• Open Access

Propylthiouracil for Treatment of Hyperthyroidism in a Horse


Corresponding author: R.H.H. Tan, VMRCVM, Department of Large Animal Clinical Sciences, Virginia-Maryland Regional College of Veterinary Medicine, Duckpond Drive, Phase II, Blacksburg, VA 24061; e-mail: tanr@vt.edu.

A 19-year-old 250-kg Quarter Horse mare was examined in the fall for evaluation of weight loss of 4 years' duration. The mare had been purchased 6 years before examination and was reported to be normal in appearance, weight, and demeanor. She had been used for trail riding. The owner initially noticed weight loss following transient postpartum diarrhea 4 years before presentation. The weight loss was mild and slowly progressed with intermittent periods of soft feces. Referral for evaluation was prompted by marked weight loss over the preceding 6 months. The mare had a ravenous appetite and polydipsia. In addition, the mare did not completely shed her hair coat during summer and was heat intolerant in ambient temperatures that exceeded 32.2 °C (90°F). This was manifested as lethargy, sweating, and increased respiratory rate.

Because of incomplete hair coat shedding and suspicion of pars intermedia hyperplasia, a dexamethasone suppression test had been performed in the fall, 5 years before presentation. This revealed incomplete suppression of serum cortisol to 59% of baseline at 15 hours (144.3 nmol/L baseline (6:00 pm), 85.5 nmol/L at 15 hours [9:00 am]). CBC, biochemistry, and fecal examination performed 3 years before presentation were within reference intervals. A test for equine infectious anemia was negative and routine vaccinations had been administered 18 months before examination. Three days before examination, increased concentrations of plasma T4 (74.9 nmol/L; reference interval: 19.3–57.9 nmol/L) and T3 (3.7 nmol/L; reference interval: 0.45–1.2 nmol/L) were measured in samples collected at 9:00 am.

The only current medication was pyrantel tartratea (1.6 mg/kg PO q 24 h). The mare's diet for the preceding 6 weeks was free-choice pasture access, 2.7 kg commercial rationb (10% fat, 14% protein, 17% fiber) per day divided into 2 feeds, 230 g rice bran pellets once daily, 125 mL corn oil once daily, and alfalfa hay.

The mare was bright, alert, and responsive at initial physical examination, with a temperature of 37.9 °C (100.3°F), heart rate of 60 beats/min and a respiratory rate of 36 breaths/min. Mucous membrane color was pink and capillary refill time was <2 seconds. Body condition score was 1 out of 9 (Henneke body condition scoring system) and bodyweight was 250 kg (ideal estimated weight for breed, height, and age was 420 kg). Bilateral soft tissue swelling in the ventral aspect of the proximal cervical region was consistent with enlarged thyroid lobes. Polyphagia (average 5.5 kg commercial rationb and 11 kg hay/day) and polydipsia (average water intake 50 L/day) were evident during hospitalization. The mare was compliant to handling and procedures, but appeared to be more hyperactive when compared with horses of a similar age.

CBC, serum chemistry, sorbitol dehydrogenase and serum bile acid concentration were within reference intervals. Variables that were abnormal were white cell count (14.7 × 103/L; reference interval: 5.4–14.3 × 103/L), neutrophil count (11.1 × 103/L; reference interval: 2.4–6.5 × 103/L), γ-glutamyl transferase (27 U/L; reference interval: 8–24 U/L), total bilirubin (2.1 mg/dL; reference interval: 0.7–1.7 mg/dL), direct bilirubin (0.6 mg/dL; reference interval: 0–0.4 mg/dL), blood urea nitrogen (11 mg/dL; reference interval: 17–28 mg/dL), albumin (3.0 mg/dL; reference interval: 3.2–4.0 mg/dL), triglycerides (4.0 mg/dL; reference interval: 11–57 mg/dL), and potassium (2.5 mEq/L; reference interval: 3.0–4.5 mEq/L). Serum IgM was within normal limits (54.0 mg/dL, reference interval >25 mg/dL). Urinalysis and rectal examination were within normal limits. Feces were negative for parasites, Clostridium difficile A and B and Clostridium perfringens alphatoxins. Fecal culture at admission was negative for Salmonella. Serum concentrations of cortisol (281 nmol/L; reference interval: 55–165 nmol/L) and adrenocorticotrophic hormone (ACTH, 47.7 pmol/L; reference interval: 2.0–7.7 pmol/L) were increased, and insulin (1 hour post-prandial) was decreased (4.3 μU/mL; reference interval: 10–40 μU/mL). The blood glucose concentration was within normal reference intervals.

