More Than Just T₄

Abstract

Clinical challenges:

In older cats presenting with clinical features of hyperthyroidism, confirmation of a diagnosis of thyroid disease is usually straightforward. However, the potential for false-negative and false-positive results exists with all thyroid function tests (especially in the context of routine screening of asymptomatic cats) and leads to clinical dilemmas. For example, a high serum T₄ value may be found in a cat that lacks clinical signs of hyperthyroidism, or hyperthyroidism may be suspected in a cat with normal total T₄ concentrations.

Practical relevance:

To avoid unnecessary treatment and potentially adverse effects in a euthyroid cat, thyroid function tests must always be interpreted in the light of the cat’s history, clinical signs, physical examination findings and other laboratory findings.

Evidence base:

In this article the author reviews the use of commonly recommended thyroid function tests, focusing on clinical scenarios that present diagnostic difficulties. In doing so, he draws on the veterinary and comparative literature, his own clinical experience, and data, unpublished to date, obtained from a series of 100 hyperthyroid cats consecutively diagnosed at his clinic.

The perfect thyroid test remains elusive

The diagnosis of hyperthyroidism in cats is primarily based on a constellation of typical clinical features, including the cat’s signalment, history, clinical signs and physical examination findings (eg, palpable thyroid nodule).^1–3 Since many non-thyroidal diseases can mimic the signs of thyroid disease in cats, a complete database (eg, complete blood count, serum chemistry profile, urinalysis) must always be evaluated to help exclude other illness. After reviewing this data, the next step is to use thyroid function tests to confirm the diagnosis of hyperthyroidism.

Over the past two decades, many advances have been made in the availability of tests that can be used to diagnose feline hyperthyroidism. However, all of the commonly used tests have limitations, especially when evaluating cats suffering from non-thyroidal illness. Clearly we still do not have the perfect thyroid test that is capable of both confirming the diagnosis of thyroid disease in all affected cats and completely excluding it in cats that do not have thyroid disease.

This article reviews the common thyroid function tests currently recommended to diagnose feline hyperthyroidism. To that end, data obtained from a serum thyroid hormone panel (ie, serum total T₄, T₃, free T₄ and thyroid-stimulating hormone [TSH]) that was collected from 100 consecutively diagnosed hyperthyroid cats at the author’s clinic has been included to illustrate the utility of these individual serum thyroid tests. In addition, data is presented comparing two different T₄ assays commonly used by veterinary laboratories in order to demonstrate that all thyroid measurements cannot always be considered equivalent. Since definitive diagnosis of this common disorder is not always clear-cut, this review concentrates on protocols used in the work-up of ‘occult’ or problem cases, in which misdiagnosis is not uncommon.

Palpating the thyroid nodule

All cats with hyperthyroidism have a thyroid nodule affecting one or both lobes, since in all cases there is underlying thyroid pathology leading to the disease (eg, thyroid adenomatous hyperplasia, adenoma or carcinoma).^4,5 Therefore, a palpable cervical thyroid nodule (goiter) is present in most, if not all, cats with hyperthyroidism.

Thyroid gland palpation is an important non-invasive and inexpensive diagnostic tool and should be a standard part of the physical examination of all cats, but especially those older than 10 years of age, when hyperthyroidism becomes more prevalent. There are two general techniques used to palpate the thyroid gland in cats, both of which should be performed as routine.^1,2,6,7 In the author’s experience, it is not uncommon to miss a thyroid nodule with one technique but pick up a nodule with the second technique on the same examination.

With the classic palpation technique (Figure 1a), ⁶ the cat’s neck should be slightly extended with the head tilted backward. The thumb and index finger are gently passed over both sides of the trachea, starting at the laryngeal area and moving downwards toward the thoracic inlet (sternal manubrium). The fingertips should remain within the jugular furrows; it is important to be gentle because if the muscle is palpated too firmly, the nodule may be pressed into the muscle and difficult to feel. A thyroid nodule is usually recognised as a somewhat mobile, subcutaneous nodule or ‘blip’ that slides or slips under the fingertips. Since the thyroid lobes of the cat are loosely attached to the trachea, the enlarged lobe(s) frequently descend(s) ventrally from the normal location adjacent to the larynx. In hyperthyroid cats in which thyroid gland enlargement is not palpable, the possibility that the affected lobes have descended into the thoracic cavity should always be considered.

Figure 1.

