=1.22, 4.12) for TgAb positivity. In contrast, no significant association was found for TPOAb positivity (60). One study from the Czech Republic in patients with Hg hypersensitivity (61) concluded that removal of Hg-containing dental amalgam may help to successfully treat autoimmune thyroiditis (AIT). In this study, 27/39 AIT patients had Hg hypersensitivity (with amalgam fillings removed only in 15/27), while 12/39 AIT patients had not (controls). Serum TgAb and TPOAb were measured both at baseline and six months later. Compared to baseline, the 15 patients in whom amalgam fillings were removed had a significant decrease in the serum levels of both TgAb and TPOAb (P=0.0007). In contrast, both TgAb and TPOAb did not change in the other two groups (61). Also, for Cd, the other heavy metal that we mentioned above, a direct relationship with thyroid autoantibodies was reported (53). In Chinese women, natural log-transformed blood levels of Cd correlated directly with natural log-transformed serum levels of TgAb (53).

There is evidence for the Hg presence in the thyroid. In a very recent autopsy-based Australian study (62), the presence of intracellular inorganic Hg was searched in paraffin-embedded thyroid tissue blocks from 115 persons (68 males, 47 females; mean age = 54 years, median age = 47 years, range= 1 to 104) with varied clinicopathological conditions. Using autometallography, Hg was found in the thyrocytes with an age-dependent frequency: 4%, 9%, and 38% of persons in the age band 1-29, 30-59, and 60-104 years, respectively. The frequency of thyroid samples containing Hg was similar in males (18%) and females (21%). Laser ablation-inductively coupled plasma-mass spectrometry not only confirmed the presence of Hg, but also detected other metals in six selected samples: cadmium (n= 6), iron (n= 5), lead (n= 4), nickel (n= 2), and silver (n= 2). The authors concluded that Hg can trigger genotoxicity, autoimmunity, oxidative damage, and be involved in the pathogenesis of AIT, hypothyroidism, and thyroid cancer (TC) (62). A previous Italian study on thyroid tissue samples removed at surgery from 77 euthyroid subjects showed that Hg and Cd are significantly more concentrated in the thyroid than in the adjacent muscle and fat of the same individual (63). Worthy of note, this Italian group (64) compared the urine concentration of several metals in a Sicilian area that features an incidence of thyroid-cancer two-fold greater than a control area. The authors found that the geometric mean value in the first area was at least two-fold higher than that in the second area for eight metals, two of which being Hg and Cd (64).

Having mentioned TC is not improper, given the abundant literature on the important predisposing role for such malignancy (particularly, papillary TC) and its advanced stages exerted not only by Hashimoto’s thyroiditis (HT) but also by serum TSH per se, even by TSH levels that are within the upper values of the reference limits (65–88). In regard to TC, it is also pertinent to remind the involvement of chemokines (89–98) and the peroxisome proliferator-activated receptors (PPARs) in the molecular oncogenesis (95–106), so that both types of molecules may serve as novel targets of TC precision therapy or prevention. Quite interestingly, the beneficial health effects of DHA and EPA on metabolic diseases are thought to arise from their binding to and activation of PPARs (107), as it was shown illustratively for T1D in Table 2.

