Susan E. Orosz, Ph.D., DVM - Jack W. Oliver, DVM, Ph.D.
Edward C. Schroeder, DVM, MS - Nancy Zagaya, LVMT - L. Kim Abney, BS

Department of Comparative Medicine
College of Veterinary Medicine
The University of Tennessee
P.O. Box 1071
Knoxville, TN 37901-1071

Abstract:    This paper describes a useful protocol for the clinical evaluation of thyroid function in Amazon parrots using thyrotropin releasing factor (TRF). When TRF was administered at 50 µg/kg IV, to blue-fronted Amazons (Amazona aestiva, n=8) and Hispaniolan (Amazona ventralis, n= 12) parrots, they had a decreased T4 level and an increased T3 level at 2 hours regardless of whether they were fasted or allowed to feed ad libitum.

Introduction
Primary hypothyroidism involves the thyroid gland and its inadequate production of thyroxine (T4) and triiodothyronine (T3). It is the most common form of hypothyroidism in mammals, and has also been reported in birds. In a strain of obese chickens, primary hypothyroidism has been determined to be the result of a hereditary thyroiditis. Its pathology resembles that of Hashimoto's disease, an autoimmune thyroiditis of humans. Thyroiditis has also been documented in other avian species including an Amazon parrot. Rudas et al. have described a malabsorption syndrome that is associated with low levels of T3 and T4. Budgergars (Melopsittacus undulatus) that have an iodine-deficient diet can develop goiter and pigeons (Columba livia) may also develop goiter secondary to goitrogenic foods. Primary hypothyroidism has also been documented in a scarlet macaw (Ara macao).
A decrease in circulating thyroid hormones affects the metabolism of many of the organ systems, resulting in diffuse signs in mammals and in birds. Plasma levels of T3 and/or T4 have been used for the diagnosis of hypothyroidism in mammals and T4 in  birds. However, baseline values can vary between studies and between species of birds. A variety of factors have been reported to influence thyroid hormone concentrations in mammals and to result in a misdiagnosis in a normal animal or an erroneous result in diseased one. These factors include: nutritional status, concomitant disease, current drug thearpy, endogenous hormones, estrus and normal daily fluctuations in plasma thyroid hormones.

In birds, factors known to affect plasma levels of T4 include drugs, handling, bleeding, food intake, enviornmental temperature, increased plasma corticosterone levels and infection with Eimera. Normal plasma concentrations of T4 in birds are much lower than those of mammals and can lead to a diagnosis hypothyroidism when in fact thyroid function is normal.

For these reasons, a stimulation test is considered to be more aaccurate for the evaluation of thyroid gland function and diagnostioc interpretation. Thyroid function test utilizing thyroid stimulation hormone (TSH) have been described and a thyroid releasing factor (TRF) test which acts to release TSH, has been developed for dogs to evaluate gland function. A TSH stimulation test has also been developed for chickens, psittacines and racing pigeons, and the test has been considered to be useful in determining thyroid function. TSH is only available at considerable expense when purchased from large chemical firms like Sigma Chemical (St. Louis, MO) and biological effect is considered variable (Bruce McNeil, UTCVM Pharmacy, pers. comm.) TRF is available from selected pharmacies and from chemical concerns at an affordable price.

The purpose of this study was to evaluate the normal response of the thyroid gland to a TRF stimulation protocol in two types of Amazon parrots, blue-fronted Amazons (Amazona aestiva) and Hispaniolan parrots (Amazona ventralis).

Materials and Methods
Two species of Amazon parrots were used for this study. The blue-fronted Amazon parrots (n=8), were housed individually or in pairs; Hispaniolan parrots (n= 12), were housed in a similar manner. Both species were kept in the same room on a 12 hour light/dark cycle and fed pellets (Lafeber Co., Cornell, IL) and water ad libitum. They also received approximately 20% of their diet as sweet potato and other vegetables in the morning. The temperature was maintained at 24.4 - 25.5'C (76 - 78'F). All birds were considered normal on physical examination prior to the beginning of the study, and were observed daily throughout the time period. These birds were not tame and each was captured, sampled and then released. Blood was collected by venipuncture from the jugular or basilic vein and placed into heparinized tubes. Plasma was separated within 4 hours after collection and kept frozen (-20'C) until analyzed. The TRFa was solubilized with a sterile saline at a concentration of 100 µg/ml, and 50 µg/kg was administered intravenously into the medial metatarsal vein.

Plasma samples were/taken throughout a 3 hour time period in both groups of birds for measurement of T3 and T4 levels. Both groups were evaluated in a nonfasted and fasted (12 ± 1 hours post food withdrawal in the pm) state.