Thyroid palpation techniques: (a) the ‘classic’ technique; (b,c) the second (Norsworthy) technique. In (b) the author is palpating for the right thyroid lobe and in (c) he is searching for the left thyroid lobe

With the second technique (sometimes referred to as the Norsworthy technique), ⁷ the clinician is positioned directly behind the cat (Figure 1b,c). The head of the cat is raised and turned 45° to the right or left, away from the side being assessed (ie, to palpate the right thyroid lobe, turn the cat’s head to the left). The tip of the clinician’s index or middle finger is placed in the groove formed by the trachea and sternothyroid muscle just below the larynx and then moved downwards in the groove to the thoracic inlet. If the thyroid lobe is enlarged, a characteristic ‘blip’ is felt as the finger passes over the goiter.

Unfortunately, the finding of a thyroid nodule on physical examination cannot always be equated with a hyperthyroid state, since smaller goiters are frequently detected in euthyroid cats without clinical and laboratory evidence of hyperthyroidism.^7–9 Put another way, the finding of a palpable thyroid nodule is a highly sensitive, but poorly specific diagnostic test. Therefore, the presence of a palpable goiter is, in itself, a poor indicator of hyperthyroidism. That said, the size of the enlarged thyroid gland can be predictive, with most larger nodules being found in cats with clinical hyperthyroidism.^7–9 Although some cats with small, palpable thyroid nodules may remain euthyroid for prolonged periods of time, many will eventually develop clinical and biochemical signs of hyperthyroidism, as the thyroid nodules continue to grow and begin to oversecrete thyroid hormone.

Because all hyperthyroid cats must have at least one (if not two and sometimes more) thyroid nodules in order to develop the disease, the finding of a thyroid nodule, either by cervical palpation or by imaging (eg, thyroid scintigraphy – see later), plays a crucial role in diagnosis in these cats. If the clinician cannot palpate a thyroid nodule, diagnosis becomes much more difficult to confirm, especially in cats with early, mild or borderline disease. As discussed below, none of the serum thyroid function tests are perfect and all can produce false-positive and false-negative results.

Measuring total T₄ concentration

In older cats presenting with clinical features of hyperthyroidism (eg, weight loss despite a good appetite), confirming the diagnosis is usually straightforward, as over 90% of hyperthyroid cats will have a serum total T₄ concentration that is clearly high.^{1–3,6,10–13} Because measurement of total T₄ is relatively cheap and readily available, this has become the screening test of choice for cats with hyperthyroidism.

Practitioners should be aware of the assay techniques being used by their commercial or in-house laboratories. Broadly speaking, serum T₄ can now be measured in five different ways (see box).

Although the correlation of serum T₄ concentrations provided by all of these assay methods is generally good (with the exception of the human T₄ kits, which are never recommended), any of these different assays can occasionally provide serum T₄ values that are falsely high or falsely low. No assay has 100% test sensitivity and specificity. Compared with RIA (the gold standard), CEIA has been shown to provide very similar test results,^14,15 whereas the in-house ELISA methods appear less precise or reliable. In one study of 50 cats, ¹⁷ serum T₄ concentrations obtained with an in-house ELISA method were compared with the results obtained with a validated RIA method. The ELISA was found consistently to overestimate the total T₄ concentration compared with the RIA. When results of the two methods were categorized (low, borderline low, normal, borderline high or high), they were discordant in up to 28 (56%) of the samples. Thus, reliance on the ELISA T₄ results might have led to inappropriate clinical decisions in half of the cases. Based on these findings, the authors concluded that the in-house ELISA kit was not accurate for determining serum total T₄ concentrations in cats. ¹⁷

Only one study comparing the results of T₄ determined by RIA and EIA has been reported in cats. ¹⁹ Although the correlation between the techniques was excellent, no clinical information was included about the thyroid status of any of the cats in that study.

To further investigate the usefulness of the EIA method for T₄ determination in the cat, the author’s clinic collected serum samples from 100 cats with a wide range of expected T₄ values – 60 with untreated hyperthyroidism and 40 that were being monitored after treatment with radioiodine. With each serum sample, T₄ was measured both by EIA and CEIA, and the results compared (Figure 2a). Regression analysis revealed that T₄ values obtained by each method were significantly correlated (R = 0.97; P <0.001). However, the EIA method tended to underestimate the T₄ value, which was reflected in the (flatter) slopes of the regression lines. This pattern was further supported by the finding that the mean and median T₄ concentration of all samples obtained from cats measured by EIA (mean ± SEM, 88.8 ± 7.7 nmol/l; median, 55 nmol/l) was significantly (P <0.0001) lower than that measured by CEIA (100.4 ± 9.0 nmol/l; median, 64.4 nmol/l).

