Unraveling the Environmental Links to Feline Hyperthyroidism: Insights from Silicone Passive Samplers

Abstract

Feline hyperthyroidism (FH) is the most common endocrine disorder affecting cats and poses significant health challenges to domestic cats and veterinary professionals. This disease is caused by the effects of excess thyroid hormone production and causes a variety of symptoms including weight loss, increased urination, and increased appetite. Despite its prevalence, the underlying cause of this condition remains unclear. While many factors have been extensively studied, there isn’t conclusive evidence linking hyperthyroidism to diet, litter, and indoor lifestyle. Recent research has suggested an association between FH and exposure to flame retardants in consumer products. Many consumer products also contain other endocrine-disrupting chemicals (EDCs) and potential endocrine-disrupting chemicals (pEDCs) in addition to flame retardants that could be linked to FH. To investigate this further, silicone passive sampling devices (PSDs) in the form of pet tags were used to measure the environmental chemical exposure of 78 cats, aged seven years and older, in Oregon and New York using a chemical screening method containing hundreds of EDCs/pEDCs. The objective of this study was to compare exposure frequencies and concentrations between hyperthyroid and non-hyperthyroid cats. While no statistically significant associations were identified, this study found higher concentrations of butyl benzyl phthalate (BBP), galaxolide, lilial, and tonalide in the tags worn by cats with FH compared to euthyroid cats. TCPP, b-ionone, lilial, cinnamal, benzyl salicylate, and tonalide have not been previously mentioned in past feline exposure studies. These chemicals are found in various personal care and consumer products such as vinyl tiles, fragrances, furniture, and cosmetics. Their presence in PSDs worn by cats that develop hyperthyroidism may indicate a potential role of these environmental chemicals in FH etiology.

Graphical Abstract

1. Introduction

Feline hyperthyroidism (FH) is an endocrine disorder that occurs due to overproduction of thyroid hormones and is a significant health concern in the feline population (Peterson 2012). This disease poses a substantial challenge to domestic cats, their owners, and veterinary professionals (Peterson 2012). Diagnosis of FH has increased substantially since it was first identified in the 1970s (Peterson 2012) and has become the most prevalent endocrine disorder observed in cats ages seven years and older (Peterson 2014, Poutasse, Herbstman et al. 2019). The diagnostic rate of FH in North America has increased from 0.3% in 1979 to 4.5% in 1985 (Scarlett, Moise et al. 1988) and further increased to 10% in 2014 (Poutasse, Herbstman et al. 2019). While this may be due to increased owner awareness and screening for this disease, it is also suspected the disease has increased in prevalence.

Cats in domestic settings are the only non-human species commonly diagnosed with hyperthyroidism, a condition analogous to toxic nodular goiter (TNG) in humans (Peterson 2014). Both diseases arise from excess amounts of thyroxine (T4) and triiodothyronine (T3) hormones (Scott-Moncrieff 2012). A diagnosis of FH is typically made by evaluating total T4 and occasionally free T4 (Peterson, Graves et al. 1990, Peterson 2006). Decreased levels of thyroid-stimulating hormone (TSH), when measured in combination with fT4 and tT4, can also be used as a diagnostic test (Peterson, Guterl et al. 2015). The overproduction of T3 and T4 hormones often results from benign adenomatous hyperplasia of one, or more commonly, both thyroid lobes (Scott-Moncrieff 2012, Kasabalis, Soubasis et al. 2013). In felines, the most common symptoms are hyperactivity, weight loss, increased appetite, vomiting, diarrhea, increased water consumption, increased urination and difficult breathing (Kasabalis, Soubasis et al. 2013, Vaske, Schermerhorn et al. 2014). Cats can also develop systemic hypertension, heart disease and seizures (Valtonen and Eriksson 1970). Treatment options include administration of anti-thyroid hormone drugs, dietary iodine restriction, radioactive iodine treatment, and surgical thyroidectomy (Mallery, Pollard et al. 2003, Carney, Ward et al. 2016).