There is cross-talk between the omega-3 PUFA and the thyroid hormone pathways, as exemplified by the augmented thyroid hormone signaling pathways in the liver, this being one mechanism used by n-3 PUFAs to affect lipid metabolism (108). For instance, specific steps of TH signaling in lipid metabolism that are influenced by n-3 PUFA include higher liver expression of the thyroid hormone nuclear receptor TRβ1 and mitochondrial α-glycerophosphate dehydrogenase (109). Starting from their previous observation showing that plasma free fatty acids concentration in some hypothyroid patients is above the normal range and that this higher concentration is associated with less severe symptoms of hypothyroidism, Makino et al. investigated the effect of highly purified EPA ethyl ester (EPA-E) derived from fish oil on thyroid function in rats with methimazole-induced hypothyroidism (110). They found that oral administration of EPA-E inhibited the reduction of thyroid hormone levels and the change of thyroid follicles in the hypothyroid rats, suggesting that n-3 PUFA may prevent methimazole-induced hypothyroidism. In a Dutch study on 13 patients with hypothyroidism caused by the ablation treatment for well-differentiated TC (111), induction of hypothyroidism decreased PUFA levels in plasma, erythrocytes and polymorphonuclear leukocytes. Another site of interaction between PUFA and thyroid hormones can be the thyroid hormone plasma transport proteins. Several studies have shown that PUFA inhibit thyroid hormone binding to such carrier proteins (112, 113), one practical consequence being increased tissue availability of the biologically active protein-unbound, free thyroid hormone. In addition, studies on the brain of aged rats that were fed fish oil (27% DHA content) for one month showed an approximately 10-fold increase in the expression of transthyretin (114). Transthyretin is the second major thyroid hormone plasma carrier, which is synthesized also in the central nervous system. Since transthyretin also operates as an amyloid-beta protein scavenger, transthyretin overexpression could prevent the formation of amyloid aggregates (114). This study is relevant because decreased PUFA levels, particularly DHA, were detected in elderly subjects and patients with Alzheimer’s disease (AD), and because there is epidemiological evidence for an association between fish consumption and low prevalence of AD (114). Of interest, AD is also characterized by brain inflammation and decreased local concentration of specialized pro-resolving mediators (SPM) (115). SPM are derived from PUFA and are key in the resolution of inflammation. Because of the technical difficulties in investigating the microglia function directly, Wang et al. took advantage of the useful model of peripheral blood mononuclear cells (PBMC) (115). In their randomized, double-blind, and placebo-controlled trial on 204 AD patients, Wang et al. administered a placebo or a supplement of DHA (1.7 g) and EPA (0.6 g) daily for 6 months. At the end of treatment, in those who received the n-3 PUFA, the plasma levels of DHA and EPA levels increased. When the culture medium of PBMC incubated with amyloid-β 1-40 was analyzed, levels of the SPMs lipoxin A 4 and resolvin D1 secreted by PBMCs were decreased in the patients supplemented with placebo, but unchanged in the patients supplemented with n-3 PUFA. Changes in the levels of SPM secreted by PBMC were positively correlated to changes in plasma transthyretin, and to cognitive changes as well (115). In the setting of tissue protection by PUFA, experimental studies in rats by Videla and colleagues showed that the combination of DHA plus thyroid hormone (T3) protects from liver injury through a synergistic action that also involves causing increased intrahepatic levels of resolvins (116–120).

Table 3 summarizes human studies that link thyroid disorders with the consumption of contaminated fish (17, 18, 121–127). Overall, the consequences are impairment of thyroid function, as measured by serum levels of thyroid hormones and/or TSH, and triggering of thyroid autoimmunity, as measured by serum levels of thyroid autoantibodies. Only studies by Benvenga and colleagues addressed Hg contamination (17, 18). The group of pregnant women who consumed swordfish selectively or predominantly among other fish species ingested the greatest amounts of Hg and had the greatest both serum levels and rates of positivity for thyroid autoantibodies throughout pregnancy compared to other groups of women (17, 18). As a result, the swordfish eaters had the greatest rate of postpartum thyroiditis, since positivity for thyroid autoantibodies is a major risk factor for such autoimmune type of thyroiditis that develops within the first 12 months after parturition (17, 18).

In the preceding section, we have reminded the role of Hg as an autoimmunity trigger (17, 44–51). However, omega-3 PUFA antagonize this effect of Hg (128, 129). Gill et al. (129) showed that dietary ingestion of n-3 PUFA (fish oil) promotes CD95 signaling by upregulating caspase 8 activation and that DHA counteracts the negative effect of Hg on CD95 signaling in T lymphocytes (128).

The clinical benefit of thyroid autoimmune disorders given by consuming Hg-poor, omega-3-rich oily fish or by taking omega-3-based supplements (17, 18, 130–132) is summarized in Table 4. The group of pregnant women who consumed oily fish, selectively or predominantly among other species of fish, ingested the lowest amount of Hg but the greatest amount of omega- PUFA; this group of women also had the lowest both serum levels and rates of positivity for thyroid autoantibodies throughout pregnancy compared to other groups of women (17, 18). As a result, the oily fish eaters had the lowest rate of postpartum thyroiditis.

Concerning Breese McCoy’s study (130), one comment is the limitation given by the fact that only thyroid function tests were monitored, while thyroid autoantibodies were not.