Figure 2.

Scatter plots of total T₄ concentrations measured by CEIA and EIA in 100 cats (60 with untreated hyperthyroidism, 40 being monitored after radioiodine treatment). The cats ranged in age from 9–19 years (median 13 years); almost all were of mixed breeding (domestic shorthair or domestic longhair), with 47 castrated males and 53 spayed females. In each graph, each point represents the measurement of total T₄ concentration by the two assays, and the line represents the best fit of the data for linear regression analysis. (a) Scatter plots for all 100 cats. The correlation between the two methods was 0.97 for the 100 serum samples. (b) Scatter plots for 66 cats with values <100 nmol/l, as measured by CEIA. The correlation between the two methods was 0.93 for this group of 66 serum samples. Shaded areas indicate reference interval limits for the two T₄ methods. The cats with discordant T₄ results are shown in red

Importantly, when 66 cats with serum T₄ values <100 nmol/l were examined separately (Figure 2b), it became clear that discordant T₄ results did occur between the two methods, which could affect the clinical outcome in some cats. For instance, three of the treated cats that had subnormal T₄ values when measured by EIA were found to have normal T₄ concentrations when measured by CEIA; none of these cats were considered to be hypothyroid, and all had normal concentrations of both free T₄ and TSH. Even more importantly, 11 cats with untreated hyperthyroidism had normal T₄ values when measured by EIA but had clearly high T₄ concentrations when measured by CEIA; based on the EIA result alone, the diagnosis of hyperthyroidism might have been missed or additional testing required to confirm the diagnosis in these cats. Finally, although not common, one of the euthyroid cats tested had a falsely high T₄ value when measured by EIA (free T₄ was normal in this cat), whereas one hypothyroid cat showed falsely high, reference interval T₄ values when measured by EIA.

Overall, when results of the two methods were categorized (ie, low, normal or high range), serum T₄ values were discordant in 16 (24%) of the 66 cats. Therefore, although the correlation between T₄ values measured by these two methods is excellent, and most of the serum T₄ values measured by EIA agree with the cat’s clinical thyroid state, these results indicate that the EIA method provides a higher rate of false-negative and false-positive results in cats than does CEIA. Reliance on the EIA T₄ results could lead to inappropriate clinical decisions for some of the cats with borderline high or low T₄ values.

The bottom line is that no matter what assay method is used for T₄ measurement, there is the potential for false-negative and false-positive results. Therefore, serum T₄ results must always be interpreted in the light of the cat’s history, clinical signs and other laboratory findings. If a high serum T₄ value is found in a cat that lacks clinical signs of hyperthyroidism, especially if no thyroid nodule is palpated, there should never be any hesitation to repeat the serum T₄ test using a different technique, with RIA or CEIA being preferred in such cases.

Problem 1: Hyperthyroidism is suspected but the cat’s serum T₄ and T₃ concentrations are normal

Although serum total T₄ is preferable as a screening test for hyperthyroidism, many previously published reports indicate that approximately 10% of all hyperthyroid cats (and 40% of cats with early or mild hyperthyroidism) have serum T₄ within the reference interval limits.^3,12,13 Thus, finding a single reference interval value for T₄ does not preclude a diagnosis of hyperthyroidism.

In agreement with this, of the 100 hyperthyroid cats that were consecutively diagnosed at the author’s clinic, 92 had serum total T₄ values above the reference interval limits. The remaining eight cats with ‘normal’ T₄ values were all within the upper third of the reference interval (Figure 3a). As a proportion of the 23 cats in the series diagnosed with mild hyperthyroidism (<75 nmol/l), these eight cats with T₄ reference values account for 35% of cases.

Figure 3.