Plasma concentrations of T3 and 4 T4 were analyzed by radioimmunoassay using commercially available kits. All samples were assayed in duplicate. Assay validation was performed on pooled plasma from the Amazon parrots as described previously. Statistical analysis of the endocrine results was performed using a commercially available package. Statistical significance was set at P<. 0 1. Values were reported as mean ± standard deviation.

Results
Triiodothyronine and T4 concentrations were measurable in the plasma from both groups of Amazon parrots. However, the baseline levels of T4 and T3 varied between the short (Fig 1 & 2, respectively) and long time intervals (Fig. 3 & 4) for the blue-fronted Amazons. Due to the small size of the Hispaniolan parrots, blood was drawn only twice during the sampling period. T4 and T3 values (Fig. 5 and 6, respectively) were lower in fasted birds in comparison to the unfasted ones. The difference between the fasted and nonfasted baseline levels were significantly different in the Hispaniolan group (P<.01).

Serum thyroxine levels at 2 hours were significantly lower (p<.0l) from baseline values in the Hispaniolan parrots (Fig. 3). This trend also occurred in the blue fronts, but the difference in levels was not statistically significant. This may be due to the small sample size and/or the fact that one parrot's values were considerably higher compared to the others in the group, which exaggerated variability. All birds, except one blue front, exhibited a decrease in their 2 hour plasma level compared to their baseline levels for T4 (Fig. 5).

Serum triiodothyronine levels for each parrot were increased over baseline at two hours, except that magnitude of response compared to T4 was less (Fig. 4 and 6). There were no significant differences between fasted and nonfasted groups from each species for T3. No significant differences were noted between sexes and species of Amazons for either T3 or T4 levels. The increase in post-TRF triiodothyronine was not statistically significant but may have biological significance.

It appears that when normal Amazon parrots are administered 50 µg/kg of TRF intravenously, their serum levels of T4 will decrease by approximately 50% while their T3 levels will increase 5-10% two hours post-administration.

Discussion
Baseline levels of T4 for blue-fronted Amazons were greater in this study than previously reported by Zenoble et al., but less than those of Lothrop et al. The baseline values from this study were similar to those in the "stressed" cockatiel group of Harms et al., but were lower than the acclimatized group prior to thyroid ablation. The values in the Amazons from our study were much lower than those of racing pigeons throughout the entire year. However, the baseline values were in the same range as that of immature chickens.

Triiodothyronine levels in this study were less than those reported by Zenoble et al. but greater than those reported by Lothrop et al. Triiodothyronine was not measured in experiments using cockatiels and racing pigeons. The values in this study were less than those reported in immature chickens.

The variability in baseline thyroid hormone values between studies underscores the advantage of using thyroid stimulation testing. Exogenously administered TSH in previous studies in parrots increased T4 levels 4-6 hours later by a minimum of 50% to a maximum of twice baseline. Peak levels appeared from 6-8 hours in the cockatiel study with a Mold increase in the acclimatized group and a 14-fold increase in the stressed group. There was a 2.5 fold increase in racing pigeons up to 32 hours post TSH administration. In the present study using TRF instead, the T4 levels decreased from 31-64%. Therefore, the changes using TSH in previous studies were much greater than in the present study using TRF. However, from a clinical perspective, the magnitude of change in the tests is not as critical as the ability to reach a diagnosis about thyroid function.

It was hypothesized for this study that TSH stimulated the release of T3 and T4, resulting in increased circulating levels of these hormones. However, it was also hypothesized that TRF primarily stimulates the release of growth hormone with a minimal effect (if any) on TSH release. Growth hormone has been reported to inhibit peripheral deiodination of thyroid hormones in birds, resulting in elevated T3 levels. In order for the latter to occur, an adequate supply of T4 would have to be available in the peripheral tissues. With increased levels of circulating T3, negative feedback would occur at the level of the pituitary, resulting in a decrease in TSH out put and subsequent T4 decrease.

The decrease in T4 levels support this negative-feedback hypothesis. The study results also suggest that a TRF stimulation test in parrots is useful in assessing normal function of the hypothalamo-pituitary-thyroid axis clinically. This test has an advantage over single value tests such as those that analyze T4 in that two values are obtained and evaluated: both a rise in the levels of T3 and a drop in T4 for normal function. These results also suggest that neither plasma levels of T3 or T4 will change from baseline in birds that are hypothyroid. This test, when combined with physical examination findings and other associated data, should be considered for the diagnosis of abnormal thyroid function in Amazons species. Further studies such as clinical trials in other avian species and/or use in a hypothyroid model, would be useful in evaluating the efficacy of this TRF test for assessing thyroid function.

