A critical review of food-associated factors proposed in the etiology of feline hyperthyroidism

Abstract

Since the first description of feline hyperthyroidism (HT) in 1979, several studies have been undertaken to define the etiology of the disease. Epidemiologic studies, after investigating non-food- and food-associated factors, suggest a multifactorial etiology. However, in the absence of prospective cohort studies that can confirm a cause-and-effect relationship between HT and associated risk factors, no causative factor for HT has been identified to date. Feline HT resembles toxic nodular goiter in humans, with autonomously functioning upregulated iodide uptake systems. Contribution of the diet to HT development remains controversial. The purpose of this paper is to review critically the reported food-associated risk factors for HT.

Introduction

Within 10 years of the first description of feline hyperthyroidism (HT) in 1979,¹ the first studies that reported dietary associated risk factors for HT were published.^2,3 Over the past two decades, several epidemiologic studies have since investigated other possible associations between diet and HT.^4–9 In these latter studies, a canned diet was found to be a food-associated risk factor. This led to hypotheses on what could be the etiologic factors for this association, such as potential contaminants like bisphenol-A (BPA) and polybrominated diphenyl ethers (PBDEs), isoflavones, or dietary levels of selenium and iodine, none of which have, to date, been confirmed by prospective studies. The purpose of this review is to provide an objective evaluation of the available information about the etiology of HT and especially the data supporting a potential role of diet.

Thyroid function and toxic nodular goiter

Thyroid hormones are produced in the colloid of thyroid follicles after stimulation by thyroid-stimulating hormone (TSH).¹⁰ TSH is released from the pituitary, after stimulation by thyrotropin-releasing hormone (TRH) from the hypothalamus. The thyroid hormones control the secretion of TRH and TSH by a negative feedback mechanism. TSH has several effects on the thyroid follicular cell. It binds to and activates the TSH receptors. TSH stimulates iodide uptake by enhanced transport by a sodium–iodide symporter, cell growth and hormone synthesis by iodide organification, as well as release of T3 and T4 into the bloodstream (Figure 1).

Figure 1

Hypothalamic–pituitary–thyroid axis and key steps in thyroid hormone formation. The thyroid is part of the hypothalamus–pituitary–thyroid axis. The hypothalamus releases thyrotropin-releasing hormone (TRH), which stimulates the pituitary to release thyroid-stimulating hormone (TSH). TSH, consisting of α and β subunits, binds to and activates the TSH receptor (TSHR) coupled to G proteins. Stimulation of the adenosine 3′,5′- cyclic monophosphate pathway results in enhanced iodide uptake, growth, differentiation and hormone synthesis. Iodide is actively transported into thyroid follicular cells by the sodium–iodide symporter (NIS) at the basolateral membrane. Thyroid peroxidase (TPO) oxidizes iodide and subsequently iodinates tyrosil residues of thyroglobulin (Tg) in the presence of hydrogen peroxide. The iodotyrosines, mono- and diiodothyrosil, are coupled to T4 and T3; this reaction is also catalyzed by TPO (coupling), and Tg is recycled in the follicular cell, and the thyronines T3 and T4 are released into the bloodstream. T3 and T4 have peripheric actions, but also act with a negative feedback on the release on TRH and TSH.¹¹ Solid lines: effect. Dashed lines: inhibition

Humans with chronic iodine deficiency can develop nodular goiter. Iodine deficiency causes TSH release and TSH stimulation with hypertrophy, hyperplasia and development of autonomous nodules in the thyroid gland. This is shown in Figure 2. This non-toxic nodular goiter becomes toxic with symptoms of human HT when iodine intake increases.²⁰ Adenomatous hyperplastic nodules in the thyroid of hyperthyroid cats resemble human toxic nodular goiter (TNG). There are comparable histopathologic features and mutations in the TSH receptor described in cats and humans.^21,22 Moreover, as observed with thyroid tissue from human patients with TNG, thyroid adenomatous tissue from cats with HT is autonomous in growth and function with upregulated systems for iodide uptake.^21,23

Figure 2

Causes and effects of preferential proliferation of thyrocytes. Several factors can cause increased thyroid-stimulating hormone (TSH)-dependent stimulatory pathways. These pathways or other factors like toxins, goitrogens or nutritional factors from the diet can cause preferential proliferation of certain thyrocytes or thyroid hyperplasia. Hyperplastic nodules of autonomous growing and functioning thyrocytes form, and clinical signs of hyperthyroidism develop. Solid lines: effect. Dashed lines: inhibition^12–19

Feline HT

Excessive production and secretion of thyroid hormones in HT is caused in 98% of cases by benign adenomatous hyperplasia of the thyroid gland.²⁴ Diagnosis is based on clinical signs caused by the increased level of thyroid hormones in the blood (polyphagia, polyuria, polydipsia, weight loss and hyperactivity), and confirmed by measurement of increased serum thyroid hormone concentration.²⁵

The etiology of HT is likely to be multifactorial. Since it was first diagnosed in 1979,¹ several case-control studies have been undertaken to find associated risk factors.^3,5–9 Non-dietary risk factors identified to be related to HT include increasing age;^3,5,7–9 indoor housing;^3,9 use of fertilizers, herbicides, flea powders and sprays;³ and use of cat litter.^6,9 Thyroid disruptors are defined as substances that alter the function of the thyroid glands. The increasing risk of HT with age may reflect chronic exposure to thyroid disruptors, which, over time, increases the risk of genetic mutations in thyrocytes. This may result in formation of hyperplastic nodules, adenomas and HT.⁹ An increased risk among indoor cats may reflect a greater exposure to a goitrogen found indoors, or a correlation to another unidentified risk factor.³

Besides the possibility of a goitrogen linked to cat litter or found elsewhere indoors, the use of cat litter and living indoors can also simply be markers of receiving better care; hence, cats enjoy longer lives and are more likely to reach the age when HT develops.⁶ Old cats that have multinodular follicular cell hyperplasia in their thyroid gland can develop functional thyroid adenomas.²⁶ Nearly every long-standing goiter and even most normal thyroids of elderly humans become nodular after a certain amount of time.¹²

The prevalence of HT differs among different regions. It is more frequently diagnosed in the UK compared with Denmark, and more in Eastern than Western Australia.²⁷ Shortly after the first description of HT in the USA, a study was published investigating the geographic distribution, which showed the incidence in the USA was higher on the west coast and the Great Lakes regions compared with more central states.³ A possible difference in regional prevalence should be taken into account when evaluating epidemiologic studies. Most studies focus either on prevalence or etiology in a specific area. Only the USA,^3,5 the UK,^9,28 Ireland²⁹ and Hong Kong⁴ are the areas in which both etiology and prevalence have been investigated.

