GBGB Logo
Site Search
Information for Vets | Thyroid Function

Thyroid Function in Greyhounds


Introductory note to paper by Dr Peter Graham
Note: Dr Peter Graham, an internationally recognised expert in the area, has produced a scientific review of the current situation regarding thyroid disease in greyhounds. This has been required because of the apparently high rate of diagnosis and treatment of hypothyroidism, in spite of the scientific literature indicating that the condition is relatively uncommon. Most of these diagnoses are being made on the basis of a single test (T4) using conventional "pet-type" concentrations when there is strong evidence that greyhound levels are much lower.

This document has been published to give guidance to both trainers and veterinary surgeons, and some of the most important points are highlighted in this introduction. The entire paper should be read to obtain the complete picture.

Summary of Dr Graham's paper:

  • Thyroid disease in greyhounds is a controversial topic. There is increasing awareness of the differences in thyroid hormone levels between sighthounds as a group (and greyhounds specifically) and the general dog population. This feature of greyhounds has been recognised for more than 10 years but knowledge of it does not appear to have been fully disseminated.

  • Most endocrinologists believe that these differences necessitate greyhound specific reference ranges for thyroid hormones much the same as the widely accepted higher greyhound reference range values for red blood cell parameters (red cell count, haematocrit, haemoglobin). They do not see evidence of hypothyroidism as a clinical syndrome in the overwhelming majority of greyhounds with thyroid hormone levels below those of the general dog reference ranges. The treatment of such normal animals with thyroid hormone preparations may have disadvantageous effects on the animals.

  • Within the greyhound community, "low" thyroid hormone concentrations are being blamed for diverse syndromes that are not typical of hypothyroidism as it has been defined and experienced until now. Manifestations cited include: "low self-esteem", "spookiness", fearfulness, shyness, failure to interact with other dogs, reduced fertility in addition to reluctance to exercise, sensitivity to cold and bald-thigh syndrome. Some believe that low serum thyroxine concentrations in greyhounds reflect a true hypothyroidism genetically selected and highly prevalent within the breed. However, Dr Graham?s conclusion is that hypothyroidism in the greyhound is probably much less common than is generally perceived.

  • Given published data and current assay capabilities, there would be no such thing as a "Greyhound with a low-T4" if greyhound specific reference ranges were used. For that reason, it is clear that hypothyroidism cannot be diagnosed based on a single T4 assay. It is also suggested that the treatment of such greyhounds using thyroid preparations may carry a risk to the greyhound.

  • Future work should focus on correlating other measures of thyroid function (e.g. T3 and TSH) with perceived clinical syndromes of hypothyroidism in the greyhound in order to generate some diagnostic performance data (diagnostic sensitivity and specificity) and clarify which clinical presentations are associated with hypothyroidism and those that are clearly not.

Thyroid function in Greyhounds
Peter A Graham BVMS PhD CertVR DipECVCP MRCVS

Dr Graham is currently Managing Director of Cambridge Specialist Laboratory Services and NationWide Laboratories. Previously he was Section Chief and Assistant Professor at Michigan State University?s Endocrine Diagnostic Section.

Introduction:
Thyroid disease in Greyhounds is a controversial topic. There is increasing awareness of the differences in thyroid hormone levels between sighthounds as a group (and Greyhounds specifically) and the general dog population. This feature of Greyhounds has been recognised for more than 10 years but knowledge of it does not appear to have been fully disseminated. Most endocrinologists believe that these differences necessitate Greyhound specific reference ranges for thyroid hormones much the same as the widely accepted higher Greyhound reference range values for red blood cell parameters (red cell count, haematocrit, haemoglobin). They do not see evidence of hypothyroidism as a clinical syndrome in the overwhelming majority of Greyhounds with thyroid hormone levels below those of the general dog reference ranges.

However, within the Greyhound community, "low" thyroid hormone concentrations are being blamed for diverse syndromes that are not typical of hypothyroidism as it has been defined and experienced until now. Manifestations cited include: "low self-esteem", "spookiness", fearfulness, shyness, failure to interact with other dogs, reduced fertility in addition to reluctance to exercise, sensitivity to cold and bald-thigh syndrome. Some believe that low serum thyroxine concentrations in Greyhounds reflect a true hypothyroidism genetically selected and highly prevalent within the breed.