Researchers have been evaluating genetic, environmental, and husbandry risk factors since the 1980s (Peterson 2012, Scott-Moncrieff 2012). More recent investigations focused on household exposure to flame retardants (FRs) as risk factors for FH (Dye, Venier et al. 2007, Poutasse, Herbstman et al. 2019). Flame retardants (FRs) are a class of chemicals commonly added to household products. Many FRs are known to act as endocrine-disrupting chemicals (EDCs), meaning they can interfere with hormone systems (Norrgran Engdahl, Bignert et al. 2017).

EDCs are recognized for their ability to interfere with the endocrine systems of animals and humans (Clotfelter, Bell et al. 2004), while potential endocrine-disrupting chemicals (pEDCs) may affect the endocrine system, but have not yet been conclusively proven to do so (Kledal, Jørgensen et al. 2000). Some chemicals in household items, including building materials and pesticides, are recognized as EDCs/pEDCs (Hwang, Park et al. 2008, Rudel and Perovich 2009). Other household objects, such as appliances, electronics, furniture, carpets, personal care products (PCPs), and mattresses are known sources of FRs and other EDCs/pEDCs (Hwang, Park et al. 2008). In addition, litter box deodorizers, cat and dog repellents, and fragrances added directly to cat litter and litter box plastics have been identified as potential sources of EDC/pEDC exposure (Lee and Ji 2022, Ashcroft, Dosoky et al. 2024).

To measure the bioavailable FR exposure in house cats, past studies have used silicone passive sampling devices (PSDs) worn as pet tags. Bioavailable chemicals are those that can be absorbed by the cat or study participant via inhalation, dermal contact, or ingestion. The cat tags measure these bioavailable chemicals through passive sampling from atmospheric exposures as well as direct contact with the home environment and the cat’s fur, mimicking primarily inhalation exposure but also some dermal contact, providing insights into the environmental exposure impacting the cat (O’Connell, Kincl et al. 2014). Silicone PSDs capture lipophilic, volatile, and semi-volatile organic chemicals (SVOCs) through diffusion, as the PSD polymer imitates the phospholipid membrane of an organism (Anderson and Hillwalker 2008). Over the past decade, there have been numerous studies using PSDs to capture personal chemical exposure (Hwang, Park et al. 2008, Doherty, McRitchie et al. 2022, Bonner, Horn et al. 2023, Samon, Rohlman et al. 2023), including the use of silicone wristbands to capture various chemical classes such as PCPs, phthalates, FRs, polycyclic aromatic hydrocarbons (PAHs), and pesticides.

Most past studies using PSDs to capture personal chemical exposure have focused on human chemical exposure, leaving a gap in research on pets in the home as either sentinels for exposure (Wise, Hammel et al. 2020), or direct animal health (Poutasse, Herbstman et al. 2019). This gap for animal exposure includes individual feline chemical exposure, especially pertaining to any association between environmental chemical exposures and FH. Furthermore, within the field of study of personal chemical exposure, domestic felines are uniquely suited for isolating chemical exposures from the indoor home environment and serving as sentinels for indoor exposure only, compared to dogs that also capture outdoor chemical exposures. Silicone PSDs worn as pet tags (Figure 1) work well for capturing this feline exposure, as they are cost-effective, user-friendly, non-invasive, and lightweight and their use can help in gaining a more comprehensive understanding of the factors contributing to FH.

Figure 1.

Silicone pet tag on cat collar.

In this study, EDC/pEDC exposures of cats diagnosed with FH were compared to those of cats without feline hyperthyroidism (NFH) by analyzing silicone pet tags worn by domestic, indoor cats aged seven years and older. A broad semi-quantitative screen of 1530 target analytes covering a wide range of chemical classes, including many EDCs/pEDCs, was conducted to investigate previously unrecognized associations between FH and chemical exposure in the home. The objectives of this study were to (1) compare how frequently EDCs/pEDCs were detected between hyperthyroid and non-hyperthyroid cats by silicone cat tags, (2) compare the EDC/pEDC cat tag concentrations between hyperthyroid and non-hyperthyroid cats, and (3) compare the EDC/pEDC cat tag concentrations with thyroid hormone concentrations in cats without medically altered hormone levels (i.e. healthy cats).

2. Materials and Methods

2.1. Materials

All solvents used for sample processing were Optima grade and purchased from Fisher Scientific (Pittsburgh, PA, USA). Analytical standards were obtained from AccuStandard (New Haven, CT, USA). All standards were prepared in ethyl acetate and stored at 4°C. All water used was ultra-pure 18 MΩ/cm. Airtight polytetrafluoroethylene (PTFE) storage bags and closures were sourced from Welch Fluorocarbon, Inc. (Dover, NH, USA).