Study design for risk factor assessment

There are two principal types of study used to assess risk factors: retrospective case-control studies and prospective cohort studies. All studies investigating risk factors for HT are retrospective case-control studies with data collection mostly performed using a questionnaire. In these studies the association between a disease and risk factor can be estimated by calculating the odds ratio (OR). Cats, preferably from the same population, are grouped based on the presence or absence of HT. OR is the ratio of the odds of exposure to a factor in cases and controls. If the OR is significantly greater than 1, the cases have been more exposed to this factor.³⁰ Case-control studies are sensitive to bias, and the exposure in the control group should be representative of the population.

Risk factors suggested from retrospective studies are not only related to the host, but also to the environment and diet. This is strengthened by HT being a ‘new’ disease, as first described in 1979.¹ It should be highlighted, however, that thyroid abnormalities were becoming more common by the mid-1960s.³¹ Therefore, it is nearly impossible to design one multivariate logistic model that includes all potential confounding factors.⁵ A low prevalence of disease is generally considered to be below 10%, and in this instance retrospective studies are less likely to identify etiologic factors. The most recent prevalence of HT reported is 1% of all cats and 6% in cats over 10 years of age.³²

Prospective cohort studies follow patients over time and allow an evaluation of the cause and effect relationship between risk factors (classically highlighted in a retrospective case-control study) and the development of disease. The drawback of this design is the delay before the appearance of the disease, which makes it less adapted to rare conditions, as well as those occurring late in life. To date, no prospective study on HT has been performed in cats, and therefore no true causative factor has been identified.

Food-associated risk factorsin feline HT

The proposed mechanism for development of HT due to dietary factors in most retrospective studies seems to be linked to decreased serum thyroid hormone concentrations, which would affect the negative feedback by thyroid hormones and stimulate TSH secretion. This chronic thyroid overstimulation would be responsible for hypertrophy and then hyperplasia of thyroid follicular cells (Figure 2).¹³ This mechanism is supported by the fact that both glands are affected,³³ and also the bilateral altered expression of G proteins in follicular tissue.^34,35 It does not explain a possible pre-existing TSH receptor mutation, mostly described in unilateral disease.²² Neither this mechanism nor parts of it have been investigated, or proven to be present in cats developing HT. This is mainly owing to the low sensitivity of the currently available TSH assays.²⁸

Canned food

Several studies have reported an OR significantly >1 for canned cat food as part or as the sole constituent of the diet.^3,5,6,9 Within canned foods, specific flavors like fish, liver or poultry,⁷ as well as a large variety of flavors,⁸ have also been associated with an increased risk.

All of the above studies have limitations in their design and materials and methods. These include, for instance, insufficient recording of dietary history,^6,8 poor matching of age between cases and controls,^5,9 some unhealthy cats in the control group⁵ or lack of information about the health status of control cats.⁸ The study of Martin et al only evaluated the current preference of wet food and therefore did not necessarily reflect consumption during the period before developing HT, did not analyze the significance of the OR and an increased risk (without listing the P value) was noted for every flavor listed.⁷ In some studies cases outnumbered controls.^6,9 Results of studies finding a significant increased risk associated with consumption of a wet diet, as well as the limitations of these studies, are summarized in Table 1.

Table 1

Studies describing an increased risk (odds ratio [OR] >1) with statistical significance (P <0.05) for wet food as part of the complete diet

Study	Cases; controls (n, total)	Cases; controls	% Wet in diet	OR	P value	95% CI	Study limitations
Scarlett et al³	56; 117	12; 29	0–50	1.60	0.020	0.50–5.70	Sick cats in controls
Scarlett et al³	56; 117	38; 55	⩾50	3.40	0.020	1.10–10.40	Sick cats in controls
Kass et al⁶	379; 351	59; 39	50–74	2.88	0.010	1.26–6.60	Current diet may not reflect diet history
Kass et al⁶	379; 351	138; 106	75–100	2.09	0.040	1.03–4.23	Current diet may not reflect diet history
Edinboro et al⁵	109; 173	43; 52	⩽50	2.74	0.002	1.43–5.27	Age difference between cases and controls
		17; 16	50	3.52	0.004	1.50–8.28
		25; 25	>50	3.32	0.002	1.56–7.06
Olczak et al⁸	125; 250	125; 125*	Variety of flavors in wet	3.80	0.005	1.50-–9.60	Current diet may not reflect diet history Health status of controls missing
Wakeling et al⁹	109; 196	30; 59	50	2.30	0.040	1.00–5.00^†	Age difference between cases and controls
		33; 39	51–99	3.80	0.001	1.70–8.40^†
		26; 8	100	14.50	<0.001	5.20–40.40^†
		54; 24	>50	2.80	0.001	1.50–5.20^‡

Control group consisted of 125 sex- and age-matched control cats from the total of 250 control cats

†

Unadjusted OR from univariable analysis

‡

Adjusted OR from multivariable analysis

CI = confidence interval

The risk of HT increases with age.⁹ An age difference between case and control groups can cause selection bias owing to possible interdependence between age and another risk factor. Sick cats in the control group may have non-thyroidal illness. Concomitant non-thyroidal illness can suppress serum total T4 concentration into reference range values, even in a cat with HT, and can therefore result in classifying them as controls rather than cases.³⁶

About 70% of European cat owners follow a mixed feeding regime for their cats, and this applies for all life stages. Periodontal disease affects over 60% of cats older than 3 years.³⁷ Dental and other intraoral disorders have been described to be over-represented in hyperthyroid cats compared with healthy controls.⁸ A preference for softer (wet) food might be associated with poorer dental health, or there might be presence of an unknown protective factor in dry food.