Hypothyroidism in general:
Hypothyroidism is the most common endocrine disorder of the dog. It results when thyroid hormone production the thyroid glands is inadequate. The thyroid glands produce thyroxine (T4) and to a lesser extent tri-iodothyronine (T3) in response to stimulation by pituitary origin thyrotropin (thyroid stimulating hormone; TSH). T4 is the pro-hormone of T3, in that an iodine atom is removed from T4 by monodeiodinase enzyme to produce T3 the metabolic active form of thyroid hormone that is able to bind to receptor sites within the nucleus of cells. Thyroid hormone positively influences metabolic rate, i.e., higher levels of thyroid hormone are associated with higher metabolic rates. Consequently when there is a deficiency of thyroid hormone, the principal clinical signs relate to metabolic rate, growth and tissue turnover. Classically, hypothyroid animals become overweight, lethargic, cold intolerant and display skin and hair abnormalities including alopecia (or failure to regrow hair), hyperpigmentation, pyoderma and seborrhoea. They are also susceptible to pyoderma.

Both T4 and T3 are highly protein-bound when circulating in plasma. Less than 0.1% of T4 is free to interact with tissues and be converted to T3.

Hypothyroidism can arise in dogs either as a result of immune mediated thyroiditis or idiopathic thyroid follicular atrophy. Du ring immune-mediated thyroiditis, antibodies to thyroglobulin, the principal thyroid antigen, can be found in the circulation (TgAA). Overall, the percentage of hypothyroid cases in which TgAA is found is about 50. There is some recent evidence to suggest that this 50:50 ratio in pathogenesis does not hold true in all breeds.
The progression of thyroiditis to the point at which functional reserves are destroyed and hypothyroidism develops is considered to be slow and it may take months or years from the first evidence of thyroiditis.

When the thyroid glands are destroyed either by immune mediated or idiopathic processes, the failure to produce thyroid hormone causes the negative feedback on the pituitary production of TSH to be lost such that it production increases and it should be detectable at a higher concentration in plasma.
Factors influencing thyroid status that complicate thyroid diagnosis

In an ideal world, low thyroid hormone circulating in plasma would mean that a person or animal had hypothyroidism. Unfortunately, at the centre of control of metabolic rate, thyroid hormone is subject to other influences that complicate the interpretation of plasma concentrations. Thyroid hormone concentrations are subject to physiological influence at each of the levels described above; centrally, by altered TSH production, in circulation by altered protein binding, and peripherally by altered de-iodinase activity.

The most important factor that needs consideration, is the effect of non-thyroidal illness. During times of illness and stress, thyroid hormone concentrations often decrease (low T4 (or T3) state of medical illness previously known as "euthyroid sick"). This is considered to be an adaptive protective mechanism that, by lowering metabolic rate, helps preserve resources to increase the chance of recovery. Physiologically, it would seem unwise to override this adaptation with exogenous thyroid hormone. In human medicine, there is no clear indication that such supplementation is beneficial and some evidence to suggest that exogenous thyroid therapy could negatively influence recovery from certain non-thyroidal illnesses. It has been observed in thyroxine treated hypothyroid humans, that some prefer or enjoy the feeling of being mildly thyrotoxic and many are reluctant to reduce their doses to appropriate levels despite the potential for long-term detrimental effects. Publications that investigate the influence on exogenous thyroxine therapy for low T4 state of medical illness in naturally occurring canine disease are not available.

The administration of certain pharmaceutical preparations can also reduce plasma thyroid hormone concentration. Classic examples are barbiturates and glucocorticoids. The effect of barbiturates is further complicated by the duration of therapy but it is thought that central effects, altered protein binding and hormone metabolism/clearance have a role in these drug effects. TSH concentrations generally do not increase during drug therapy and it is likely that this is another example of an adaptive process that reduces thyroid hormone concentrations appropriately and in the long-term interests of the animal.
However the "sulpha" group of antimicrobials given for long enough at sufficient doses can actually cause a reversible primary hypothyroidism including both decreased thyroid hormones and increased TSH.