Box plots of serum concentrations of total T₄, total T₃, free T₄ and TSH in 100 cats with untreated hyperthyroidism. These hyperthyroid cats were consecutively diagnosed and ranged in age from 8–20 years (median 13 years); almost all were of mixed breeding (domestic shorthair or domestic longhair), with 45 castrated males and 55 spayed females. (a) Serum total T₄ concentrations, determined by CEIA. (b) Serum total T₃ concentrations, determined by CEIA. (c) Serum free T₄ concentrations, determined by equilibrium dialysis. (d) Serum TSH, determined by canine TSH CIEA. In each graph, the box represents the interquartile range (ie, 25th to 75th percentile range or the middle half of the data). The horizontal bar in the box represents the median value. For each box plot, the T-bars represent the main body of data, which in most instances is equal to the range. Outlying data points are represented by open circles. The shaded areas indicate the reference interval for each hormone. For the TSH assay, the detection limit (not indicated) is 0.03 ng/ml

As a diagnostic test for hyperthyroidism, serum total T₃ is not very helpful, at least as the sole screening test. It has previously been reported that a third of hyperthyroid cats have T₃ values that remain within reference interval limits (ie, not diagnostic for hyperthyroidism).^3,10,12,13 Results of the current series of 100 consecutively diagnosed hyperthyroid cats concur with this – 42 cats (42%) had normal serum T₃ concentrations as measured by CEIA (Immulite Total T₃; Seimens Healthcare) (Figure 3b).

For the purposes of cats with early or mild hyperthyroidism that have high–normal to only slightly high serum T₄ values (ie, the cats that are more difficult to diagnose), determination of serum T₃ becomes a useless diagnostic test. In support of that, of the 23 cats in the current series with mild hyperthyroidism (total T₄ concentration <75 nmol/l; Figure 2a), all maintained a serum T₃ value well within the reference interval.

Why might a hyperthyroid cat have normal T₄ and T₃ values?

In general, there are three possible reasons why a hyperthyroid cat could have normal T₄ and T₃ values and still show clinical signs of the disease.

• Early or subclinical hyperthyroidism In cats with hyperthyroidism that is still at an early or subclinical stage, either no or only very mild clinical signs would be expected. The thyroid nodule(s) in these cats should be small and may even be difficult to palpate. In early or mildly hyperthyroid cats (with no concurrent illnesses) serum total T₄ concentrations will eventually increase into the diagnostic thyrotoxic range upon retesting a few weeks to months later, as the thyroid tumor grows and secretes more T₄ into the circulation.

• Fluctuating T₄ concentrations In cats with mild hyperthyroidism, serum total T₄ concentrations can fluctuate in and out of the reference interval. ²⁰ Fluctuations in thyroid hormones occur in all hyperthyroid cats but are of little diagnostic significance in cats with severe hyperthyroidism and markedly elevated T₄ concentrations.

• Severe non-thyroidal illness Serum total T₄ concentrations can be suppressed to below the reference interval in euthyroid cats with severe non-thyroidal illness.^6,13,21–23 Similarly, marginally elevated serum total T₄ concentrations may be suppressed to the middle to high end of the reference interval in cats with mild hyperthyroidism and concurrent moderate to severe non-thyroidal disease. Once concurrent disease resolves, the serum total T₄ concentrations will again increase into the diagnostic thyrotoxic range.

If hyperthyroidism is suspected based on history, clinical signs (weight loss despite a good appetite) and physical examination findings (tachycardia, palpable thyroid nodule), but serum T₄ remains within the upper third of the reference interval (eg, >35–40 nmol/l), hyperthyroidism is still possible. There is no definitive approach to the diagnosis of hyperthyroidism in this scenario, but there are several options for further work-up, as described below.

Options for further work-up

Repeat total T₄ concentration

In cats in which overt, manageable underlying disease is identified, such concurrent disease should be addressed first, before proceeding with further thyroid testing. Once concurrent disease is resolved, most hyperthyroid cats will develop a clearly high T₄, confirming the diagnosis.^1–3

Conversely, in cats without overt underlying disease, simply repeating the serum T₄ concentration after 2 weeks may be diagnostic if the T₄ is fluctuating in and out of the reference interval.^6,20 In some cats with preclinical disease, it may take a number of weeks or even months for the serum total T₄ concentrations to increase into the ‘high’ range diagnostic for hyperthyroidism.^1–3,24 Again, the veterinarian should never hesitate to repeat the serum T₄ test using a different technique, with RIA or CEIA being preferred in such cats (see box, page 770).

Thyroid scintigraphy

Thyroid scintigraphy (thyroid imaging) is a nuclear medicine procedure that produces a visual display of functional thyroid tissue based on the selective uptake of radionuclides by thyroid tissue (Figure 4).^1,2,25–29

Figure 4.