Nine criteria (Bradford Hill criteria) are used in epidemiology for assessing whether an association involves a causal component: strength, temporality, consistency, specificity, dose–response, plausibility, coherence, experiment and analogy.³⁸ The more criteria that are met, the stronger the evidence for a causal component will be. In epidemiologic HT studies that reported an OR significantly >1 for canned food as part or the sole constituent of the diet, only three of those criteria have been met:

Strength – a higher OR means a stronger association and is therefore more likely to have a causal component. ORs range from 2.1–3.8 up to 14.5 when fed a 100% wet diet;^3,5,6,8,9

Temporality – HT is observed after exposure to canned diets;

Consistency – an OR >1 for wet has been found in populations from the USA,^3,5–7 UK⁹ and New Zealand.⁸ In the last study, an OR >1 (3.8) was found for feeding a variety of canned cat food flavors in cases compared with age- and sex-matched controls, but feeding canned food >50% with a non-significant OR >1 (2.1) was only investigated in cases compared with non-matched controls. It should be noted that the incidence of HT varies between and within different regions of the world,^{3–5,27,28,39} and there are only a few areas where both prevalence and etiology have been investigated. This warrants further research to investigate consistency in other geographical areas and populations.

The remaining six criteria were not met for the following reasons:

Specificity – HT is a multifactorial disease. A simple factor affecting a specific population at a specific site has not been found;

Dose–response – no dose–response study investigating the link between the amount of canned food in the diet and the course of disease or severity of symptoms of HT has yet been performed;

Plausibility – several hypotheses have been suggested but not tested;

Coherence – because of the variability in the methodology used in the various studies, it is difficult to compare findings;

Experiment – causation is more likely if it can be confirmed by prospective randomized experiments. These have not been performed for wet food and HT in cats;

Analogy – the same effect is not shown for comparable exposures and outcomes. Not every cat exposed to canned food will develop HT.

Constituents associated with canned food

The monomer BPA is used to make epoxy resins by linking two BPA molecules with epichlorhydrine to make bisphenol-A-diglycidyl ether (BADGE).⁴⁰ BPA has structural similarity to thyroid hormones and can act as a thyroid hormone receptor antagonist.⁴¹ BPA is therefore suspected to be an endocrine disruptor and associated with thyroid dysfunction in humans. This has led, in recent years, to the removal of BPA from plastics intended to contain human food and drink. The US Food and Drug Administration is also taking steps to reduce human BPA exposure in the food supply and supports recommendations to reduce exposure in infant formulas and food preparation.⁴² The European Union (EU) restricts BPA use in plastic infant feeding bottles.⁴³

To date, no study has been carried out to show BPA in the blood or tissue of cats with HT. European regulations for BADGE allow a maximum migration of BPA into canned wet food of 9 µg/g (EU Directive 1895/2005). BPA has been detected in some canned cat food in Japan at a concentration of 0.013–0.136 µg/g,⁴⁰ and in the USA at a concentration ranging from below the detection limit (<0.002 µg/g) to 0.032 µg/g,⁴⁴ well below the current EU maximum. A suspected relationship between can size and the amount of BPA has so far not been substantiated,¹³ and is based on non-scientific studies. A significant association was found for consumption of pop-top cans and HT.⁵ Pop-top cans are typically coated with organosol coatings, which is needed for flexibility, and not with the more rigid BPA-containing epoxy resins. The low level of exposure does not strongly support BPA in adult life stages as a cause of HT. However, in humans, it is the exposure during gestation and development that has been more associated with thyroid disorders.⁴⁵ Exposure to BPA early in life, combined with the slow drug metabolism for phenolic compounds in cats,⁴⁶ remains to be investigated as a risk factor for HT.

PBDEs are flame retardants and known endocrine disruptors owing to the structural similarities of various PBDE isomers with thyroid hormone. No case-control studies have been carried out to show an association between PBDE exposure and HT. A recent study found that total serum PBDE in feral cats was significantly lower than in client-owned cats. Moreover, total PBDE concentrations were significantly higher in dust from homes of hyperthyroid cats compared with euthyroid ones.⁴⁷ PBDE has been found in the serum of hyperthyroid cats from California,⁴⁷ as well as from Sweden⁴⁸and Pakistan,⁴⁹ but there was no significant difference in serum PBDE between healthy cats and cats with HT.^47,50–52 PBDEs have been analyzed in cat food, ranging from 0.17–1.75 ng/g and from 0.6–2.9 ng/g in canned and dry food of different flavors, respectively,⁵⁰ and from 0.42–3.10 ng/g in fish-based canned foods.⁴⁷ The theoretical PBDE intake from wet food represents only 0.017% of the lowest single acutely toxic PBDE dose in laboratory animals.⁵³ The PBDE values in canned food are relatively small compared with PBDE found in dust from the homes of hyperthyroid cats, being 1100–95,000 ng/g. When the exposure of domestic cats to PBDE from canned food and dust is estimated, it is suggested that the ingestion of household dust is the primary source. Dust PBDE, and not serum PBDE, was significantly correlated with serum T4.⁴⁷ These findings do not support canned food PBDE as the causal factor for HT. Because BPA and PBDE have similar structures, a concomitant exposure can be hypothesized to have additive effects.

Flavonoids

Flavonoids are phyto-estrogens that can be divided into flavones, flavanones and isoflavones. Soybeans contain the isoflavones daidzein, genistein and biochanin A. Several effects of isoflavones on thyroid hormone metabolism have been proven, though others can still be hypothesized. In vitro studies have demonstrated that these isoflavones inhibit thyroid peroxidase, owing to the similarity between their chemical structure and the structure of T3 and T4, thereby acting as an alternative substrate for iodination.^54–56 A soybean supplemented and iodine-deficient diet synergistically stimulated growth of the thyroid gland in rats;^57,58 however, a further study suggested that this effect was not due to the isoflavones in soybean alone.⁵⁹ The inhibition of thyroperoxidase due to isoflavones genistein and daidzein was reversed following the addition of iodine in studies performed in vitro.^54,55 These data suggest that a goitrogenic effect of soy is possible when there is a deficiency in dietary iodine. Soy isoflavones have also been described to inhibit 5′-deiodinase, the enzyme responsible for conversion of T4 to biologically active T3,⁶⁰ in pigs,⁶¹ hamsters^62,63 and rats.⁵⁹ These actions can decrease the concentrations of thyroid hormones with subsequent stimulation of TSH production.