Other factors that can influence thyroid hormone concentrations include exercise, diet and the reproductive cycle. However, during the normal course of events these influences are not of sufficient magnitude to compromise diagnostic interpretation with regard to the presence or absence of hypothyroidism.

Current general diagnosis - Evidence based medicine
The mainstay of thyroid diagnosis for dogs in general is the combination of serum T4 and TSH with FT4d and TgAA for additional suport (Graham and Mooney 2005). Total T3 is not currently favoured.

Total T4 (TT4) is relatively inexpensive and easily available. It has good diagnostic sensitivity, in that few cases of hypothyroidism are missed by TT4 measurement. An exception is when hypothyroid dogs have T4 cross-reacting species of TgAA (T4AA) in their circulation which cases false normal or false high T4 values to be generated by most assay systems. Unfortunately, mostly for the non-thyroidal reasons mentioned above, TT4 has very poor diagnostic specificity with many subnormal T4 values being due to things that are not hypothyroidism. Diagnostic specificity has been reported as low as 75% for TT4 (Dixon and Mooney 1999), meaning that of dogs that present with signs similar to hypothyroidism but that do not have the disease, 25% will have subnormal T4 concentration (due to physiological response to non-thyroidal illness) and a false positive diagnosis.

TSH would be expected to elevated in cases of primary thyroid failure, but unfortunately, it is not a perfect test for hypothyroidism either. TSH measurement does have better diagnostic specificity than TT4, i.e., few non-thyroidal conditions cause a high TSH value. However, disappointingly, not all hypothyroid dogs have the expected high TSH. The proportion that do not has been variably reported but compared to gold-standard TSH response testing the figure probably lies between 15 and 20% (Dixon and Mooney 1999).

It makes sense that the combination of TT4 and TSH has the potential to benefit from the advantages of each (good sensitivity of TT4 and good specificity of TSH). Indeed, at prevalence levels of 10%, the positive predictive value (PPV) of T4/TSH ratio exceeds 80% when TT4 alone would have a PPV of less than 30%.

That means that if a positive result is obtained from a T4/TSH ratio, the clinician could be more than 80% confident that the patient was truly hypothyroid. In contrast, faced with a positive (subnormal) TT4 result alone less than 30% of such cases will actually have the disease. At the same time negative predictive value (NPV) of the T4/TSH ratio is excellent at around 99% meaning that almost complete confidence can be had that a dog with a normal T4/TSH ratio does not have hypothyroidism.

Table 1: Diagnostic performance of TT4 used alone for the diagnosis of hypothyroidism

 

TT4
hypoT4
 Not
 Total  
 ve result
89
225
314
28%
PPV
 -ve result
1388289599%NPV
 Total100
900
1000
  
 87%
98%
10%
  
 SensitivitySpecificityPrevalence  
    
Table 2: Diagnostic performance of T4/TSH ratio used for the diagnosis of hypothyroidism in the dog.

TT4/TSH
hypoT4NotTotal
  
ve result87
18
105
83%
PPV
-ve result13
882
895
99%
NPV
Total100
900
1000
  
 87%
98%
10%
  
 Sensitivity
Specificity
Prevalence
  

Free T4 is a theoretically better test for assessing thyroid hormone status. It is, after all, the fraction that is free to interact with tissues and be de-iodinated. There are several measurement techniques available but the only one that works properly measures T4 by sensitive radioimmunoassay following equilibrium dialysis (FT4d). The dialysis step removes any cross-reacting antibodies and eliminates the influence of binding proteins on hormone measurement. This makes the FT4d test more specific than TT4 because the reduced influence of the protein effects of non-thyroidal illness means that there are less false positives; by removing T4AA diagnostic sensitivity is improved. However, for reasons including displacement by free fatty acids and deterioration of the sample during prolonged shipment, the FT4d test does not have perfect diagnostic sensitivity.