Acknowledgements
The authors would like to thank Ms. Sonia Doss and Elizabeth Bailey for their technical assistance. This study was funded by the Lafeber Co., Cornell, Illinois, USA, the Department of Comparative Medicine and the College of Veterinary Medicine, The University of Tennessee, Knoxville,TN, USA.

REFERENCES
1. Jeffers JG. Recognizing and managing the effects of canine hypothyroidism. Vet Med. 1990; 1294-1380.

2. Beale KM. Current diagnostic techniques for evaluating thyroid function in the dog. Vet Clin N Amer. 1990; 20:1420-144 1.

3. Cole RK, Kite JH, Wick G, Witebsky. Inherited autoimmune thyroiditis in the fowl.

4. Zipper J, Tamaschke CH. Pathologishe Schilddrusen be ftmde bei Vogeln In: Verhandlungsberichte XIV Intern S~mp Krankh Zootiere, Wroclaw. Berlin, Akademie Verlag, 1972:113-122.

5. Rudas P, Salyi G, Szabo J. Decreased thyroxine, triodothyronine and 5-deiodination levels in malabsorption syndrome. Avian Dis 1985; 30:293-297.

6. Blackmore DK. The pathology and incidence of thyroid dysplasia in budgerigars (Mel opsittacus undulatus). Vet Rec; 1965; 75:1068-1072.

7. Rosskopf WJ, Woerpel RW. Remission of lipomatous growths in a hypothyroid budgerigar in response to L-thyroxine therapy. VMSAC 1983; 78:1415-1418.

8. Schulmberger HG. Spontaneous goiter and cancer of the thyroid in animals. Ohio J Sci 1955; 55:23-43.

9. Hollander WF, Riddle 0. Goiter in domestic pigeons. Poult Sci 1946;25:20-27.

10. Lumeij JT. Endocrinology. In: Ritchie BW, Harrison GJ & Harison LR (ed) Avian Medicine: principles and application. Lake Worth, FL: Winger's Pub Inc., 1994:482-606.

11. Oglesbee BL. Hypothyroidism in a scarlet macaw. JAVMA 1992; 201:1599-1601.

12. Bruyette DS, Nelson RW, Bottoms GD. Effect of thyrotropin storage on thyroid- stimulating hormone response testing in normal dogs. J Vet Int Med 197; 1:91-94.

13. Rosskopf WJ, Woerpel R, Rosskopf G, Vander Water D. Normal thyroid values for common pet birds. Vet. Med. Sm Anim Clin 1982; March 1982:409-412.

14. Mellen WJ, Wentworth BC. Observations on radiothyroidectornized chickens.

15. Bilezikian JP, Loeb JN, Gammon DE. Induction of sustained hyperthyroidism and hypothyroidism in the turkey: Physiological and biochemical observations. Poult Sci 1980; 59:628-634.

16. Belshaw BE. Thyroid diseases. In: Ettinger SJ (ed). Textbook of Veterinary Internal Medicine. Philadelphia: WB Saunders, 1983:1592-1614.

17. Lothrop CD, Loomis MR, Olsen JH. Thyrotrophin stimulation test for evaluation of thyroid function in psittacine birds. JAMA 1985; 186:47-48.

18. Zenoble RD, Kemppainen RJ, Young DW, Clubb SL. Endocrine responses of healthy parrots to ACTH and thyroid stimulating hormone. JAMA 1985; 187:1116-1118.

19. Lumeij JT, Westerhof 1. Clinical evaluation of thyroid function in racing pigeons (Columba livia domestica) Avian Path. 1988; 17:63-70.

20. Williamson RA, Davison TJ. The effect of a single injection of thyrotrophin on serum concentrations of thyroxine, triiodothyronine and reverse triiodothyronine in the immature chicken (Gallus domesticus) Gel & Comp Endo 1985; 58:109-113.

21. Rosychuk KAW. Thyroid hormones and antithyroid drugs. Vet Clin North Amer 1982; 12:111-148.

22. Chopra IJ, Hershman IM,, Pardridge WM. Thyroid function in nonthyroidal illness. Ann Intern Med 1983; 9:946-957.

23. Utiger RD. Decreased extrathyroidal triiodothyronine production in nonthyroidal illness: benefit or harm? Am J Med 1980; 69:807-8 10.

24. Chopra IJ. Assessment of daily production and significance of thyroidal secretion of 3.3',5-triiodothyronine (reverse T3) in man. J Clin Invest. 1976; 58:32-40.

25. Nelson RW, Evinger JV, Bottoms GD. Evaluation of serial serum thyroid hormone concentrations in the dogs. Proc Amer College Vet Internal Med 1985; 127.