Two studies showed that genistein and daidzein were present in cat foods from New Zealand⁶⁴ and the USA.⁶⁵ In both studies, the authors hypothesized that the amounts detected can have a biological effect; however, this is based on extrapolations from other species. Measurable isoflavones were present in a higher number of dry compared with wet diets.⁶⁵ This contradicts the increased risk of HT associated with canned food described earlier.^3,5,6,8,9 To date, there has only been one study evaluating the effect of soy in diet on thyroid function.⁶⁶ This short-term study in healthy cats showed a slight but significant increase in total T4 and free T4 concentrations without changes in T3 in cats receiving the soy-containing diet compared with cats receiving a soy-free diet. With inhibition of deiodination, T3 would be expected to decrease, but as a decrease in T3 triggers TSH release, the T3 concentrations are restored but with an additional release of T4 from the thyroid.⁶⁶ Besides the difference in soy content, the diets also differed in other factors such as the content of iodine and fish meal. Moreover, the concentration of daidzein and genistein in the soy-containing diet was much higher than levels reported in commercial feline diets,^64,65 and the iodine content of the soy-containing diet was higher than the legal maximum concentration.

In the study in healthy cats, the ingredients contributing to isoflavone content in the soy-containing diet were soybean meal and soy flakes.⁶⁶ When purified sources of soy, like soy isolates and soy hydrolysates, are used, they contain naturally limited levels of isoflavones because the protein concentration is higher and therefore the isoflavone concentration lower. As early as the 1960s, the goitrogenic effects of soy-based infant formula were overcome by using soy protein isolate instead of soy flour, thereby reducing the goitrogenic constituents of the formula.⁶⁷ Besides the type of raw material used, a variation in the composition of isoflavones due to, for instance, genetic and environmental influences can cause changes in isoflavone content in raw materials.⁶⁸

Selenium in the diet

Selenium is known to be implicated in thyroid hormone metabolism. Selenium deficiency is hypothesized to cause increased synthesis of T3 and T4 by the thyroid gland due to a decrease in selenoperoxidase acting on the rate-limiting step of thyroid hormone synthesis. However, selenium deficiency is also hypothesized to suppress activity of thyroxine 5′-deiodinase, which is a selenoenzyme, with decreased conversion of T4 to T3 in peripheral tissues and increased TSH stimulation. High selenium intake can result in overexpression of thioredoxin reductase, which is a selenoprotein, with increased cell growth and autonomous nodules in the thyroid.²⁷

To date, only two studies have investigated the effects of selenium in the diet on thyroid function in cats. One study found a significant increase in serum total T4 and decrease in serum total T3 in kittens fed a low selenium diet for 8 weeks.⁶⁹ A second study looked at whole blood and plasma selenium concentrations in 43 euthyroid cats and seven hyperthyroid cats from four global regions with different incidences of HT. The authors found no significant difference between cats from different regions, nor between euthyroid and hyperthyroid cats from regions with a high incidence of HT. Sample sizes of euthyroid and hyperthyroid cats were relatively small and not matched for age. Such as for flavonoids, it remains unclear what the role of selenium is in the multifactorial etiopathology of HT.²⁷

Iodine level in commercial cat food

Dietary iodine has been proposed as a causal factor of HT in a variety of ways: fluctuant iodine intake, as well as insufficient or excessive iodine (especially in fish-flavored wet diets) intake would contribute to the development of thyroid disorders. Specific flavors like fish,⁷ or a variety of flavors in wet food,⁸ have been associated with an increased risk for HT, and this has strengthened the hypothesis that dietary iodine could affect thyroid function and HT. Ocean fishes tend to be high in iodine.⁷⁰

In the first editions of the National Research Council’s (NRC) publications,^71,72 iodine requirements were based on studies showing that supplementation of 100 µg of iodine per day in iodine-deficient kittens (being fed heart exclusively) prevented hyperplastic changes.^73,74 This dosage can be estimated as 2.8 mg/kg dry matter (DM) or 6.4 mg/Mcal. In 1978 and 1986, the NRC reported minimum recommended iodine levels in published literature ranging from 1.40–4.00 mg/kg DM (0.35–1.0 mg/Mcal based on a diet of 4000 kcal/kg), and that the minimum requirement for kittens was 0.35 mg/kg dry diet (0.07 mg/Mcal based on a diet containing 5000 kcal/kg).^75,76 In 2006, the value was revised based on the study by Ranz et al in 2002⁷⁷ to 0.35 mg/Mcal (1.4 mg/kg dry diet of 4000 kcal/kg).⁷⁸ The recommendations from the Fédération Européenne de l’Industrie des Aliments pour Animaux Familiers (FEDIAF, 2008, 2011) were below the NRC’s (2006)⁷⁸ until recently,^79,80 when the FEDIAF (2012)⁸¹ increased the minimum recommended concentration to levels equal to the NRC recommendations (Table 2).^79–81 No safe upper limit has been defined by the NRC. European authorities established a legal maximum (when supplemented) of 11 mg iodine per kg DM or 2.8 mg/Mcal metabolizable energy.⁸¹

Table 2

Historical overview of recommended minimum and legal maximum iodine concentration in commercial cat food from the National Research Council (NRC) and the Fédération Européenne de l’Industrie des Aliments pour Animaux Familiers (FEDIAF)

	Minimum concentration		Maximum concentration
	mg/kg dry diet	mg/Mcal	mg/kg dry diet	mg/Mcal
NRC (1962),⁷¹ (1972)⁷²	2.8 (kittens)*	6.4 (kittens)*	Not defined	Not defined
NRC (1978),⁷⁵ (1986)⁷⁶	1.40–4.00	0.35–1.0	Not defined	Not defined
	0.35 (kittens)	0.07 (kittens)
NRC (2006)⁷⁸	1.40	0.35	Not defined	Not defined
FEDIAF (2008)⁷⁹	0.60	0.15	11	2.8
FEDIAF (2011)⁸⁰	0.60	0.15	11	2.8
FEDIAF (2012)⁸¹	1.30	0.33	11	2.8