2.2. Cat Population and Recruitment

Approval was received from Oregon State University and Columbia University Institutional Animal Care and Use Committees (IACUCs) (OSU ACUP 4963) and (CU ACUP AC-AAAT5454) for recruiting 78 cats (39 FH, 39 NFH) from various clinics in New York and Oregon between December 2017 and October 2018 (Poutasse, Herbstman et al. 2019) (refer to SI Section 1 for more on feline recruitment). All cats underwent a series of tests including a serum thyroid panel as well as a physical exam conducted by a recruiting veterinarian to assess the hyperthyroid status of the cat (more information on the thyroid panel and diagnoses can be found in SI Section 2). General statistics on each population of cats, including hormone levels of healthy cats, breeds, and comorbidities, can be found in Table 1 and Tables S1 and S2 in SI Section 1.

Table 1.

Population and litter use questionnaire answers for 75 FH and NFH cats across both locations.

Characteristic	Oregon (n=40)	New York (n=35)
Thyroid Illness
Hyperthyroid (n=38)	17	21
Non-Hyperthyroid (n=37)	23	14
Sex
Female	21	16
Male	17	21
Age Range (years)	7~17	7~18
Weight Average ± SD (kg)	5.1 ± 4.5	4.5 ± 3.7
Average Thyroid Hormone Level ± SD*	NFH (n=37)	NFH (n=37)
Thyroid Stimulating Hormone (ug/dL)	0.05 ± 0.05	0.1 ± 0.1
Total T3 (ug/dL)	37 ± 4.5	35 ± 9.53
Total T4 (ug/dL)	2.3 ± 0.44	2.3 ± 0.55
Free T4 (ug/dL)	1.2 ± 0.36	1.3 ± 0.50

^*These are only among the NFH cats (n=37).

2.3. Silicone Tag Preparation and Extraction

Silicone pet tags were purchased from 24hourwristbands.com (Houston, TX, USA) and prepared as previously described (Anderson, Points et al. 2017, Poutasse, Herbstman et al. 2019). Briefly, the tags were rinsed with ultra-pure water and conditioned in a vacuum oven at 270°C at <1 torr for at least 180 minutes. After confirming the conditioning process met data quality objectives (DQOs), each tag was mailed/given to households with felines matching our study objectives (Poutasse, Herbstman et al. 2019). Households were asked to attach the tags to pet collars (Figure 1) for seven days. After deployment, tags were gathered by study personnel or shipped to the laboratory to be stored at −20°C until analysis.

Prior to extraction, pet tags were first rinsed twice with 18MΩ water and then once with isopropanol to remove any surface particulate matter. They were then stored at −20°C until extraction. As reported previously (O’Connell, Kincl et al. 2014), compounds were extracted from silicone with two rounds of 100 mL each of ethyl acetate that were combined and reduced to 1 mL under nitrogen gas (Turbo-Vap L, Biotage, Charlotte, NC, USA; RapidVap, LabConco, Kansas City, MO, USA).

2.4. Chemical Analysis

Analysis was performed using an Agilent 7890A gas chromatograph (GC) coupled with an Agilent 5975C mass spectrometer (MS) running in full-scan mode with electron ionization (70 eV) (Wilmington, DE). An Agilent DB-5MS column (30 m × 0.25 mm) was used and inlet pressure was retention-time locked to chlorpyrifos (19.23 ± 0.20 minutes) (Bergmann, Points et al. 2018), and additional instrument parameters are located in SI Section 3. The analysis used Automated Mass Spectral Deconvolution and Identification System (AMDIS) software version 2.66 (Agilent, Wilmington, DE) to deconvolute and integrate peaks for quantification (see SI Section 3, Table S4 for more details). AMDIS software was paired with the 2008 National Institute of Standards and Technology (NIST) library and Chemstation software (Agilent, Wilmington, DE). A list of target chemicals is available at https://fses.oregonstate.edu/methods/1530. Each detection by the software was confirmed by an analytical chemist to meet data quality objectives and ensure positive identification as described previously (Bergmann, Points et al. 2018).