(a) Thyroid scintigraphy in a euthyroid cat. In normal cats, the thyroid gland appears on thyroid scans as two well-defined, focal (ovoid) areas of radionuclide accumulation in the cranial to middle cervical region. The two thyroid lobes are symmetrical in size and shape and are located side by side. Activity in the normal thyroid closely approximates activity in the salivary glands, with an expected ‘brightness ratio’ of 1:1. (b,c) Thyroid scintigraphy in two cats with early hyperthyroidism. Both cats had borderline high serum concentrations of total T₄ with slightly high free T₄ values. The cat in (b) has a unilateral left thyroid adenoma, whereas the cat in (c) has bilateral adenomas. In both cases, the uptake of the radionuclide by the functional thyroid adenoma(s) is higher than the uptake by the cat’s salivary tissue. For both cats, a high thyroid:salivary ratio was calculated, which was diagnostic for hyperthyroidism

In normal cats, the thyroid gland appears on thyroid scans as two well-defined, focal (ovoid) areas of radionuclide accumulation in the cranial to middle cervical region. The two thyroid lobes are symmetrical in size and shape and are located side by side (Figure 4a); the thyroid and salivary glands would be expected to be equally bright (a 1:1 brightness ratio). In addition to visual inspection, the percentage thyroidal uptake of the radioactive tracer and/or the thyroid:salivary ratio can be calculated; both are strongly correlated with circulating thyroid hormone concentrations and provide an extremely sensitive means of diagnosing hyperthyroidism.^25,26

In cats with suspected hyperthyroidism in which serum T₄ is normal or borderline (see box, page 770), thyroid scintigraphy provides an extremely sensitive diagnostic test (Figure 4b,c). Likewise, thyroid imaging can be valuable for confirming or excluding a diagnosis of hyperthyroidism in cats found to have high–normal to high total T₄ or free T₄ concentrations on routine screening, especially if a thyroid nodule cannot be palpated (see box, page 772).

Because thyroid scintigraphy directly visualizes functional thyroid tissue and the ‘uptake’ of the radioisotope can be estimated by determining the thyroid:salivary ratio, thyroid imaging can diagnose hyperthyroidism before laboratory tests are consistently abnormal (Figure 4b,c). Thyroid scintigraphy can also exclude the diagnosis of hyperthyroidism in euthyroid cats that have false-positive elevations in their serum T₄ or free T₄ values. Therefore, not every cat with a high total T₄ or free T₄ value will be truly hyperthyroid, and treatment for hyperthyroidism would be contraindicated and may be harmful in those cases.

Unfortunately, because of the expense and the special licensing required to perform nuclear imaging, few veterinarians have access to the equipment needed to obtain thyroid images or determine thyroid uptake. If thyroid scintigraphy is available, however, this is definitely the ‘gold standard’ method for confirming (or excluding) a diagnosis of mild or occult hyperthyroidism in cats.

Free T₄ concentration

In cats with mild hyperthyroidism and normal total T₄ values, free T₄ concentrations can also aid in diagnosis. In support of that, serum free T₄ concentrations measured by equilibrium dialysis are more consistently elevated in hyperthyroid cats than are total T₄ concentrations.^13,30 Of the 100 hyperthyroid cats in the current series, only two had a free T₄ concentration within the reference interval when measured by equilibrium dialysis (Figure 3c).

Although free T₄ is more sensitive than total T₄ for diagnosing hyperthyroidism, the test specificity for free T₄ is poor, with up to 20% of sick (and some clinically normal) euthyroid cats having false-positive free T₄ results.^13,23,30,31 These sick cats with high free T₄ concentrations generally have total T₄ values in the lower half or below the reference interval.

Caution is, therefore, advised in using serum measurements of free T₄ by equilibrium dialysis as the sole diagnostic test for hyperthyroidism. As a thyroid function test, free T₄ should always be interpreted with a corresponding total T₄ measurement. A T₄ value within the upper third of the reference range (>35–40 nmol/l), combined with a high free T₄ concentration, is consistent with mild hyperthyroidism, whereas a low or low–normal T₄ with a high free T₄ is usually associated with non-thyroidal illness.^1–3,13 However, as always, it is important to combine these thyroid test results with the cat’s clinical features and the presence of a palpable thyroid nodule to make the correct diagnosis.

Complicating the free T₄ assay situation further, the current equilibrium dialysis test, which is the method utilized by commercial veterinary laboratories around the world, may not be exactly the same assay as was used in published studies that have evaluated free T₄ in cats with thyroid disease.^13,23,30 Nichols Diagnostics, the company that produced the original assay, sold the distribution rights for the assay (Free T₄ [ED] Test) to Antech Diagnostics a few years ago, and results of a recent study in cats ³¹ indicate a much higher prevalence of false-positive test results (33%) than reported over a decade ago.^13,23 The bottom line is that although the current free T₄ by dialysis assay still appears to be a highly sensitive test for diagnosis of hyperthyroidism (Figure 3c), the assay needs to be re-evaluated in a large number of normal and sick cats to verify its specificity and accuracy as a diagnostic test.