Supplementation of 100 μg of iodine per day in kittens that were iodine deficient, after being fed heart exclusively, prevented hyperplastic changes in the thyroid^73,74

DM = dry matter

Wide ranges in iodine content have been found in commercial cat food in New Zealand (1.0–36.8 mg/kg DM),⁸² Germany (0.22–6.40 mg/kg DM in canned food and 0.5–3.2 mg/kg DM in dry food),⁷⁷ and two studies from the USA (0.71–21.19 mg/kg DM² and 0.2–154.6 mg/kg DM).¹⁴

These findings led to the hypothesis that fluctuations in dietary iodine content over time could affect thyroid function, leading successively to hyperplasia, adenomatous changes and HT (Figure 2).^8,83 The ability of cats to respond or adapt to those fluctuations in iodine could differ, thereby implicating host-related factors in HT.¹⁵ It can be seen from Table 2 that iodine recommendations have varied over the years.⁸³

Excessive iodine intake

Extrapolating from humans, high iodine ingestion has also been suggested as a cause of HT. In humans with healthy thyroid function or a thyroid disorder that is still sensitive to control by TSH from the pituitary, administration of large quantities of iodine result in an inhibition of hormone synthesis (Wolff–Chaikoff effect).⁸⁴ An escape phenomenon causes this inhibition to be transient and thyroid hormone synthesis resumes after approximately 48 h.⁸⁵ In humans with a pre-existing non-toxic nodular goiter it becomes toxic with symptoms of HT when iodine intake increases.²⁰

When three diets with different levels of iodine (0.10, 2.22 and 13.79 mg/kg DM) were fed successively, each over 2 weeks, in healthy cats, serum-free T4 concentrations responded inversely to dietary iodine.⁸⁴ Those results were recently confirmed by increased dietary iodine inducing decreased concentrations of T4, T3 and free T3 in healthy cats.⁷⁷ In a study conducted in cats with low (0.11 mg/kg DM) and high (21.14 mg/kg DM) iodine diets fed for 5 months, no significant difference in serum free T4 was noted between the diets.¹⁵ Analysis of difference between time points within diet is not reported, so the escape mechanism cannot be evaluated in those cats. Serum free T4 repeatedly increases and subsequently decreases when fed the diet high in iodine. This suggests discontinuous TSH stimulation to maintain thyroid hormone levels in physiologic ranges.^15,86 This TSH stimulation might cause hypertrophy, hyperplasia and development of autonomous nodules in the thyroid, as shown in Figure 2.

Excessive iodine intake as a cause of HT was also proposed when discussing the association between HT and canned food, especially regarding the stronger association with fish flavors.^3,7 Indeed, ocean fishes tend to be high in iodine.⁷⁰ The legal definition of ‘flavor’ is qualitative and not quantitative and therefore pet food labels do not provide the absolute contribution of fish ingredients to the total iodine content.² In a retrospective study, no increased risk for HT was found for tuna flavor.⁷ The studies from Johnson et al and Mumma mention whether fish is in the ingredient list,^2,82 though they did not investigate the effect of fish as an ingredient on the iodine concentration. Commercial diets containing fish can have wide ranges in iodine content, from 0.94–21.2 mg/kg DM⁸² and from 1.8–36.8 mg/kg DM.² A recent study showed that median iodine intake on a DM basis and estimated daily iodine intake were not significantly different for foods with fish ingredients compared with foods without them¹⁴. Moreover, adequate iodine concentration (as estimated by the respective author to be ⩾100 µg/day) was not guaranteed by the presence of seafood in the ingredient list. Median iodine concentrations were higher in foods without fish than in foods with fish listed as one of the first five ingredients.¹⁴

Iodine deficiency as a cause of HT

Humans with endemic goiter due to iodine deficiency will develop increased thyroid hormone synthesis in the autonomous nodules (iodine-induced HT) when small increases in iodine intake occur (Jod-Basedow effect).⁸⁷ This HT develops in the thyroid gland without control of the pituitary TSH. Iodine-induced HT in a human population has a transient nature, with a decreasing prevalence back to baseline values 10–20 years after the start of the increased iodine intake.^6,16,20,88 To date, this decreased prevalence has not been described in cats.

A recent retrospective case-control study showed that cats fed commercial diets, without iodine supplementation according to listed ingredients, were four times more likely to develop HT compared with cats eating iodine-supplemented cat foods.⁸⁹ Feline HT resembles TNG in humans with thyrotoxicosis and thyroid nodules. Further research is needed to investigate whether long-term inadequate iodine intake can cause pathologic changes in the thyroid due to chronic stimulation by TSH, with the development of HT after switching to iodine-rich diets, as is the case in humans (Figure 2).

Wakeling et al compared urinary iodine excretion between hyperthyroid and euthyroid cats,⁹ and although the 95% ranges overlapped, the mean urinary iodine excretion was significantly lower in hyperthyroid cats compared with euthyroid ones. The authors suggested that the lower urinary iodine excretion observed in hyperthyroid cats could reflect a period of low iodine ingestion with minimal iodine storage. Another explanation could be the increased thyroid mass that is hyperactive and taking up large amounts of iodine, thereby leaving less iodine available for excretion. Wedekind et al looked at urinary iodine concentrations of healthy cats fed diets with iodine content ranging from 150–9240 µg/kg DM.⁹⁰ The cats eating diets with iodine content ranging from 170–460 µg/kg DM (below the FEDIAF’s current minimum recommended nutrient level of 1.3 mg/kg DM) showed thyroid/salivary gland technetium uptake ratios (T/S ratios) ranging from 2.59–1.43 after 12 months on the diet.⁹⁰ T/S ratios ⩾1.2 have been described in cats with HT.^91,92 The increased T/S ratio in HT is likely due to an overactivity of the thyroid and thyroid hormone metabolism, while in healthy cats the T/S ratio can be increased by a higher number of sodium–iodine symporters during iodine deficiency.