2.5. Quality Control

For the analysis, chemicals with a 60% or greater match to library reference spectra were manually reviewed according to DQOs (Bergmann, Points et al. 2018). Compounds that did not meet all the criteria were rejected as a false positive: retention time shift (from Chemstation peak to AMDIS peak) was greater than 45 seconds, sample peak was below 3:1 signal-to-noise ratio, sample peak shape didn’t match AMDIS peak shape, or the sample spectra did not match the NIST reference library spectra (i.e. missing the parent ion).

Instrument blanks and continuing calibration verifications (CCVs) were analyzed at the beginning and end of every analytical batch (n=12). A total of 20 CCVs were analyzed, and all met the DQOs: greater than 70% detection of target compounds, retention time within 45 seconds, and more than 60% of target chemicals were detected within 2.5 times of their true standard concentration. A total of 35 instrument blanks were analyzed and no target compounds were detected by response. To demonstrate instrument precision, a duplicate analysis was performed resulting in a median relative percent difference (RPD%) of 18% across all detected analytes. A matrix spike analysis was performed on a sample to assess accuracy. Across all detected analytes, there was a median percent recovery of 100%.

Three QC blanks were analyzed with the samples. One was a trip blank that traveled round trip in the same shipping package as a sample. The second was a lab processing blank that went through all lab procedures. The third was a reagent blank that consisted only of the ethyl acetate solvent used during extraction and underwent extraction with the samples. Out of the three QC blanks analyzed, the compound bis(2-ethylhexyl) phthalate (DEHP) was detected in the reagent blank. The concentration of DEHP (203 nmol/tag) found in the reagent blank was background subtracted from all DEHP concentrations that were detected in the samples.

2.6. Participant Population

All tags were worn for seven days and of the 78 total pet tags distributed and returned to the lab, 75 were included in this analysis. Three tags were excluded because the samples had become compromised during storage following the original analysis (Poutasse, Herbstman et al. 2019). Other factors affecting the analysis of data included questionnaire compliance and confounding variables. A 98% (74/75) partial compliance rate was achieved for the questionnaire with individual question compliance ranging from 3% (2/75) to 98% (74/75). The confounding variables considered included the location of the cat, age, damage to the tag (i.e., bite marks and chunks chewed off), time spent outdoors, cohabitation with other recruited cats, and the time of year during which sampling took place. Ten of the pet tags were returned with bite marks, two of which had sections missing, presumably chewed off by the recruited cat. These tags were weighed upon return, and the new mass was substituted in for all calculations.

2.7. Statistical Analysis

Statistical analyses were only conducted for analytes detected in at least 15% of the pet tags to ensure that spurious associations were not made due to non-detects in left-censored data. Concentrations below the method limit of detection (LOD) were substituted with a value equal to half the method limit of quantification (LOQ) of the respective chemicals. Thyroid hormone concentrations in healthy cats were log-normally distributed. FH cats did not have their thyroid hormones measured because treatment for hyperthyroidism artificially changed the hormone concentrations in the FH cats in the study. All analyte concentrations were converted to units of nmol/tag and were log10 transformed for statistical tests since none were found to be normally distributed.

The data for Oregon and New York were combined and a Fisher’s test (F-test) for equal variance was conducted on the data for each chemical. This was followed by a one-sided Student’s t-test with the appropriate equal or unequal variance determined by the F-test. The one-sided Student’s t-test was used to determine if the concentration of a chemical in the FH pet tags was greater than the concentration of a chemical in the NFH pet tags at p < 0.10. A Fisher’s exact odds ratio was used to assess the significance of differences in the detection of chemicals in FH tags and NFH tags. The concentrations of total T4, free T4, total T3, and TSH were then graphed against the log10 concentration of chemicals in NFH cats, and a linear regression was applied to the graph at an R² > 0.2 significance level. Four graphs were created for each chemical to assess the effect of the chemical on each of the four hormones independently in healthy cats.