Some comments deserve the questionnaire-based study on 232 hypothyroid patients from Poland (132), a country where 15.8% of women and 2.5% of men suffer from thyroid disease, with an increase of 4.0 percent points in the year 2019 compared to 2014 (132). First, as properly described by the authors, 24% of participants were diagnosed with additional diseases (8% with polycystic ovary syndrome, 4% with depression, and 2% with insulin resistance). Second, there was enormous variability in the 197/232 hypothyroid patients who took supplements. Various was not only the spectrum of supplements taken (Table 4 of their paper) but also the periodicity of taking them (every day, 74%; irregularly, 16%; every other day, 10%). Concerning the main source of information on supplements, patients chose websites (74%), physicians (52%), family and friends (46%), and social media (43%). About two-thirds of participants took supplements according to the leaflet (71%), while one-third according to physicians’ or pharmacists’ guidelines. Results for the 197 participants (85% of 232) who took the supplements are illustrated in Table 1 of their paper (132). In that table, patients were stratified into 8 categories based on the nutraceutical taken (vitamin D, vitamins B, iron, zinc, multivitamins, selenium, omega-3 acids), with data summarized as a percent of patients reporting a benefit for 8 items, the denominator being the said 197 participants. We think that the results of this study (132) should be interpreted cautiously, mainly because it is not possible to distinguish the effect of a given nutraceutical taken alone from that of the same nutraceutical taken in association with others, and because the numbers presented in the said Table 1 are internally inconsistent. For instance, in the initial text of the Results section, the authors state that zinc supplementation was taken by 8% of the participants, therefore by 16 participants (0.08 x 197). However, upon adding the percentages reported in Table 1, the sum is 26, meaning that the number of participants claiming a benefit from zinc supplementation was 51 (0.26 x 197). For the omega-3 users, the sum of percentages is 23, meaning that the number of participants claiming a benefit from such supplementation was 45. Yet, from the initial text in the Results section, omega-3 users had to be 29 or 30 (0.15 x 197 = 29.5). Other obvious limitations of this Polish study (132) are the lack of details on the degree of hypothyroidism (overt, subclinical, either), the magnitude of change in serum levels of TSH, and having neglected to obtain information on changes in serum thyroid autoantibodies.

Concerning the mechanism behind the protection from/amelioration of AIT and the decrease in serum levels of thyroid autoantibodies caused by the omega-3 PUFA, two recent studies point to the involvement of resolvins (133, 134). Resolvins, lipoxins, maresins, and protectins are specialized pro-resolving mediators (SPM), which are downstream derivatives of PUFA (133). Resolvins of the E series (RVE1-RVE3) derive from EPA, while resolvins of the D series (RVD1-RVD6) derive from DHA. Concerning RVE-1 (5S,12R,18R-trihydroxy-eicosapentaenoic acid), which is the resolvin investigated by Song et al. (133), it triggers all aspects of the pro-resolution cascade, from inhibiting lymphocyte aggregation at the inflammation site to efferocytosis (that is, the process by which apoptotic cells are removed by phagocytic cells). The cell receptor for REV-1 is chemokine-like receptor 1 (chemR23 or CMKLR) (135), with the ligand-receptor interaction triggering anti-inflammatory responses that include marked inhibition of proinflammatory cytokines and chemokines (136). In one article, Song et al. (133) measured serum levels of REV-1 and serum thyroid parameters (TSH, T4, FT4, FT3, TPOAb, and TgAb) in 30 untreated HT patients (median and interquartile range [IQR] of TSH= 2.62, 1.95–3.05 µIU/ml) and 27 age- and sex-matched healthy controls (TSH= 2.38, IQR= 1.54–2.94, P= 0.178 vs. patients). Levels of RVE1 in patients were significantly lower compared to controls (median and IQR= 24.09, 15.76–34.38 pg/mL vs 28.51, 20.76–51.23 pg/mL). RVE1 levels correlated inversely with increasing TgAb levels in both the unadjusted model (OR= 0.945, P= 0.002) and adjusted models (OR= 0.938, P= 0.005). RVE1 levels were the lowest (19.21, 13.81–27.34 pg/mL) in the highest quartile group of TgAb levels (TgAb >361 IU/mL), and the highest (37.70, 24.66–99.16 pg/mL) in the lowest quartile group (TgAb <34 IU/mL) (19.21, 13.81–27.34 pg/mL vs 37.70, 24.66–99.16 pg/mL). Concerning TPOAb, the association with serum levels of REV1 was an inverted U-shaped curve with the lowest RVE1 levels (median and IQR= 19.21, 15.11–26.01) in the highest quartile of TPOAb levels (>431 U/mL) and the highest 31.39 (23.79–62.31) in the second quartile of TPOAb levels (13.6–106 U/mL). RVE1 levels correlated negatively also with T3 and FT3 (133). In another article, Song et al. (134) addressed RVD1 (7S, 8R, 17S-trihydroxy-4Z, 9E, 11E, 13Z, 15E, 19Z-docosahexaenoic acid) by measuring its serum levels in 30 patients with HT and 33 healthy controls. Serum concentrations of RVD1 in patients were significantly lower than in controls (median and IQR= 134.76, 85.35-201.36 pg/mL vs 187.64, 131.01-326.85 pg/mL). There was a significantly negative correlation between the TPOAb levels and the RVD1 levels. The authors also measured serum levels of the inflammatory chemokine IP-10/CXCL-10, and found that this chemokine had a significant negative association with TPOAb in the patients (134). Noteworthy, the administration of DHA (daily doses of 300 mg/kg) combined with T3 (0.05 mg/kg) to rats increased significantly the content of RVD1 and RVD2 in the liver, but left unchanged that of RVE1 and RVE2, indicating a synergistic effect compared to treatment with DHA alone and T3 alone (118). The authors concluded that co-administration of T3 and DHA enhances the capacity of liver for the resolution of inflammation by increasing RVD1 and RVD2 availability, so such co-administration constitutes an important hepatoprotective protocol for clinical purposes (118).