A d-xylose absorption test was performed after fasting the mare for 12 hours.1 The plasma concentrations of xylose and absorption pattern were within normal limits (peak xylose concentration of 20 mg/dL at 60 minutes). A rectal mucosal biopsy specimen indicated moderate infiltration of the lamina propria and muscularis with numerous eosinophils, lymphocytes, plasma cells, and occasional neutrophils. This finding was of questionable clinical significance considering the absence of diarrhea and a normal d-xylose absorption test.

Thoracic radiography, ultrasound evaluation of the thoracic and abdominal cavity (transabdominal and transrectal), and cytological examination of abdominal fluid was unremarkable. Ultrasound examination of the right and left thyroid lobes was performed. The left lobe was 67 mm in length, 51 mm in width, and 44 mm in depth. The right lobe was 64 mm in length, 50 mm in width, and 58 mm in depth. Both lobes were markedly enlarged and abnormal in appearance. The left thyroid was of mixed echogenicity with areas of hyperechoic striations distributed diffusely through the parenchyma. There were cystic regions within the left thyroid lobe with thin internal septa. The right thyroid was also of mixed echogenicity but overall was more echogenic than the left gland with few smaller cystic regions. Blood flow was detected in both glands by color-Doppler imaging.

Plasma collected at 6:00 pm on the day of examination had elevations in T4 (84 nmol/L; reference interval: 19–58 nmol/L), T3 (2.4 nmol/L; reference interval: 0.45–1.2 nmol/L), and free T4 (>129 pmol/L; reference interval: 15.4–23.2 pmol/L).c Repeat sampling 2 days after presentation showed similar elevations (T4 78 nmol/L, T3 1.8 nmol/L, and free T4 >129 pmol/L). A T3-suppression test was performed utilizing 2.5 mg 3,5,3′-triiodothyro-l-thyronined diluted in 5 mL sterile saline solution (0.15 M NaCl) by intramuscular injection at 6:00 pm on day 1, and at 8:00 am and 6:00 pm on days 2, 3 and 4. Plasma T4 and T3 (Fig 1) concentrations were determined 5 minutes before each dose of T3 and at 8:00 am and 6:00 pm on days 5 and 7, and at 8:00 am on day 10. T4 concentrations failed to be suppressed during administration of T3 and during the following 6 days. The failure of exogenous T3 administration to suppress endogenous thyroid hormone concentrations was consistent with hyperthyroid disease.

Figure 1.

 Plasma T3(▪) and T4(▴) concentrations during T3 suppression test.

Thyroid stimulating hormone (TSH) concentrations were measured by double-antibody radioimmunoassay with equine TSH antiserum and equine TSH antigen.e,2,3 These samples were frozen at −70 °C until analysis could be performed (80 days after admission). However, the lack of repeatability in measuring TSH concentrations in these samples made it difficult to ascribe biological significance to the data.

An overnight dexamethasone suppression test4 was performed (0.04 mg/kg dexamethasone sodium phosphate IM). Blood samples for cortisol measurement were collected 5 minutes before dexamethasone administration and at 8, 12, 16, 20, and 24 hours after injection. Although cortisol concentration was markedly lower (37 nmol/L) 8 hours after administration of dexamethasone, it was incompletely suppressed (Fig 2).

Figure 2.

 Cortisol concentrations after administration of dexamethasone (↑).

Ventral, oblique, and lateral images of the cervical region were obtained utilizing nuclear scintigraphy. Static acquisition images were made 60 minutes after injection of pertechnetate 30 mCi Na+ 99mTcO4). The left thyroid lobe was enlarged with increased radiopharmaceutical uptake (Fig 3). The uptake was heterogenous with a slight photopenic region in the caudal pole of the left thyroid lobe. There was low-intensity radiopharmaceutical uptake in the right thyroid lobe.