3. Results & Discussion

3.1. Chemical Detections

Eighteen chemicals, including EDCs, pEDCs, and non-endocrine active chemicals, were detected in at least 15% (n=12) of all pet tags across both states (see SI Section 6, Table S8 for detection frequencies of each), with the total number of chemicals detected in an individual tag varying from 1–16. The EDCs detected were butyl benzyl phthalate (BBP), butylated hydroxyanisole (BHA), butylated hydroxytoluene (BHT), bis(2-ethylhexyl) phthalate (DEHP), diethyl phthalate (DEP), diisobutyl phthalate (DIBP), di-n-butyl phthalate (DNBP), and galaxolide. The pEDCs detected were lilial and tonalide. The remaining chemicals were benzophenone, benzyl benzoate, benzyl salicylate, b-ionone, cinnamal, coumarin, N,N-diethyl-m-toluamide (DEET), and tris(chloropropyl) phosphate (TCPP). Fifteen of these detected chemicals are classified as fragrances by the EPA (SI Section 5, Table S7), and only three chemicals (BBP, DEET, and TCPP) are not considered fragrances. Total detections did not appear to be different between the health phenotypes (FH: 10 ± 3 vs NFH: 11 ± 2).

Figure 2 compares detection frequencies of chemicals found in FH cats in Oregon to those in New York. Twelve compounds (67%) were detected more frequently in tags worn by FH cats in Oregon than by FH cats in New York (Figure 2). One compound, BHA, was detected equally frequently in tags worn by FH cats in both Oregon and New York (Figure 2). Figure 3 compares detection frequencies of chemicals found in FH cats to those of chemicals found in NFH cats separately by state. Across both states, DEET was detected more frequently in tags worn by FH cats, but by comparison, BHA, cinnamal, DEP, galaxolide, and tonalide were detected more frequently in tags worn by NFH cats (Figures 3a and 3b). In New York, DEHP was detected in 95% of tags worn by FH cats; DEHP, DIBP, DNBP, and galaxolide were detected in 100% of tags worn by NFH cats; and benzophenone was detected in 93% of tags worn by NFH cats (Figure 3). In Oregon, galaxolide was detected in 94% of tags worn by FH cats and 96% of tags worn by NFH cats; benzophenone, DEHP, DIBP, and DNBP were detected in 100% of tags worn by NFH cats; and BBP was detected in 91% of tags worn by NFH cats (Figure 3).

Figure 2.

Compound detection frequency (%) for FH cats (n=38) in Oregon (n=17) and New York (n=21).

Figure 3.

Compound detection frequency (%) for (a) FH (n=21) and NFH (n=14) cats in New York (n=35) and (b) FH (n=17) and NFH (n=23) cats in Oregon (n=40).

3.2. Odds Ratios

To address the objective of comparing chemicals between feline disease phenotypes, a contingency table was created, and a Fisher’s exact odds ratio was performed for each chemical and the associated number of detections in both states (Figure 4). In instances where there were zero detections in a cell of the contingency table, the Haldane-Anscombe correction was applied (Anscombe 1956, Haldane 1956) and 0.5 was added to all values in the contingency table. All log odds ratios showed no significant difference (p < 0.10), as indicated by their confidence intervals crossing the value of zero, meaning there was no association between the detection of any of the chemicals in the tags and the cat disease phenotype.

Figure 4.

Fisher’s exact odds ratios of detection frequencies for all 75 cats with 90% confidence intervals on a log scale.

3.3. Chemical Concentrations

While there were no statistically significant chemical concentration results between FH and NFH cats (one-sided Student’s t-test (p > 0.10), SI Section 7, Table S9), some trends were observed where BBP, galaxolide, lilial, and tonalide, were detected at higher concentrations (≥138%) in tags worn by FH cats in both states. The average concentrations of BBP, galaxolide, lilial, and tonalide separated by state and hyperthyroid status are shown in Figure 5 (see SI Section 7, Figure S1 for the other compounds). By contrast, DEP, DIBP, benzophenone, benzyl benzoate, cinnamal, and coumarin were detected at higher concentrations (≥109%) in tags worn by NFH cats in both states. Percentages were calculated using a ratio of the average concentration of a chemical in FH cat tags to the average concentration of a chemical in NFH cat tags in each state. Total concentration of chemicals detected per cat tag varied from 5.49 nmol/tag to 533 nmol/tag. The average total concentration did not appear to be different between the cat health phenotypes (FH: 206 ± 107 nmol/tag vs NFH: 218 ± 113 nmol/tag). Importantly, the presence of multiple chemicals within individual pet tags can lead to synergistic toxic effects, where the combined impact on health is greater than the sum of the individual effects (Cedergreen 2014). This highlights the need to consider the interactions between chemicals within a single tag when assessing potential risks.