Experimental studies that have evaluated the thyroidal effects of the omega-3 PUFA (137–143) are summarized in Table 5. Omega-3 PUFA were shown to have a neuroprotective (138, 139) and cardioprotective (143) action on unfavorable effects caused by thyroid hormone deficiency. On the other hand, the omega-3 PUFA mitigate, ameliorate certain hyperthyroidism-induced unfavorable effects, such as arrhythmias (141) and hepatic dysfunction (137). Such favorable effects of the omega-3 PUFA on consequences in the peripheral tissues that are caused by opposite states of thyroid dysfunction confer to the omega-3 PUFA a modulatory role that is reminiscent of another dietary nutrient, L-carnitine. Indeed, beneficial effects of L-carnitine were reported both in the hyperthyroidism setting (144, 145) and in the hypothyroidism setting (146, 147), with a modulatory role also demonstrated in the tissue glucocorticoid hormone action (148).

The very few studies available in the English-language literature on the omega-3 PUFA-based nutraceuticals in the clinical setting of autoimmune thyroid disorders (130, 131) (Table 4) contrast with the large number of studies that have tested seleno-L-methionine alone (149–151) or seleno-L-methionine combined with myo-inositol (151–157), which is a hexahydroxycyclohexane (C₆H₁₂O₆) and one of the nine stereoisomers of inositol that plays a pivotal role in many metabolic pathways (158). In the studies with seleno-L-methionine combined with myo-inositol, the benefit was generally greater than that given by seleno-L-methionine alone or myo-inositol alone. Such benefit was based on the reduction of TSH, TgAb, TPOAb levels, and, when studied (154), on the reduction of serum chemokine levels (CXCL10/IP-10). Myo-inositol+seleno-L-methionine treatment protected blood mononuclear cells (PBMC) of either HT or healthy patients from H2O2–induced stress. Furthermore, the association of myo-inositol and seleno-L-methionine decreased the expression of the chemokines CXCL10/IP-10, CCL2/MCP-1 and CXCL9 (159). In another study, which evaluated the thyroid toxicity of cadmium and protection from this toxicity by nutraceuticals (160), Cd induced a marked overexpression MCP-1/CCL2 and CXCL10 in the thyroid. Again, the protection given by seleno-L-methionine combined with myo-inositol was significantly greater than that of either nutraceutical alone (160). In vitro experiments on blood cells or other cells have shown that omega-3 PUFA decrease the secretion of CXCL10/IP-10 (161–163), CXCL9 (164) and CCL2/MCP-1 (11, 165–176).