Total T3 (TT3) is not currently a test of choice in the diagnosis of hypothyroidism in the dog. The main reason for this is the high prevalence of T3 cross-reacting TgAA (T3AA) in hypothyroid dog sera (>30%) in addition to some of the same problems of non-thyroidal illness and drug therapy that befall TT4. T3AA interfere with T3 measurement techniques causing false normal or false high values to be generated.

The measurement of TgAA can identify immune-mediated thyroid pathology often even before it has progressed to the stage of hypothyroidism. Also the cross-reacting T3 and T4 autoantibodies (T3AA, T4AA) are subsets of TgAA so a negative TgAA result means that there cannot be any interfering cross reacting antibodies and TT4 and TT3 can be relied upon.

In the general canine population, the laboratory diagnosis of hypothyroidism is less than perfect but the combination of a minimum of T4 and TSH greatly improves the chance of a correct diagnosis over the use of TT4 alone. The addition of FT4d can help when the presence of non-thyroidal concurrent illness, drug therapy or T4AA is suspected or known.

Given that hypothyroidism is seldom a life-threatening disease, re-testing at a later date following recovery from known or suspected illness or withdrawal of drug therapy is another powerful option.

Special considerations in Greyhounds

Reference range studies
There have been several attempts to generate reference ranges for thyroid parameters in Greyhounds, sometimes directly and sometimes as part of larger studies of influences on Greyhound thyroid function. A summary of those available in the literature is included in table 3.

It can been seen from the summary table that the Greyhound reference range is considerably less than that of the general canine range for TT4 and that it extends to zero for FT4d. However, the ranges for TT3 and TSH are much more close to the general range. This combination of lower T4 but "normal" T3 suggests either a different T4:T3 production ratio by the Greyhound thyroid glands or a higher de-iodinase activity than in the general canine population.

Table 3: Mean, standard deviation and calculated reference range in several Greyhound thyroid studies

Total T4
(nmol/L)

   
13.9 +/- 6
2.1 - 37
Gaughan 2001
N = 98
18.1 +/- 8.1
2.9 - 34.7
Beale 1992
N = 206
7.2 +/- 2.5
2.3 - 12.1
Hill 2001
N = 9 (neutered & rested for 3 months)
15 - 67 MSU general canine range
13 - 52 CSLS/NWL UK general canine range
FT4d
(pmol/L)
   
11.3 +/- 6.2
0-23.5
Gaughan 2001
N = 98
0 +/- 23.5
0-17
Hill 2001
N = 9 (neutered & rested for 3 months)
6 - 42 MSU general canine range
7.0 - 40 CSLS/NWL UK general canine range
TT3
(nmol/L)
   
2.2 +/- 0.76
0.7 - 3.7
Beale 1992
N = 206
1.3 +/- 0.3
0.7 - 1.9
Hill 2001
N = 9 (neutered & rested for 3 months)
1.0 - 2.5 MSU general canine range
0.3 - 2.5 CSLS/NWL UK general canine range
TTSH
(ng/mL)
   
0.21 +/- 0.18
0 - 0.56Gaughan 2001N = 98
0.22 +/- 0.05
0.07 - 0.42Hill 2001N = 9 (neutered & rested for 3 months)
0 - 0.68 MSU general canine range
0 - 0.42 CSLS/NWL UK general canine range

Note: All above ranges calculated on mean +/- 1.96 SD which may not have been appropriate in all cases but was limited by the information presented in publications.

It should be noted that the lower limit of the Greyhound TT4 ranges reported above is below the capabilities of most commercial laboratories? TT4 measurement techniques. Most laboratories can report canine TT4 to 6 nmol/L and some to 4 nmol/L but none can accurately report values as low as 2 nmol/L with adequate precision. Practically speaking, if the data above are applicable to the current commercially available measurement techniques, this means that we cannot use TT4 or FT4d to confirm hypothyroidism in Greyhounds. Adjusting the references ranges to Greyhound specific ones will not help if we cannot reliably measure concentrations below the new reference range. Therefore, at present, TT4 and FT4d can only be reliably used to confirm euthyroidism. When TT4 or FT4d are detectable in Greyhound serum (>6 nmol/L, >2 pmol/L respectively) euthyroidism is likely. The converse of undetectable values supporting hypothyroidism does not hold true. This leaves total T3 and TSH as potential tests of choice for confirming hypothyroidism. Total T3 will still be subject to the effects of non-thyroidal illness (low-T3 state of medical illness) and will therefore still be subnormal in some animals that do not have hypothyroidism as is the case with TT4 in the general canine population.