Figure 5.

Average concentrations of chemicals in nmol/tag from both New York and Oregon comparing FH to NFH for BBP, galaxolide, lilial, and tonalide with standard deviation.

3.4. Individual Exposures

Individual chemical exposures for each cat are depicted in the SI (Section 8), with five representative distribution charts shown in Figure 6. Each cat had a unique distribution of chemical exposures that was difficult to accurately represent through summary statistics, but cats in Figure 6 were chosen based on a visual comparison between the pie charts (refer to SI Section 8). Average concentrations showed that DEHP was present in the highest concentration of any chemical for both health groups and states, which is reflected in the individual exposure pie charts. Conversely, BBP was among the top six contributors to total chemical exposure based on the average concentrations of each compound across all tags. However, the pie charts show an example of a single individual with a BBP concentration comprising the highest proportion of the individual’s chemical exposure (Chart 38 in Figure 6); whereas it is barely perceptible in other individuals’ pie charts. This broad variation in concentrations and proportions of overall exposure illustrates the uniqueness of each cat’s exposure profile.

Figure 6.

Chemical exposure distributions of five individual cats with the total concentration of all chemicals (nmol/tag) in the center. Charts 6, 24, and 38 depict FH cats while charts 42 and 64 depict NFH cats.

3.5. Hormones

An analysis of hormone concentrations and chemical concentrations in NFH cats was performed to determine if there was any correlation between increasing chemical exposure and changing thyroid hormone levels as a possible indicator of NFH cats developing FH in the future due to chemical exposure. Hormone levels were exclusively assessed in NFH cats as the hyperthyroidism treatments altered the hormone concentration in the FH cats so they would not correlate to chemical exposure. Only five graphs comparing the levels of the individual hormone in NFH cats and the concentrations of a specific chemical above the LOD had an R² value of greater than 0.2, showing a possible association between chemical exposure and hormone concentration (SI Section 9, Figure S3). Four of the five graphs had a linear regression that contradicted the expectation that FH is associated with an increase in T4 and T3 and a decrease in TSH (Peterson, Guterl et al. 2015). Only BHT showed a positive correlation between increasing chemical concentration and increasing total T3 concentration in NFH cats, indicating an increased risk of FH, as evidenced by a positive linear regression (R² = 0.36) between the log concentration of BHT and the total T3 concentration. A literature search did not show any other causes of consistently increased total T3 in cats.

3.6. Discussion

This is the first study to conduct a broad screen for EDCs/pEDCs detections, among household cats with and without hyperthyroidism. Furthermore, this is the first study to identify TCPP, b-ionone, lilial, cinnamal, benzyl salicylate, and tonalide as feline exposures. The EDCs/pEDCs detected were galaxolide, tonalide, DEHP, BBP, DNBP, DEP, DIBP, BHA, BHT, and lilial (SI Section 5, Table S7). Chemicals detected in this study that are not classified as EDCs/pEDCs were benzophenone, benzyl salicylate, benzyl benzoate, DEET, coumarin, TCPP, b-ionone, and cinnamal (SI Section 5, Table S7). Benzophenone, galaxolide, benzyl salicylate, and tonalide are frequently used in fragrances, PCPs, and household objects, such as candles and room sprays (Johnson, Favela et al. 2022). Benzophenone and benzyl salicylate are common ingredients used in sunscreens and cosmetics and can persist on clothes and potentially be absorbed through the dermis of humans (Morrison, Bekö et al. 2017).

Several frequently detected chemicals in this study (e.g., coumarin, galaxolide, and lilial) are found in fragrances used in cat litters as deodorizers and fragrance boosters (SI Section 5, Table S7), indicating that cat litter with these fragrances may be a potential chemical exposure pathway. Cat litter and cat products frequently cite “fragrance” as an ingredient instead of the individual chemical components, preventing an assessment to determine if litter is the source of exposure for some of the detected chemicals. This generalized labeling also applies to human products including beauty and personal care products (Steinemann 2009). These proprietary “fragrances” include a complex mixture of chemicals including EDCs/pEDCs (Nicolopoulou-Stamati, Hens et al. 2015). Without knowing what chemicals are in these mixtures, an association cannot be determined between cat litter use, chemical exposures, and FH occurrence. This also has potential health implications for humans exposed to these unknown chemical mixtures.