Conclusions

Epidemiologic studies investigating environmental and nutritional exposures suggest HT to be a complex multifactorial disease. Several risk factors have been associated with HT, including diet and feeding canned food. However, retrospective studies have their limitations in design, which has led to the search for other dietary risk factors. None of those risk factors have yet been confirmed to induce HT. Lifelong prospective longitudinal studies based on the strongest epidemiologic evidence to date are needed to investigate the role of nutritional risk factors in the development of HT.

Footnotes

Acknowledgements

This review has been written after a request from the Fédération Européenne de l’Industrie des Aliments pour Animaux Familiers (FEDIAF). The FEDIAF is the European pet food industry federation, representing this industry in 26 countries via 18 national or regional associations. The aim of FEDIAF is to support a legislative framework for the production of safe, nutritious and palatable pet food. This review fulfills the need for an objective evaluation of the available information on feline hyperthyroidism and its suggested association with petfood.

Conflict of interest

I van Hoek and V Biourge were employees of Royal Canin SAS at the time of writing the manuscript.

Funding

This research received no specific grant from any funding agency in the public, commercial or not-for-profit sectors.

Accepted: 29 September 2014

References

Peterson

Johnson

Andrews

. Spontaneous hyperthyroidism in the cat. Proceedings of the American College of Veterinary Internal Medicine. Seattle, WA: ACVIM, 1979, p 108.

Mumma

Rashid

Shane

, et al. Toxic and protective constituents in pet foods. Am J Vet Res 1986; 47: 1633–1637.

Scarlett

Moise

Rayl

. Feline hyperthyroidism: a descriptive and case-control study. Prev Vet Med 1988; 6: 295–309.

De Wet

Mooney

Thompson

, et al. Prevalence of and risk factors for feline hyperthyroidism in Hong Kong. J Feline Med Surg 2009; 11: 315–321.

Edinboro

Scott-Moncrieff

Janovitz

, et al. Epidemiologic study of relationships between consumption of commercial canned food and risk of hyperthyroidism in cats. J Am Vet Med Assoc 2004; 224: 879–886.

Kass

Peterson

Levy

, et al. Evaluation of environmental, nutritional, and host factors in cats with hyperthyroidism. J Vet Intern Med 1999; 13: 323–329.

Martin

Rossing

Ryland

, et al. Evaluation of dietary and environmental risk factors for hyperthyroidism in cats. J Am Vet Med Assoc 2000; 217: 853–856.

Olczak

Jones

Pfeiffer

, et al. Multivariate analysis of risk factors for feline hyperthyroidism in New Zealand. N Z Vet J 2005; 53: 53–58.

Wakeling

Everard

Brodbelt

, et al. Risk factors for feline hyperthyroidism in the UK. J Small Anim Pract 2009; 50: 406–414.

10.

Cohen

Wondisford

. Factors that control thyroid function. In: Braverman

Utiger

(eds). The thyroid A fundamental and clinical text. Philadelphia, PA: Lippincott Williams and Wilkins, 2005, pp 159–213.

11.

Campos

van Hoek

Peremans

, et al. Recombinant human thyrotropin in veterinary medicine: current use and future perspectives. J Vet Intern Med 2012; 26: 853–862.

12.

Ferguson

. Pathogenesis of hyperthyroidism. In: August

(ed). Consultations in feline medicine. St Louis, MO: WB Saunders, 1994, pp 133–140.

13.

Peterson

. Hyperthyroidism in cats: what’s causing this epidemic of thyroid disease and can we prevent it? J Feline Med Surg 2012; 14: 804–818.

14.

Edinboro

Pearce

Pino

, et al. Iodine concentration in commercial cat foods from three regions of the USA, 2008–2009. J Feline Med Surg 2013; 15: 717–724.

15.

Tarttelin

Ford

. Dietary iodine level and thyroid function in the cat. J Nutr 1994; 124: 2577S–2578S.

16.

Peterson

Randolph

Mooney

. Endocrine diseases. In: Sherding

(ed). The cat: diseases and clinical management. 2nd ed. New York: Saunders, 1994, pp 1403–1506.

17.

Ferguson

Freedman

. Goiter in apparently euthyroid cats. In: August

(ed). Consultations in feline internal medicine. St Louis, MO: Elsevier Saunders, 2006, pp 207–215.

18.

Peterson

Ward

. Etiopathologic findings of hyperthyroidism in cats. Vet Clin North Am Small Anim Pract 2007; 37: 633–645.

19.

Wakeling

Elliott

Petrie

, et al. Urinary iodide concentration in hyperthyroid cats. Am J Vet Res 2009; 70: 741–749.

20.

Laurberg

Bülow Pedersen

, et al. Environmental iodine intake affects the type of nonmalignant thyroid disease. Thyroid 2001; 11: 457–469.

21.

Peter

Gerber

Studer

, et al. Autonomy of growth and of iodine metabolism in hyperthyroid feline goiters transplanted onto nude mice. J Clin Invest 1987; 80: 491–498.

22.

Watson

Radford

Kipar

, et al. Somatic mutations of the thyroid-stimulating hormone receptor gene in feline hyperthyroidism: parallels with human hyperthyroidism. J Endocrinol 2005; 186: 523–537.

23.

Peter

Gerber

Studer

, et al. Autonomous growth and function of cultured thyroid follicles from cats with spontaneous hyperthyroidism. Thyroid 1991; 1: 331–338.

24.

Meeking

. Thyroid disorders in the geriatric patient. Vet Clin North Am Small Anim Pract 2005; 35: 635–653.

25.

Thoday

Mooney

. Historical, clinical and laboratory features of 126 hyperthyroid cats. Vet Rec 1992; 131: 257–264.

26.

Apen

. Pathophysiology of the thyroid gland. In: Dunlop

Malbert

C-H

(eds). Veterinary pathophysiology. Ames, IA: Wiley-Blackwell, 2004, pp 444–469.

27.

Foster

Thoday

Arthur

, et al. Selenium status of cats in four regions of the world and comparison with reported incidence of hyperthyroidism in cats in those regions. Am J Vet Res 2001; 62: 934–937.