Figure 3.

 Images acquired 60 minutes post radionuclide injection: right lateral (R), ventral (V) and left lateral (L) views of proximal cervical region. Note intense uptake within left lobe, low intensity uptake within right lobe, and normal salivary gland uptake.

A Tru-Cut biopsy specimen of each thyroid lobe was obtained percutaneously and fixed in formalin. The majority of follicles in the right thyroid lobe were devoid of colloid, and the remaining follicles were irregularly sized and distended with colloid. No distinguishable follicles were present in the left thyroid lobe, which consisted of polygonal cells arranged in nests or acinar structures supported by scant fibrovascular stroma. The cells had scant to abundant, eosinophilic cytoplasm, poorly defined cellular borders, variably sized nuclei, prominent nucleoli, and infrequent mitotic figures. A diagnosis of left-sided thyroid adenocarcinoma was made. The histopathology of the right thyroid was inconclusive, but suggestive of atrophy, with scattered cystic follicles.

Computed tomography was performed after induction of general anesthesia. The pituitary gland was enlarged and protruded dorsally from the sella turcica. These findings were consistent with pituitary adenoma or hyperplasia. The only other intracranial abnormality was slight asymmetry of the lateral ventricles, which was considered unlikely to be of clinical significance. Both thyroid lobes were markedly enlarged and heterogenous.

A diagnosis was made of hyperthyroidism. Concurrent pituitary–adrenocortical axis dysfunction, with elevations in cortisol and ACTH, were considered secondary to hyperthyroidism, primary pituitary pars intermedia dysfunction (PPID) or both.5–8 Pancreatic dysfunction with decreased insulin concentrations could also be attributed to thyrotoxicosis.9–11 Therefore, treatment of the hyperthyroidism was instituted. Left-sided thyroidectomy was offered to the owner or the novel treatment, 6-n-propyl-2-thiouracilf (PTU). Treatment with PTU was elected by the owner and was instituted at a dosage of 8 mg/kg PO q24h (2 g total dose). The powdered drug was mixed with food, which was readily consumed by the horse. Thyroid hormone concentrations were measured 5 minutes before initial dosing and repeated at 3-day intervals for 15 days. An increase in bodyweight was evident by day 4 (261.4 kg) and a reduction in heart rate (48 beats/min, Fig 4) by day 10 of PTU administration. At day 14, results of thyroid hormone analysis became available and indicated a reduction in T3, T4, and free T4 (Figs 5a, b and 6) to below reference intervals had occurred by day 9 of treatment. The dosage of PTU was reduced to 2 g PO q48h on day 14. The horse was discharged on day 15. Dietary recommendations after discharge included free choice hay, pasture access, water, salt, and mineral, in addition to 7.3 kg of a commercial feedb per day divided into 4 meals.

Figure 4.

 Heart rate after commencement of PTU administration (↑).

Figure 5.

 T3 concentrations during PTU administration with upper (-) and lower (–) reference intervals.

Figure 6.

 T4 concentrations during PTU administration with upper (-) and lower (–) reference intervals.

Blood sampling for thyroid hormone concentrations and measurement of heart rate and estimation of bodyweight by weight tape were performed once weekly by the referring veterinarian after discharge. At a dosage of PTU at 2 g, thyroid hormone concentrations remained within, or in close association with, reference intervals (Figs 5–7). Heart rate remained stable with continued weight gain (306.8 kg at day 55). Observation of polyphagia and polydipsia also ceased at day 21 of PTU administration and no adverse effects of were detected. The plasma cortisol concentration returned to normal reference intervals (94 nmol/L) at day 48 of PTU administration although plasma ACTH remained elevated (82 pmol/L). The size of both thyroid lobes remained unchanged during this period.

Figure 7.

 Free T4 concentrations during PTU administration with upper (-) and lower (–) reference intervals.