Galaxolide is a common ingredient in household products including air fresheners, cleaning supplies, and cat litter deodorizing sprays. Previous studies have found no significant link between FH and the use of cat litter (Kass et al. 1999, Peterson and Ward 2007, Peterson 2012). However, it is important to consider that the use of cat litter is a strong indicator that a cat lives indoors (Kass et al. 1999). Indoor cats often receive better care, including regular veterinary visits, consistent access to food and water, and a safer environment, which can contribute to a longer lifespan. This extended lifespan allows them to reach an age where they are more likely to develop conditions such as FH, which predominantly affects senior cats. Therefore, while the use of cat litter itself may not be linked to FH, it serves as an indicator of the indoor lifestyle and the associated better care that can lead to a longer life and a higher likelihood of developing age-related conditions like FH.

The presence of lilial, which is a known fragrance in cat litter, in cat tags is important because it indicates that humans themselves, through the purchase of cat litter with lilial fragrances, are a possible source of chemical exposure to their pets. The EU and Britain banned lilial in 2009 due to its effects on fertility and fetal development (Charles and Darbre 2009, UNION 2009). As of early 2025, lilial is unregulated in the US and is commonly used in personal care products such as perfumes, lotions, and hair products (Darbre and Harvey 2022).

Tonalide exposures were higher for FH cats than NFH cats in both Oregon and New York. In the US, tonalide is not considered an EDC/pEDC but it is under evaluation for endocrine disruption under an EU legislation (SI Section 5, Table S7). Tonalide is commonly found in personal care products such as perfumes, soaps, and detergents (Jyoti and Sinha 2023). Previous studies have found that tonalide may pose risks to invertebrates and fish, potentially reducing their ability to reproduce (Cahova, Blahova et al. 2023, Jyoti and Sinha 2023). Tonalide is ubiquitous in personal care products, and its widespread use contributes to environmental contamination (Valdersnes, Kallenborn et al. 2006). Given the differences in geographical locations and environmental conditions, further investigation is necessary to understand the potential link between tonalide exposure and FH in cats.

Similarly, BBP was one of the most frequently detected chemicals in the cat tags from both Oregon and New York, indicating that exposure occurs across diverse regions. BBP is classified as a phthalate and is considered an EDC (SI Section 5, Table S7). BBP is one of the most common EDCs found in plasticizers and in many vinyl products (Christova-Bagdassarian, Tishkova et al. 2017). It is also found in some personal care products such as nail polish and hairsprays (Houlihan, Brody et al. 2002, Young, Allen et al. 2018).

In this study, BHT was most frequently detected among FH cats from Oregon (Figure 3b). BHT is an antioxidant added to both human and pet foods, food packaging, cosmetics, pharmaceuticals, gasolines, oils, paints, and inks (Pop, Berce et al. 2013, Schmidtkunz, Küpper et al. 2020). Very little is known about the effects of BHT on human health (Wang and Kannan 2019) and as of 2025, no studies have been reported on the effects of BHT on feline endocrine health. BHT has been shown to exhibit endocrine disrupting activity in rats (Pop, Berce et al. 2013), specifically, BHT exposure in rats has been shown to accelerate the iodine cycle, thus increasing thyroid hormone production (Briggs, Lok et al. 1989). Notably, BHT metabolism varies between different animal species (Schmidtkunz, Küpper et al. 2020), so more studies are necessary to adequately understand its effects on feline hyperthyroidism.

Inhalation was the primary route of exposure influencing the chemical concentrations in the cat tags in this study. As such, the use of consumer products and household products that release these chemicals in the vapor-phase is the main factor influencing exposure. Many EDCs are measured at higher concentrations in indoor air than in outdoor air (Pocar, Grieco et al. 2023), due to the release of chemicals from multiple sources in an enclosed space such as heating, cooking, smoking, and use of chemical-based consumer products. A common element of all chemicals detected in this study is their use in personal care products, household products, and home building materials indicating that the choices made by pet owners regarding what products are used in the house are the primary factor influencing EDC/pEDC exposure to their pets.