26. Williamson RA, Misson BH, Davison TJ. The effect of exposure to 40'C on the heat production and the serum concentration of triiodothyronine, thyroxine and corticosterone in immature domestic fowl. Gen & Comp Endocrin 1985; 60:178-186.

27. Davison TF, Rea J, Freeman BM. Effect of increased circulating corticosterone on the 4 plasma concentrations of thyroxine and triiodothyronine in chickens. IRCS Med Sci 1985; 13:692-693.   

28. Davison TF, Chapman HD, Harvey S. Endocrine changes in the fowl during infection with Eimeria maxima. Res Vet Sci 1985; 38:296-300.

29. Kuhn ER, Decuypere E, Huybrechts LM, Darras VM. Hormonal control of the hepatic 5'-monodeiodination activity and its relation to the somatomedin production in the chicken. In: Wada M, Ishii S, Scanes CG (ed). Endocrinology of Birds. Molecular to Behavioral. Berlin: Springer Verlag 1990:129-142.

30. Oliver JW, Waldrop V. Sampling protocol for thyrotropin stimulation test in the dog. J Am Vet Med Assoc 1983; 182:486-489.

31. Evinger JV, Nelson RW, Bottoms GD. Thyrotropin-releasing hormone stimulation testing in healthy dogs. Am J Vet Res 1985; 46:1323-1325.

32. Harms CA, Hoskinson JJ, Bruyette DS, Carpenter JW, Galland J, Veatch JK, Wilson SC, Baier JG. Development of an experimental model of hypothyroidism Jn cockatiels (Nymphicus hollandicus). Am J Vet Res 1994; 55:399-404.

33. Harvey S, Sterling RJ, and Phillips JG. Diminution of thyrotrophin releasing hormoneinduced growth hormone secretion in adult domestic fowl (Gallus domesticus). J Endo 1981; 89:405-410.

34.  Newcomer WS, Haung FS. Thyrotropin-releasing hormone in chicks. Endo 1984; 95:18-320.

Figures
Figure 1. Mean ± SD plasma T4 levels before and < 1 h post TRF stimulation (50 µg/kg, IV) in blue-fronted Amazon parrots. To convert nmol/1 to ng/ml divide by 1.29 for T4. Data represent baseline comparison with student's t test for grouped data.

Figure 3. Mean ± SD plasma T4 levels before and 1 h post TRF stimulation (50 µg/kg, IV) in blue-fronted and Hispaniolan parrots. *T4 levels of the Hispaniolan parrots 2 h post TRF were significantly less than different baseline (p<.Ol) using student's t test.for grouped data.

Figure 2. Mean ± SD plasma T3 levels before an < 1 h post TRF stimulation (50 µg/kg, IV) in blue-fronted Amazon parrots. Plasma levels at 45 minutes were increased significantly (p <.05) over baseline. To convert nmol/1 to ng/ml divide by 1.54 for T3.

Figure 4. Mean ± SD plasma T3 levels before and > 1 h post TRF stimulation (50 µg/k, IV) in blue-fronted and Hispaniolan parrots. Although T3 levels increased over baseline, values were not statistically significant.

Figure 5. Comparison of unfasted and fasted mean ± SD plasma T4 levels between baseline and 2 h post TRF (50 µg/g, IV). The 2 hour unfasted and fasted groups were deceased significantly (p < .0 1) from baseline using a student's t test for grouped data. The same trend occurred in the blue-fronts but one individual remained unchanged and elevated.

Figure 6. Comparison of unfasted and fasted mean ± SD plasma T3 levels between baseline and 2 h post TRF (50 µg/kg, IV). The 2 hour groups were increased from 10-30% over baseline but not significantly.

A decrease in circulating thyroid hormones affects the metabolism of many of the organ systems in the body. This results, typically in mammals, in diffuse signs that are not pathognomonic for hypothyroidism (Bruette, 1987). Plasma levels of triiodothyronine (T3) and thyroxine (T4) are often used to help establish a diagnosis of hypothyroidism in mammals (Belshaw, 1979), and while plasma levels of thyroxine are used to diagnose this condition in birds (Lumeij, 1994). However, a variety of factors can influence thyroid hormone concentrations and result in a misdiagnosis in a normal animal or an erroneous result in a diseased animal. Factors reported in mammals include: nutritional status, concomitant disease, current drug therapy, endogenous hormones, estrus and normal daily fluctuations in plasma thyroid hormones (Rosychuk, 1982; Chopra, 1983; Utiger, 1980; Chopra, 1976; Nelson, 1985).