Benign adenomatous enlargement of the thyroid gland is common in older horses and is most commonly unilateral.12,13 Hyperthyroidism is a rare disease of horses with 2 previous cases being reported.14,15 In both these horses, unilateral thyroid enlargement was identified and associated with a primary, endocrinologically active adenoma and adenocarcinoma, respectively. Similar to other species, clinical signs in horses included emaciation, polyphagia, hyperexcitability, polydipsia, and tachycardia.14,15 In this horse, changes to the parenchyma on ultrasonographic exam were both solid tissue with some cystic regions and were most consistent with bilaterally adenomatous hyperplasia. Increased echogenicity of the right thyroid lobe in relation to the left was suggestive of increased fibrosis in this gland. No capsular disruption was present that suggested an invasive neoplastic process. Histopathology of the left thyroid lobe, however, was consistent with adenocarcinoma and inconclusive for the right thyroid lobe.

A TSH-secreting pituitary adenoma (TSH-oma) causing bilaterally hyperfunctioning lobes can explain bilaterally enlarged thyroid lobes.18 TSH-omas are a rare cause of hyperthyroidism in humans and have not been reported in the horse. Concentration of TSH in affected humans is highly variable and can be normal to greater than 3 times reference intervals.16 TSH molecules secreted by pituitary tumors in humans are heterogeneous and may have normal, reduced, or increased biological and immunological activities.17 Unfortunately, normal values for aged horses or those with concurrent disease have not been established and further research is required to validate equine TSH testing.2 Magnetic resonance imaging may have been useful to further evaluate pituitary architecture although definitive diagnosis of TSH-oma would require histopathological examination. Surgical resection is the recommended treatment for TSH-secreting pituitary tumors in humans.16,18 If surgical resection is not possible or unsuccessful, medical management consists of radiotherapy or long-acting somatostatin analogs (octreotide, lanreotide).16 It is hypothesized that TSH may also be involved in tumorigenesis in the thyroid gland.19

The lack of uptake by the right thyroid lobe during nuclear scintigraphy was thought to be caused by infarction, atrophy, and an associated decrease or absence of endocrinological function. Spontaneous infarction or hematomas of functional tumors, for example thyroid and parathyroid adenomas, are uncommonly reported in humans and can be associated with acute inflammatory changes and temporary resolution of hyperthyroid disease.20–22 Some thyroid tumors in humans also can undergo extensive necrosis, fibrosis, cystic degeneration, and infarction after fine-needle aspiration procedures.23,24

It is of interest that nuclear scintigraphy was the only diagnostic modality able to elucidate the difference in function between the lobes. This would be clinically significant in those horses in which unilateral hemithyroidectomy could be considered as a treatment option in animals with bilateral thyroid enlargement. It is evident in this horse that bilateral enlargement of lobes might not be associated with equivalent function. The removal of the nonfunctional lobe in these horses would not result in clinical improvement. In human cases of parathyroid adenoma infarction, parathyroidectomy is still considered as regeneration of the parathyroid adenoma may occur.20 The use of nuclear medicine for the evaluation of function, detection of metastatic disease and measurement of metabolic tumor volume in thyroid disease is well established in human and small animal medicine.25,26

Concurrent PPID and adrenal dysfunction with elevated ACTH, cortisol, and abnormal dexamethasone suppression test results have not been reported in previous equine cases of hyperthyroidism. In humans, cats and cows with hyperthyroid disease, elevations in ACTH and cortisol are reported and thought to be associated with activation of the pituitary-adrenocortical axis.5–8 In a human study of Cushing's syndrome, 8.4% of patients were found to have primary hyperthyroid disease.27 Marked elevations in serum cortisol concentrations have also been measured in a canine case report of hyperthyroidism.28 This was thought to be associated with alterations in cortisol metabolism or stress of the hyperthyroid state. In humans with severe hyperthyroidism, cortisol concentrations are reduced following a dexamethasone suppression test, but response to low-dose ACTH stimulation is lower in the hyperthyroid than in the euthyroid state, indicating impairment of adrenocortical reserve.29 Treatment of hyperthyroidism results in normalization of ACTH, cortisol, and responses to hormone stimulation tests.5,30 Treatment of rats with PTU has also been found to counteract both basal and ACTH-induced adrenal steroidogenesis through attenuation of the activity of 11 β-hydroxylase and cAMP production in rat zona fasciculata-reticularis cells.31 Long-term monitoring of ACTH, cortisol, and dexamethasone suppression tests would be required to assess whether normalization of adrenocortical axis occurs in this horse. It is interesting to note that the cortisol concentration decreased to within normal reference intervals within 48 days of PTU treatment, although ACTH remained elevated. It is possible that PPID could be independent or exacerbated by hyperthyroidism in this horse and future treatment with pergolide may be necessary if adrenocortical dysfunction persists.