Another nutraceutical, Aloe vera (also known as Aloe barbadensis), decreased serum levels of thyroid antibodies and improved subclinical hypothyroidism (177). It is pertinent to remind Aloe vera because, among the several nutritional substances contained in its leaves, there are selenium and PUFA, including the omega-6 linoleic acid and the omega-3 linolenic acid, the latter being the second most abundant after one saturated fatty acid (palmitic acid) (178). Noteworthy, in the Aloe vera juice used in the study by Metro et al. (177) the fat content is 0.2%, with saturated fatty acids being absent. As written in that paper (177), “one of the authors decided to take Aloe Barbadensis Miller juice (ABMJ), at the dose of 50 ml every morning on an empty stomach, as a skin soother and laxative”. This author had a history of HT-associated subclinical hypothyroidism, which she monitored with periodical checks of serum thyroid function tests and thyroid autoantibodies. “At the biochemical check performed three months after having started taking ABMJ, she was struck by the remarkable improvement of all indices (Table 1). The improvement was even more impressive six months later” (177). Based on this experience, the effects of ABMJ administration were tested, in a 9-month duration trial, on HT women with levothyroxine-untreated subclinical hypothyroidism and high levels of TPOAb (n= 30). Controls were 15 HT women with untreated subclinical hypothyroidism who were matched for age and baseline levels of TPOAb, TSH, FT4 and FT3. In the Aloe-treated group, TSH, FT4 and TPOAb improved significantly already at month 3 and further (−61%, +23% and −56%, respectively) at month 9. At baseline, 100% of women had TSH > 4.0 mU/L and TPOAb > 400 U/ml, but at month 9 rates fell to 0% and 37%, respectively. In contrast, the control group had no significant changes in any index (177).

Of the three aforementioned chemokines (CXCL10/IP-10, CCL2/MCP-1 and CXCL9), only one (MCP-1) was evaluated in terms of response to Aloe vera, and only by two studies (179, 180). In one study (179), a wound dressing containing an Aloe vera extract plus gelatin decreased the production of MCP-1 by 75% in human mesenchymal stem cells treated with TGFβ. In the other study (180) the oral administration of two antidiabetic phytosterols isolated from Aloe vera lophenol (lophenol and cycloartanol) decreased serum and hepatic concentrations of MCP-1 in Zucker diabetic fatty (ZDF) rats. Nevertheless, there is abundant literature on Aloe vera being able to decrease the production of pro-inflammatory cytokines and chemokines (181–188).

Based on an article of May 2018 by the main Italian press agency ANSA (189), Italy is the first European country for nutraceutical products based on per capita expenditure, that is, euro 40 compared to euro 28 of the EU. Italians spend more than euro 3.2 billion on dietary/supplements/nutraceutical products. A major driver for this expenditure is wishing to prevent and/or treat metabolic diseases (189) Based on the survey “The Italian food supplement supply chain, 2019-2020” by the FederSalus Research Centre (that is the national agency of the Italian producers and distributors of food supplements), in late 2019, the food supplement market in Italy reached a value of approximately euro 3.6 billion (190), corresponding to 27% of the euro 13.2 billion value of the European market, thus preceding Germany and France (18 and 8%, respectively). In particular, 32 million persons used supplements (65% of the adult Italian population) and 261 million packs (equal to 8 per capita) were sold in 2019 (190). At Italian pharmacies, the main sales channel, supplements are confirmed to be the second category after prescription drugs and give the greatest boost to growth, with 28.6 million medical prescriptions issued for food supplements in 2019 (190). Interestingly, the leading medical category that accounts for most of the 28.6 million medical prescriptions is represented by the general practitioners (21%) (190). Endocrinologists are absent in the first 8 positions, but the document fails to specify which categories are represented in the 9th category (“others”), a category that accounts for 12% of the prescriptions (190). In terms of comparison with the United States, based on data from a decade ago, annual supplement sales were USD 23 billion, with approximately 40,000 supplements products being on the market (191). In 2015, the American market for supplements was valued at USD 37 billion (151), but expected to reach USD 56.7 billion by 2024 (192) or USD 117.92 billion by 2027 (193).