Finally, many commercial diagnostic laboratories are now using many other free T₄ assays, most of which do not use RIA or incorporate the equilibrium dialysis step that was once believed to be so important.^31,32 Only one of these assays, a free T₄ chemiluminescent immunoassay (Immulite Veterinary Free T₄; Siemens Healthcare), has been critically evaluated in hyperthyroid cats. ³¹ In that study, the assay showed excellent test performance in euthyroid and hyperthyroid cats, with a comparable or even better test accuracy than the free T₄ by dialysis method. A major flaw in that study, however, was that a group of cats with moderate to severe non-thyroidal illness was not included, so the prevalence of false-positive test results with this free T₄ chemiluminescent immunoassay method in sick cats that do not have hyperthyroidism is unknown. This is a critical issue; if it turns out that the prevalence of false-positive free T₄ results is lower with this chemiluminescent immunoassay, this would represent a major advantage over the free T₄ by dialysis method. Conversely, it may be that the rate of false-positive free T₄ results is as high or even higher than that reported with equilibrium dialysis.

Overall, it is unclear how much additional useful information is truly gained by present dialysis or chemiluminescent free T₄ assays over the use of total T₄ estimations alone. Many veterinarians mistakenly believe that the finding of high free T₄ in a cat is completely diagnostic for hyperthyroidism, especially when total T₄ is within the middle to upper half of the reference interval. However, since up to 30% of these cats may turn out to be euthyroid, ³¹ it is clear that the free T₄ test can never be considered a ‘gold standard’ diagnostic test for thyroid disease in cats. Use of free T₄ testing can lead to more confusion than clarity in some hyperthyroid cats; certainly, reliance on free T₄ results risks misdiagnosis of hyperthyroidism in many euthyroid cats, especially those with non-thyroidal illness.

Thyroid-stimulating hormone

In human patients, measurement of circulating TSH (thyrotropin) concentration is now firmly established as the first-line test for assessment of thyroid function in most clinical situations, including hyperthyroidism.^33–35 The finding of a very low serum TSH value is consistent with a diagnosis of hyperthyroidism; if the serum TSH concentration is within the reference interval, then a diagnosis of hyperthyroidism is effectively ruled out. In overt hyperthyroidism, high serum levels of total or free T₄ nearly always accompany a suppressed TSH value, which confirms the diagnosis. Less commonly, patients have subclinical hyperthyroidism, defined as a low serum TSH concentration with normal serum thyroid hormone concentrations. This pattern of biochemistry may reflect mild thyroid hormone excess but may also reflect hypothalamic or pituitary disease, non-thyroidal illness or ingestion of drugs that inhibit TSH secretion. The prevalence of subclinical hyperthyroidism is 1–2% in the general population, but higher in people older than 60 years of age, and in women.

In cats, a feline-specific TSH assay has not yet been developed for commercial use. However, assays for measuring canine TSH (cTSH) are widely available (Immulite Canine TSH; Siemens Healthcare), and it has been suggested that TSH measurement may also provide diagnostic information in cats with suspected hyperthyroidism.^30,36–40

Theoretically, as in people, it could be expected that serum TSH levels would be low in early stages of hyperthyroidism, even before T₄ is clearly elevated.^33–35 A recent prospective study provided evidence that an undetectable TSH concentration in euthyroid cats is associated with an increased risk of the subsequent diagnosis of hyperthyroidism. ³⁸ However, it is important to note that almost one-third of all cats had undetectable TSH values at baseline but only half of those cats with low TSH values went on to develop hyperthyroidism. Therefore, TSH measurements are not a good tool, at least in the author’s opinion, for predicting which cats will develop hyperthyroidism in the future.