The reduction in the post-prandial insulin concentration can be explained by defective insulin response and production during hyperthyroidism. Human hyperthyroid patients can display both an inability to increase the insulin response appropriately to hyperglycemia and increased proinsulin concentrations both in the fasting state and in response to a meal.9 In rats, thyroid hormones cause a decrease in glucose-induced insulin secretion associated with genomic and non-genomic effects of thyroid hormone on glucose-induced insulin secretion.10 An increased rate of β-cell death caused by apoptosis causes a decrease in insulin content and glucose-induced insulin secretion from the pancreas in hyperthyroidism has also been identified.11 Further investigation of insulin and glucose concentrations with multiple sampling and dynamic testing would be required to definitively identify glucoregulatory dysfunction.

Propylthiouracil is a thioureylene that antagonizes thyroid peroxidase. This enzyme catalyzes both the incorporation of iodine into the tyrosine residues and the coupling of the outer phenol ring to the inner, thus blocking synthesis of thyroxin and triiodothyronine. It is commonly used in human medicine for the treatment of hyperthyroidism. Its use has been documented at 4–6 mg/kgPO q24h in horses to induce hypothyroidism, as a model to study the disease.32–34 The use of an increase in previously described doses at 8 mg/kg daily dosage was subjective and based on the marked elevations in thyroid hormone concentrations, lobe size and expected weight gain. The authors were, however, surprised by both the magnitude and rapid reduction in hormone concentrations after initiation of PTU treatment. Previous studies in euthyroid horses had reported the induction of a hypothyroid state with reduction in concentrations of T3 and T4 by day 28,33 T3 and T4 by day 52,34 or T3 by day 7, T4 by day 35, and free T4 by day 28.32 As with other reports in horses,32–34 no adverse effects were detected with its use and oral dosage in food was effective. Cost of PTU per dose was $2.40.

In conclusion, concurrent adrenocortical and glucoregulatory dysfunction may exist in equine cases of hyperthyroidism. PTU at an initial dose of 8 mg/kg PO at a 48-hour interval appears to be a safe, easy to administer, relatively inexpensive, and effective treatment for hyperthyroidism in the horse. The use of nuclear scintigraphy for the evaluation of thyroid function is invaluable and should be performed before hemithyroidectomy is undertaken.


aPfizer Animal Health, New York, NY

bTriple Crown Nutrition, Inc, Town Wayzata, MN

cAnimal Health Diagnostic Center, College of Veterinary Medicine, Cornell University, NY

dSigma Chemical Co., St Louis, MO

eDivision of Animal Sciences, University of Missouri, Columbia, MO

f6-n-propyl-2-thiouracil, Sigma Chemical Co.


This study would not have been possible without the technical expertise and knowledge of Prof. D.H. Keisler (Division of Animal Sciences, University of Missouri, Columbia, MO) and Prof. V.K. Ganjam (Department of Biomedical Sciences and Veterinary Medicine and Surgery, College of Veterinary Medicine, University of Missouri, Columbia, MO) for TSH radioimmunassay, and Dr N.J. Place, Dr B.J. Schanbacher and Mr S.V. Lamb (Animal Health Diagnostic Center, College of Veterinary Medicine, Cornell University, NY) for thyroid hormone analysis. Additional assistance was also provided by Dr T. LeRoith (Department of Pathobiology, Virginia-Maryland Regional College of Veterinary Medicine, Blacksburg, VA).