The availability of the TSH response test is limited nowadays although an expensive recombinant human preparation is available (Sauve and Paradis 2000). When available the TSH response test is probably still the gold standard for assessing thyroid function although again it should be noted that normal Greyhounds do have a lower response than the general canine population (Gaughan and Bruyette 2001).

Unpublished work from Rob Shiel and Carmel Mooney (46 Greyhounds) at University College Dublin and Kent Refsal (49 sighthounds) and Markus Rick (100 Greyhounds) at Michigan State University are generally consistent with that in the literature. T4 ranges extend to zero or have a significant proportion below the assay limit of detection, but TT3 similar (although perhaps with a higher lower limit) to the general dog reference range. Further unpublished work on 100 Salukis yields similar conclusions.

Other Greyhound considerations
Several studies have attempted to address thyroid status with regard to what have been seen as specific Greyhound issues including the influence of oestrus suppression by testosterone, the effects of training and exercise, reproductive and racing performance and the condition known as Bald Thigh Syndrome.

Cowan (Cowan et al. 1997) failed to find differences in thyroid parameters between females treated with testosterone and males and Gaughan (Gaughan and Bruyette 2001) similarly found no differences in baseline thyroid parameters due to testosterone treatment . However, testosterone did appear to have some impact on the thyroid hormone response to TSH injection.

In a small number of dogs (n = 9) Hill (Hill et al. 2001) investigated the effects of exercise on thyroid parameters and found that haemoconcentration-adjusted concentrations of Total T4 increased dramatically but that TT3, FT4d and TSH did not immediately following a 500m race. After neutering and a 3 month non-racing period, TT4 concentrations increased slightly (from mean 2.7 ± sd 1.8 to 7.2 ± 2.5 nmol/L) but still not to the standard canine reference range. The other parameters did not change in this "detrained" period. In sled dogs exposed to long periods of exercise, thyroid hormone concentrations decrease by the end of a race (Panciera et al. 2003; Lee et al. 2004) and interesting T4 concentrations were below the general canine reference range in a significant proportion of trained pre-race animals (23-26%). Inconsistent thyroid hormone responses have been found following both short and long periods of exercise in man.

Bald thigh syndrome (BTS) is a common complaint among Greyhounds. In one post mortem study of 230 Kansas GH, 19% were found to be affected (Schoning and Cowan 1993). Bald thigh syndrome is a non-scarred alopecia principally affecting the caudal thighs but sometimes extending ventrally (even as far as the elbow) and sometimes with hyperpigmentation. Cowan (Cowan et al. 1997) investigated the possible associations with thyroid function of testosterone levels in 29 BTS and 28 normal Greyhounds and found none. There were no significant differences in TT4, TT3, FT4d, TSH or testosterone between the affected and unaffected animals.

Both reproductive and racing performance were investigated by Beale (Beale et al. 1992) using owner reported evaluations. She found no association between thyroid hormone concentrations and either poor reproductive or poor racing performance.

Much of the above would give the impression that hypothyroidism does not occur in the Greyhound. Of course this is incorrect and well documented cases of Greyhound hypothyroidism do exist but several pieces of evidence suggest that it is relatively uncommon compared to the prevalence in other breeds. Beale (Beale et al. 1992) noted that the prevalence of TgAA in Greyhound samples (3.6%) was much lower than in a general canine hospital population of the time (13.2%). Similarly, of 1,409 Greyhound samples at MSU 2% were TgAA positive compared with 10% of 143,800 canine samples in total. Of these 1,409, only 42 (3%) had thyroid results including elevated TSH consistent with hypothyroidism compared to 8% of general 143,800 canine samples. From a pathogenesis point of view, only 1/42 hypothyroid Greyhounds were TgAA positive, quite different from the overall 50:50 distribution of TgAA positive vs. negative hypothyroidism. So hypothyroidism does exist in the Greyhound, but compared to other breeds, it's relatively uncommon and probably is more often due to idiopathic follicular atrophy than immune mediated thyroiditis.