But can a serum TSH measurement be used as a diagnostic test for hyperthyroid cats? It has been reported that hyperthyroid cats will almost always have a low TSH value, at or below the limit of detection of the assay (0.03 ng/ml).^{30,36–38,40} However, in one study that included 40 euthyroid cats (20 healthy cats and 20 cats with chronic kidney disease), five (12.5%) of the cats also had undetectable levels of TSH, indistinguishable from the values in the hyperthyroid cats. ³⁰ Similarly, in a recent larger study of 187 euthyroid cats, 34 (18%) were found to have undetectable levels of TSH. ⁴⁰ Therefore, the main downside of the current cTSH assay lies in its inability to reliably differentiate euthyroid cats from those with hyperthyroidism. This is related to the assay’s suboptimal sensitivity when used to measure feline TSH. In other words, the current cTSH assay cannot accurately measure low TSH values in cats (ie, cannot distinguish a low–normal TSH concentration from an undetectable value).

In the current series of 100 consecutively diagnosed hyperthyroid cats, 96 had undetectable serum concentrations of TSH (<0.03 ng/ml), whereas four had detectable TSH concentrations that remained within the reference interval (Figure 3d). Although it might be anticipated that cats with mild hyperthyroidism would be more likely to have detectable serum TSH concentrations, that does not always appear to be the case. All eight of the cats with reference interval total T₄ values had low or undetectable TSH values, consistent with a mild hyperthyroid state suppressing pituitary TSH secretion. More surprisingly, of the four hyperthyroid cats that had detectable TSH values (Figure 3d), two had mild hyperthyroidism (serum T₄, 55 and 75 nmol/l) and two had moderate to severe disease (serum T₄, 115 and 230 nmol/l).

The reason why some hyperthyroid cats, especially those with more advanced disease, fail to completely suppress circulating TSH concentrations is not clear, but assay artefact or laboratory error are possible explanations. Clearly, a better TSH assay is needed to help diagnose cats with mild hyperthyroidism – a feline-specific TSH assay that has adequate sensitivity to reliably distinguish a normal value from a low one. For the meantime, caution is advised to avoid overinterpreting values in cats, since it can be so difficult to distinguish low and low–normal values from the suppressed values expected in cats with hyperthyroidism.

The best use for currently available TSH measurements, it might seem, would be to help exclude hyperthyroidism (ie, by finding a mid- to high–normal value rather than a suppressed value). However, it is also clear that the finding of a detectable serum TSH concentration (>0.03 ng/ml) cannot completely exclude that diagnosis. As a diagnostic test for feline hyperthyroidism, serum TSH should never be run alone, but should be measured in tandem with T₄ (and free T₄) concentrations.

Dynamic thyroid function testing

In the majority of hyperthyroid cats with normal total T₄ concentrations, identification of a thyroid nodule or concurrent disease, repeat total T₄ analysis (by RIA or CEIA) or simultaneous measurement of free T₄ allows confirmation of the diagnosis (see box, page 770). Further diagnostic tests are rarely required.

Dynamic thyroid function tests (T₃ suppression and thyrotropin-releasing hormone [TRH] stimulation tests) have been recommended in the past as being helpful in confirming a diagnosis of hyperthyroidism (Table 1).^{1–3,6,41,42} Nowadays, use of these tests is only considered when repeated total T₄ concentration remains within the reference interval, when the free T₄ concentration is equivocal, when a thyroid nodule cannot be palpated or when thyroid scintigraphy is unavailable.

Table 1.

Commonly used protocols for dynamic thyroid function tests to diagnose cats with mild hyperthyroidism

	T3 suppression test	TRH stimulation test
Drug	Liothyronine	TRH
Dose	20 µg q8h for seven doses	0.1 mg/kg
Route	Oral	Intravenous
Sampling times	0 and 2–4 h after last dose	0 and 4 h
Assay	Total T4	Total T4
Interpretation*
Euthyroidism	<20 nmol/l with >50% suppression	>60% increase
Hyperthyroidism	>20 nmol/l ± <35% suppression	<50% increase

TRH = thyrotropin-releasing hormone, ND = not determined

^*Values quoted are guidelines only. Each individual laboratory should furnish its own reference interval

Problem 2: Palpable thyroid gland but cat is not hyperthyroid

Non-functional enlargement of the thyroid glands (goiter) has been recognized since the 1960s but has taken on new significance since functional hyperthyroidism arose as an entity in the late 1970s. Non-functional goiter was ‘re-recognized’ about a decade ago.^24,43,44

Many believe that clinical hyperthyroidism has a prodromal period (also referred to as subclinical hyperthyroidism or pre-hyperthyroidism).^38,43,45,46 However, it is not clear whether all goiters are indicative that the cat will develop hyperthyroidism, and some cats will remain healthy and euthyroid for years.