28.

Wakeling

Elliott

Syme

. Evaluation of predictors for the diagnosis of hyperthyroidism in cats. J Vet Intern Med 2011; 25: 1057–1065.

29.

Gallagher

Mooney

. Prevalence and risk factors for hyperthyroidism in Irish Cats from the Greater Dublin area. ACVIM Forum; 2013 Jun 13–15; Seattle, WA, USA.

30.

Lilienfeld

Stolley

. Observational studies: I. Cohort studies. In: Foundations of epidemiology. 3rd ed. New York: Oxford University Press, 1994, pp 198–225.

31.

Feldman

Nelson

. Canine and feline endocrinology and reproduction. 3rd ed. St Louis, MO: Saunders, 2004.

32.

Hospital

. State of pet health report. http://www.stateofpethealth.com/content/pdf/Banfield-State-of-Pet-Health-Report_2013.pdf (2013, accessed 2014).

33.

Peterson

Broome

. Thyroid scintigraphy findings in 2096 cats with hyperthyroidism. Vet Radiol Ultrasound 2015; 56: 84–95.

34.

Hammer

Holt

Ward

. Altered expression of G proteins in thyroid gland adenomas obtained from hyperthyroid cats. Am J Vet Res 2000; 61: 874–879.

35.

Ward

Achenbach

Peterson

, et al. Expression of inhibitory G proteins in adenomatous thyroid glands obtained from hyperthyroid cats. Am J Vet Res 2005; 66: 1478–1482.

36.

Peterson

Gamble

. Effect of nonthyroidal illness on serum thyroxine concentrations in cats: 494 cases (1988). J Am Vet Med Assoc 1990; 197: 1203–1208.

37.

Clarke

Cameron

. Relationship between diet, dental calculus and periodontal disease in domestic and feral cats in Australia. Aust Vet J 1998; 76: 690–693.

38.

Thygesen

Andersen

. A philosophical analysis of the Hill criteria. J Epidemiol Community Health 2005; 59: 512–516.

39.

Wakeling

Melian

Font

. Evidence for differing incidences of feline hyperthyroidism in London UK and Spain. European Congress of Veterinary Internal Medicine; 2005; Glasgow, UK. Poster abstract 43, p 220.

40.

Kang

Kondo

. Determination of bisphenol A in canned pet foods. Res Vet Sci 2002; 73: 177–182.

41.

Moriyama

Tagami

Akamizu

, et al. Thyroid hormone action is disrupted by bisphenol A as an antagonist. J Clin Endocrinol Metab 2002; 87: 5185–5190.

42.

US Food and Drugs Administration. Bisphenol A (BPA): use in food contact application. http://www.fda.gov/NewsEvents/PublicHealthFocus/ucm064437.htm2013 (2013, accessed June 2014).

43.

European Union. Amending Directive 2002/72/EC as regards the restriction of use of Bisphenol A in plastic infant feeding bottles. Official Journal of the European Union. Commission directive 2011/8/EU. http://eur-lex.europa.eu/legal-content/EN/TXT/HTML/?uri=CELEX:32011L0008&from=EN. (accessed 2014).

44.

Schecter

Malik

Haffner

, et al. Bisphenol A (BPA) in U.S. food. Environ Sci Technol 2010; 44: 9425–9430.

45.

Lakind

Naiman

. Daily intake of bisphenol A and potential sources of exposure: 2005–2006 National Health and Nutrition Examination Survey. J Expo Sci Environ Epidemiol 2011; 21: 272–279.

46.

Court

Greenblatt

. Molecular basis for deficient acetaminophen glucuronidation in cats: an interspecies comparison of enzyme kinetics in liver microsomes. Biochem Pharmacol 1997; 53: 1041–1047.

47.

Mensching

Slater

Scott

, et al. The feline thyroid gland: a model for endocrine disruption by polybrominated diphenyl ethers (PBDEs)? J Toxicol Environ Health A 2012; 75: 201–212.

48.

Norrgran

Jones

Lindquist

, et al. Decabromobiphenyl, polybrominated diphenyl ethers, and brominated phenolic compounds in serum of cats diagnosed with the endocrine disease feline hyperthyroidism. Arch Environ Contam Toxicol 2012; 63: 161–168.

49.

Ali

Malik

Mehdi

, et al. Organohalogenated contaminants (OHCs) in the serum and hair of pet cats and dogs: biosentinels of indoor pollution. Sci Total Environ 2013; 449: 29–36.

50.

Dye

Venier

Zhu

, et al. Elevated PBDE levels in pet cats: sentinels for humans? Environ Sci Technol 2007; 41: 6350–6356.

51.

Kupryianchyk

Hovander

Jones

, et al. Hyperthyroidism, a new disease in cats – is it caused by exposure to environmental organic pollutants? Organohalogen Compounds 2009; 71: 6.

52.

Guo

Park

Wang

, et al. High polybrominated diphenyl ether levels in California house cats: house dust a primary source? Environ Toxicol Chem 2012; 31: 301–306.

53.

Mensching

Beasley

. The feline thyroid gland: a model for endocrine disruption by polybrominated diphenyl ethers? http://hdl.handle.net/2142/10742 (2009, accessed 2014).

54.

Divi

Chang

Doerge

. Anti-thyroid isoflavones from soybean: isolation, characterization, and mechanisms of action. Biochem Pharmacol 1997; 54: 1087–1096.

55.

Divi

Doerge

. Inhibition of thyroid peroxidase by dietary flavonoids. Chem Res Toxicol 1996; 9: 16–23.

56.

Doerge

Sheehan

. Goitrogenic and estrogenic activity of soy isoflavones. Environ Health Perspect 2002; 110 Suppl 3: 349–353.

57.

Ikeda

Nishikawa

Imazawa

, et al. Dramatic synergism between excess soybean intake and iodine deficiency on the development of rat thyroid hyperplasia. Carcinogenesis 2000; 21: 707–713.

58.

Ikeda

Nishikawa

Son

, et al. Synergistic effects of high-dose soybean intake with iodine deficiency, but not sulfadimethoxine or phenobarbital, on rat thyroid proliferation. Jpn J Cancer Res 2001; 92: 390–395.