Summary
  • Thyroid diagnosis has progressed greatly in recent years with the availability of TSH, TgAA and improved availability of FT4d
  • In the general dog population there is still some difficultly in the laboratory confirmation of hypothyroidism
  • The use of exogenous thyroxine in euthyroid animals has unknown but possibly deleterious effects as it over-rides a normal physiological phenomenon
  • Greyhounds present a special difficulty due to an apparent lower reference range for TT4 and FT4d
  • Greyhound reference ranges for TT4 and FT4d probably extend below the limit of detection for currently available assays
  • Greyhound TT3 and TSH reference ranges are similar to the general canine population
  • There is no proven relationship between basal thyroid hormone concentrations and bald thigh syndrome, poor racing or reproductive performance.
  • Training may decrease baseline TT4 concentrations
  • Hypothyroidism does occur in Greyhounds but appears to uncommon compared to other breeds and may more commonly be due to idiopathic follicular atrophy than immune mediated thyroiditis

Conclusions
In conclusion, hypothyroidism in the Greyhound is probably much less common than is generally perceived. Given published data and current assay capabilities, there would be no such thing as a "Greyhound with a low-T4" if Greyhound specific reference ranges were used. Future work should focus on correlating other measures of thyroid function (e.g. T3 and TSH) with perceived clinical syndromes of hypothyroidism in the Greyhound in order to generate some diagnostic performance data (diagnostic sensitivity and specificity) and clarify which clinical presentations are associated with hypothyroidism and those that are clearly not.

References
Beale K, Bloomberg MS, Van Gilder J, Wolfson BB and Keisling K (1992) Correlation of racing and reproductive performance in Greyhounds with response to thyroid function testing. J Am Anim Hosp Assoc 28, 263-269.

Cowan LA, Refsal KR, Nachreiner R and Schoning P (1997) Thyroid hormone and testosterone concentrations in racing greyhounds with and without bald thigh syndrome. J Vet Intern Med 11, 142.

Dixon RM and Mooney CT (1999) Evaluation of serum free thyroxine and thyrotropin concentrations in the diagnosis of canine hypothyroidism. J Small Anim Pract 40, 72-8.

Gaughan KR and Bruyette DS (2001) Thyroid function testing in Greyhounds. Am J Vet Res 62, 1130-3.

Graham PA and Mooney CT (2005). Laboratory evaluation of hypothyroidism and hyperthyroidism. In BSAVA Manual of Canine and Feline Clinical Pathology. 2nd Ed, E. Villiers and L. Blackwood: 260-277.

Hill RC, Fox LE, Lewis DD, Beale K, Nachreiner R, Scott KC, Sundstrom DA, Jones GL and Butterwick R (2001) Effects of racing and training on serum thyroid hormone concentrations in racing Greyhounds. Am J Vet Res 62, 1969-1972.

Lee JA, Hinchcliff KW, Piercy RJ, Schmidt KE and Nelson S (2004) Effects of racing and nontraining on plasma thyroid hormone concentrations in sled dogs. J Am Vet Med Assoc 224, 226-231.

Panciera DL, Hinchcliff KW, Olson J and Constable PD (2003) Plasma thyroid hormone concentrations in dogs competing in a long-distance sled dog race. J Vet Intern Med 17, 593-596.

Sauve F and Paradis M (2000) Use of recombinant human thyroid-stimulating hormone for thyrotropin stimulation test in euthyroid dogs. Can Vet J 41, 215-9.

Schoning P and Cowan LA (1993) Gross and microscopic lesions of 230 Kansas greyhounds. J Vet Diagn Invest 5, 392-7.