With regard to the omega-3 products, their global market was valued at USD 2.10 billion in 2020 (USD 554.8 million in the US alone) (194), and it is projected to go up at a compound annual growth rate (CAGR) of 7.4% during the from 2020 to 2025 (195) and 7.8% from 2020 to 2028 (194). The main factors responsible for such increased consumption include the rising frequency of cardiovascular diseases (CVDs), changing dietary habits, the rising importance of immunity development post-COVID-19 and an increasing number of omega-3-based pharmaceutical product launches. DHA dominated the market in 2020, but EPA is expected to grow faster because of the increasing demand for immunity-boosting supplements (194).

The marine source segment of omega-3 supplements held the largest revenue share of over 82.8% in 2020. Fish oil (which contains both DHA and EPA) is the major marine source, and it is derived mainly from anchovy fish. Prices of omega-3 PUFA reflect the extraction and processing costs of fish oil, and they change depending on the availability of anchovy fish. Increasing contamination of fish by Hg and other pollutants is expected to negatively affect the prices of fish oil (194). Fish oils are currently considered the best and generally a safe source of omega-3 (196). However, the declining fish population due to overfishing has led to searching for more sustainable sources. Krill oil (which contains both DHA and EPA) and algae oil (which contains only DHA) have gained greater importance in recent years as alternatives, with algae serving as an option for the vegetarian population (197). In the US market, 9% of grocery shoppers buy high-omega-3 food or beverages in a typical shopping trip, with the proportion of adults who take fish oil supplements have increased from 8% in 2006 to 17% in 2011 (198). Omega-3 PUFA are also derived from plant sources including walnuts, pumpkin seeds, soybean oil, flaxseed oil, and canola oil, with plant oils being major sources of alpha-linolenic (ALA).

By comparison, the European market of the omega-3 products during the period (2019–2024) is expected to reach USD 14.61 billion by 2024 growing at a CAGR of 7% (199). The European demand for DHA is forecasted to increase significantly due to the favorable regulations in the European Union, which made DHA a mandatory ingredient in infant formula from 2020. Chia seeds are gaining popularity in Europe because of their nutritional and health properties that derive from their content in omega-3. Germany is the top European importer of chia seeds, preceding the Netherlands, Spain, and the United Kingdom (199).

In 2010, Friend of the Sea (FoS), an international nongovernment organization with the mission of promoting environmental conservation, introduced specific standards for producers of fish oil, fishmeal, fish feed and omega-3 supplements (200). Accredited third-party certification bodies certify that the oil originates from fisheries that are compliant with FoS sustainable fishing requirements, and that a full chain of custody occurs throughout the supply and the production chain (200). As of September 2018, 439 companies (compared to only 76 in 2015) adhere voluntarily to FoS standards for fish oil, fishmeal, fish feed and omega-3 supplement. Certified oils originate mostly from approved Peruvian anchovy fisheries and fleet (Engraulis ringens, 29%), Antarctic krill (Euphausia superba, 22%), European sardine (Sardina pilchardus, 8%), European anchovy (Engraulis encrasicolus, 7%), Chub mackerel (Scomber japonicus, 7%), Atlantic cod (Gadus morhua, 3%) (195). FoS presence in the nutraceutical industry has grown considerably in the United States, where it now accounts for over 50% of total FoS certified supplements (268 companies) (195). The United States are followed by France, Canada, Norway, United Kingdom and Italy, these six countries representing the top 6 countries for FoS labeled supplements.

Some studies are described now to illustrate the bioavalability of different types of omega-3 products. A 4-week randomized, placebo-controlled, double-blinded Icelandic study on 99 adults (of whom 77 completed the study) investigated the bioavailability of LC n-3 PUFAs from microencapsulated powder compared with ready-to-eat meals enriched with liquid fish oil (201). Participants were randomized into three groups 38 received 1.5 g/d EPA and DHA as meals enriched with liquid fish oil; 30 received the same amount of these LC n-3 PUFA as microencapsulated fish oil powder and regular meals; and 31 (controls) received placebo powder and regular meals. The authors found similar bioavailability between ready-to-eat meals enriched with liquid fish oil and LC n-3 PUFAs in encapsulated powder (201).