Some limitations of this study to address in future research include the increase of study size and population diversity, as well as consideration of breed and comorbidities as confounding factors. Some breeds have been shown to be less prone to FH than others (Scarlett, Moise et al. 1988, Crossley, Debnath et al. 2017) and research into comorbidities, such as chronic kidney disease is ongoing (Geddes and Aguiar 2022). There is also the need to expand the chemicals sampled and analyze with higher sensitivities in future studies. It is possible that there is an association between feline hyperthyroidism and a compound that the silicone PSDs are unable to uptake and retain or from a chemical not able to be detected by this chemical screen method. For example, both vapor phase and particulate-bound chemicals are relevant for exposure, but silicone PSDs are only capable of capturing chemicals in the vapor phase (O’Connell, Kincl et al. 2014) so another study utilizing different PSD technology may be warranted. While biological samples offer more direct exposure assessments which encompass all routes of exposure, PSD technology is less invasive and is better suited for the large-scale feline exposure studies required for statistical significance (Poutasse, Herbstman et al. 2019). Additionally, the lack of transparency from manufacturers regarding product ingredients made it difficult to identify which products used in homes were the sources of chemical detections. Another important limitation of this study was the lack of follow-up with the NFH cats. While the current study identified a possible link between BHT exposure and elevated levels of total T3 in NFH cats, a future study should follow the cats over a longer time frame to collect information about their health status related to long-term exposure. Since the data for this study was collected in 2017 and 2018, significant global changes have occurred due to COVID-19 (Nicola, Alsafi et al. 2020, Mitra, Chu et al. 2022). Given the increased time people spend working from home since COVID-19, more investigation of the sources and effects of EDCs commonly found in household and consumer products is urgently needed.

4. Conclusion

The elevated levels of EDCs/pEDCs in hyperthyroid cats from both Oregon and New York suggest a potential association with the disease’s occurrence. However, there was insufficient statistically significant data to establish a definitive link between higher concentrations of BBP, galaxolide, lilial, and tonalide in silicone pet tags and hyperthyroidism in cats. These findings emphasize the need for further investigation into the role of environmental chemical exposure in feline hyperthyroidism. Future studies should explore additional environmental and dietary factors and consider human exposure pathways to better understand potential shared risks and the mechanisms contributing to feline hyperthyroidism.

Figure 7.

Linear regression between log BHT concentration (log nmol/tag) and total T3 concentration (ng/dL) with R² of 0.36 in NFH cats.

Highlights:

• Eighteen chemicals were consistently detected in all cat samplers.

• BBP, galaxolide, lilial, and tonalide had higher concentrations (≥138%) in hyperthyroid cats.

• BHT concentration was positively correlated (R² = 0.36) with T3 concentration.

Abbreviations

AMDIS

Automated Mass Spectral Deconvolution and Identification System

DEHP

bis(2-ethylhexyl) phthalate

BBP

butyl benzyl phthalate

BHA

butylated hydroxyanisole

BHT

butylated hydroxytoluene

CCVs

continuing calibration verifications

DQOs

daily quality objectives

DEP

diethyl phthalate

DIBP

diisobutyl phthalate

DNBP

di-n-butyl phthalate

ED

endocrine disruptor

EDCs

endocrine-disrupting chemicals

EPA

Environmental Protection Agency

FH

Feline Hyperthyroidism

FR

flame retardants

GC

gas chromatograph

IARC

Internation Agency for Research on Cancer

LOD

limit of detection

LOQ

limit of quantification

MS

mass spectrometer

DEET

N,N-diethyl-m-toluamide

NIST

national institute of standards and technology

NFH

no feline hyperthyroidism

PCPs

personal care products

PSDs

passive sampling devices

PAHs

polycyclic aromatic hydrocarbons

PTFE

polytetrafluoroethylene

pEDCs

potential endocrine-disrupting chemicals

RPD%

relative percent difference

SVOCs

semi-volatile organic compounds

TSH

thyroid stimulating hormone

T4

thyroxine

TNG

toxic nodular goiter

T3

triiodothyronine

TCPP

tris(chloropropyl) phosphate

WHO

World Health Organization