Surgical removal of non-functional goiters has been proposed as a preventive measure, ⁴⁴ but there is no evidence to support this approach since almost all of these tumors are benign. In general, careful observation and monitoring of serum T₄ concentration and the size of the thyroid nodule is recommended, with appropriate treatment instituted when needed.

Problem 3: Cat is misdiagnosed as hyperthyroid based on falsely high serum total T₄ (or free T₄) concentration

A high circulating total T₄ concentration is the biochemical hallmark of hyperthyroidism and is extremely specific for its diagnosis.^1–3,12,13 False-positive results (ie, a high T₄ in a cat without hyperthyroidism) are relatively rare but are being seen with increasing frequency, especially with the automated T₄ immunoassays and in-house testing kits. In agreement with results from the series of 100 cats (Figure 2), one investigator reported that 1–2% of all cats diagnosed as hyperthyroid based on the finding of a high serum T₄ concentration turn out to be euthyroid based on results of thyroid scintigraphy (MR Broome, 2013, personal communication).

More important, however, is the situation in cats that maintain reference interval serum T₄ values and are diagnosed as hyperthyroid on the basis of a high serum free T₄ concentration. Up to 30% of these cats presenting with a mid- to high–normal total T₄ value, together with high free T₄ concentrations by dialysis, will turn out to be euthyroid based on results of thyroid scintigraphy. ³¹

The reason for the higher prevalence of false-positive results may be related to the increasing use of T₄ screening as part of the cat’s annual ‘wellness’ program. When serum T₄ is determined in large numbers of healthy, asymptomatic cats to screen for hyperthyroidism, a percentage of those euthyroid cats can be expected to have high ‘outlying’ T₄ values that could be misinterpreted as hyperthyroidism.

What is the probability of thyroid disease given a positive test result?

For a cat in the clinic that returns a positive result for a particular test (in this case, a high T₄ or free T₄ concentration), the question that you and your client would want an answer to is ‘what is the chance (probability) of thyroid disease given the positive test?’

What is often overlooked is that, in clinical practice, the usefulness of a test result for an individual patient depends primarily on the prevalence of the disease in the population being tested.^47,48 The diagnostic value of a test will be much improved if the test is limited to those cats that are likely to have the disease in question, based on patient history and clinical assessment. A positive or a negative result is then more likely to be meaningful, compared with a test that is indiscriminately applied to all patients, even those without any clinical features or evidence of the disease. A diagnostic test should be used to supplement rather than as a substitute for good clinical judgment.

What next?

So what next if a high serum T₄ value is found in an asymptomatic cat in which no goiter is palpable?

The first step should always be to repeat the cervical palpation, looking for a thyroid nodule and to verify the high T₄ concentration (see box, page 772). Once again, the veterinarian should never hesitate to repeat the serum T₄ test using a different technique, with RIA or CEIA preferred in such cats. If there is any doubt about the diagnosis, thyroid scintigraphy should be considered (Figure 4).

In many of these asymptomatic cats in which the diagnosis is unclear (eg, borderline T₄ concentration, no obvious thyroid nodule), the best approach is to proceed cautiously using close observation, rather than start any treatment. The cat should be rechecked at 2–3 month intervals, at each visit monitoring the cat’s body weight, carefully re-palpating for a thyroid nodule and repeating the serum T₄ (and possibly free T₄) concentration. If the cat is truly hyperthyroid, the thyroid tumor will continue to grow, the serum T₄ will eventually rise into the hyperthyroid range and the cat will lose weight.

Ultimately, hyperthyroidism is a clinical diagnosis and should never be based on a total T₄, free T₄ or TSH concentration alone. These laboratory results must be combined with the cat’s signalment, history, clinical features and the presence of a thyroid nodule to make the diagnosis.

Key Points

• Although hyperthyroidism is relatively easy to diagnose in classically presenting cats, the increased frequency of testing cats with early, mild disease, as well as the routine screening of large numbers of asymptomatic cats, has had significant implications for the diagnostic performance of many of the routine tests currently used.

• Therefore, all thyroid function test results must always be interpreted in the light of the cat’s history, clinical signs and other laboratory findings.

• If a high serum T₄ value is found in a cat that lacks clinical signs of hyperthyroidism, especially if no thyroid nodule is palpated, the serum T₄ test should be repeated or other testing undertaken to verify the diagnosis.

• None of the treatments used for hyperthyroidism are benign. Therefore, it is of paramount importance that a misdiagnosis of hyperthyroidism is avoided in a euthyroid cat to prevent unneeded and potentially adverse treatment effects.