59.

Son

Nishikawa

Ikeda

, et al. Lack of effect of soy isoflavone on thyroid hyperplasia in rats receiving an iodine-deficient diet. Jpn J Cancer Res 2001; 92: 103–108.

60.

de Souza

Dos

Santos

Goncalves

Vaisman

, et al. Impact of flavonoids on thyroid function. Food Chem Toxicol 2011; 49: 2495–2502.

61.

Scholz-Ahrens

Hagemeister

Unshelm

, et al. Response of hormones modulating plasma cholesterol to dietary casein or soy protein in minipigs. J Nutr 1990; 120: 1387–1392.

62.

Balmir

Staack

Jeffrey

, et al. An extract of soy flour influences serum cholesterol and thyroid hormones in rats and hamsters. J Nutr 1996; 126: 3046–3053.

63.

Potter

Pertile

Berber-Jimenez

. Soy protein concentrate and isolated soy protein similarly lower blood serum cholesterol but differently affect thyroid hormones in hamsters. J Nutr 1996; 126: 2007–2011.

64.

Bell

Rutherfurd

Hendriks

. The isoflavone content of commercially-available feline diets in New Zealand. N Z Vet J 2006; 54: 103–108.

65.

Court

Freeman

. Identification and concentration of soy isoflavones in commercial cat foods. Am J Vet Res 2002; 63: 181–185.

66.

White

Freeman

Mahony

, et al. Effect of dietary soy on serum thyroid hormone concentrations in healthy adult cats. Am J Vet Res 2004; 65: 586–591.

67.

Hughes

. Phytoestrogens and health. London: Committee on Toxicity of Chemicals in Food, Consumer Products and the Environment, Food Standards Agency UK, 2003.

68.

Teekachunhatean

Hanprasertpong

Teekachunhatean

. Factors affecting isoflavone content in soybean seeds grown in Thailand. Int J Agron 2013: 11. http://dx.doi.org/10.1155/2013/163573

69.

Howard

Wedekind

, et al. A low-selenium diet increases thyroxine and decreases 3,5,3’triiodothyronine in the plasma of kittens. J Anim Physiol Anim Nutr (Berl) 2002; 86: 36–41.

70.

Kidd

Trowbridge

Goldsby

, et al. Sources of dietary iodine. J Am Diet Assoc 1974; 65: 420–422.

71.

National Research Council of the National Academies. Nutrient requirements of laboratory animals. Washington, DC: National Research Council of the National Academies, 1962.

72.

National Research Council of the National Academies. Nutrient requirements of laboratory animals (revised edition). Washington, DC: National Research Council of the National Academies, 1972.

73.

Roberts

Scott

. Nutrition of the cat. 5. The influence of calcium and iodine supplements to a meat diet on the retention of nitrogen, calcium and phosphorus. Br J Nutr 1961; 15: 73–82.

74.

Scott

Greaves

Scott

. Nutrition of the cat. 4. Calcium and iodine deficiency on a meat diet. Br J Nutr 1961; 15: 35–51.

75.

National Research Council of the National Academies. Nutrient requirements of cats (revised edition). Washington, DC: National Research Council of the National Academies, 1978.

76.

National Research Council of the National Academies. Nutrient requirements of cats (revised edition). Washington, DC: National Research Council of the National Academies, 1986.

77.

Ranz

Tetrick

Opitz

, et al. Estimation of iodine status in cats. J Nutr 2002; 132: 1751S–1753S.

78.

National Research Council of the National Academies. Nutrient Requirements of dogs and cats. Washington, DC: National Research Council of the National Academies, 2006.

79.

Fédération Européenne de l’Industrie des Aliments pour Animaux Familiers (FEDIAF). The reference document on cat and dog nutrition. Brussels: FEDIAF, 2008.

80.

Fédération Européenne de l’Industrie des Aliments pour Animaux Familiers (FEDIAF). The reference document on cat and dog nutrition. Brussels: FEDIAF, 2011.

81.

Fédération Européenne de l’Industrie des Aliments pour Animaux Familiers (FEDIAF). The reference document on cat and dog nutrition. Brussels: FEDIAF, 2012.

82.

Johnson

Ford

Tarttelin

, et al. Iodine content of commercially-prepared cat foods. N Z Vet J 1992; 40: 18–20.

83.

Edinboro

Scott-Moncrieff

Glickman

. Feline hyperthyroidism: potential relationship with iodine supplement requirements of commercial cat foods. J Feline Med Surg 2010; 12: 672–679.

84.

Tarttelin

Johnson

Cooke

, et al. Serum free thyroxine levels respond inversely to changes in levels of dietary iodine in the domestic cat. N Z Vet J 1992; 40: 66–68.

85.

Braverman

. Iodine and the thyroid: 33 years of study. Thyroid 1994; 4: 351–356.

86.

Kyle

Tarttelin

Cooke

, et al. Serum free thyroxine levels in cats maintained on diets relatively high or low in iodine. N Z Vet J 1994; 42: 101–103.

87.

Woeber

. Iodine and thyroid disease. Med Clin North Am 1991; 75: 169–178.

88.

Stanbury

Ermans

Bourdoux

, et al. Iodine-induced hyperthyroidism: occurrence and epidemiology. Thyroid 1998; 8: 83–100.

89.

Edinboro

Scott-Moncrieff

Glickman

. Review of iodine recommendations for commercial cat foods and potential impacts of proposed changes. Thyroid 2004; 14: 722.

90.

Wedekind

Blumer

Huntington

, et al. The feline iodine requirement is lower than the 2006 NRC recommended allowance. J Anim Physiol Anim Nutr (Berl) 2010; 94: 527–539.

91.

Daniel

Sharp

Nieckarz

, et al. Quantitative thyroid scintigraphy as a predictor of serum thyroxin concentration in normal and hyperthyroid cats. Vet Radiol Ultrasound 2002; 43: 374–382.

92.

Fischetti

Drost

DiBartola

, et al. Effects of methimazole on thyroid gland uptake of 99mTC-pertechnetate in 19 hyperthyroid cats. Vet Radiol Ultrasound 2005; 46: 267–272.