Starting from the fact that no conclusive information had been published on the relative bioavailability of omega-3 supplements taken as fish oil or krill oil, with few studies suggesting that the phospholipid form (krill) is absorbed better than the fish oil ethyl ester (EE) or triglyceride (TG) forms, Yurko-Mauro et al. (202) compared the oral bioavailability of the same dose of both DHA and EPA in fish oil-EE vs. fish oil-TG vs. krill oil after a four-week supplementation. In this double-blind, randomized, parallel study, 66 healthy adults were supplemented with a 1.3 g/day dose of EPA+DHA (approximately 816 mg/d EPA + 522 mg/d DHA, regardless of formulation) as either fish oil-EE (n=22 participants), fish oil-TG (n=22) or krill oil capsules (n=22). The authors found similar plasma and red blood cell levels of EPA+DHA across fish oil and krill oil products when matched for dose, DHA and EPA concentrations, indicating comparable oral bioavailability irrespective of formulation (197). For instance, mean total plasma DHA+EPA at 672 h were 90.9 ± 41 µg/mL (fish oil-EE), 108 ± 40 µg/mL fish oil-TG (fish oil-TG), and 118.5 ± 48 krill oil (krill oil), with a probability value just borderline significant (P= 0.052). Indeed, upon perusing figures in this paper (202), a decreasing hierarchy is evident for both plasma and red blood cell levels of DHA+EPA across all time points: krill oil > fish oil-TG > fish oil-EE. Furthermore, in a Canadian double-blinded, randomized, placebo-controlled, crossover trial on 24 volunteers, krill oil resulted more effective than fish oil in increasing n-3 PUFA both in plasma (P = 0.0043) and in red blood cells (P = 0.0011) (203). This study consisted of three treatment phases including krill or fish oil, each providing 600 mg of n-3 PUFA or placebo control, corn oil in capsule form (203).

DHA and EPA are added to several commercially available foods, such as infant and pet formulas, and they are also supplemented in animal feed to incorporate them in consumer dairy, meat, and poultry products (204). The main sources of EPA and DHA are fish oils or purified preparations from microalgae (204). In a one-month duration Australian study (205), 16 healthy males were provided with a range of foodstuffs naturally containing LC n-3 PUFA (fresh and canned fish, canola oil, flaxseed meal) and items fortified with fish oil (sausages, milk, margarine spread, luncheon meat, French onion dip); food choices were left to the discretion of each participant. Blood and cell levels of ALA, EPA and DHA increased highly significantly after 4 weeks (P<0.001) (205).

Because of the systemic action of thyroid hormones and their pleiotropic effects, thyroid disrupting chemicals (such as those mentioned in Table 3) represent a public health issue, making the comprehension of the mechanisms through which they interfere on thyroid homeostasis considerably importance.

Considering (i) the increasing incidence of both HT and TC worldwide (206–213); (ii) the aforementioned predisposing role for TC (particularly, papillary TC) and its advanced stages exerted not only by HT but also by even trendwise high serum TSH levels per se (65–88); (iii) the risk for developing metabolic and cardiovascular disorders conferred by both elevated/trendwise elevated serum TSH levels and thyroid autoimmunity (214–230), then it would be beneficial to contrast the appearance and/or duration of HT as well as to correct the slightly elevated serum TSH levels of subclinical hypothyroidism, the leading etiology of which is AIT. Furthermore, HT is frequently associated with other endocrine and nonendocrine autoimmune diseases, on which omega-3 PUFA proved to be beneficial [see above, Introduction]. For instance, 19.5% of 3,069 HT patients had evidence of at least one other autoimmune disease compared to 3.6% of 1,023 patients with multinodular goiter (231). As a corollary, there would be the place for the use of nutraceuticals to prevent/delay/minimize the onset/burden of autoimmune thyroiditis and the magnitude of TSH elevation. Taking into account their aforementioned anti-autoimmunity and anti-cancer (including TC) properties, the omega-3 PUFA would be appropriate nutraceuticals to be used, either alone or combined with other supplements.

Writing - original draft: SB and FF. Writing - review and editing: SB, FF, LP, AA, GB, FV and MM. Supervision: SB, FV and MM. All authors have read and approved the final manuscript and all materials before submission.

The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.

The reviewer SF declared a past collaboration with the authors AA, SB to the handling editor.

All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.