Thyroid enlargement and its relationship to clinicopathological parameters and T₄ status in suspected hyperthyroid cats

Abstract

To relate thyroid size to routine blood parameters and T₄ status the ventral neck of 161 cats with clinical signs consistent with hyperthyroidism was examined by two independent observers using a semi-quantitative palpation system. Thyroid gland size of each side was scored from 0 (non-palpable) to a maximum of 6 (>25 mm). In 127 of the 161 cats, at least one thyroid gland was palpable. The palpation score was significantly correlated with the T₄ concentration. The 17 hyperthyroid cats had significantly higher palpation scores than the 110 euthyroid cats. Euthyroid animals with a palpation score ≥3 were significantly older, had higher body weights, lower alkaline phosphatase, alanine aminotransferase, phosphate, and urine specific gravity, but higher lipase and creatinine concentrations than hyperthyroid cats. Our study demonstrates that although no reliable conclusion on the functional status of the thyroid can be drawn based on its size the likelihood of hyperthyroidism increases with increasing size of the gland.

Hyperthyroidism is the most common endocrine disorder affecting older cats. Clinical signs, such as weight loss, polyphagia, vomiting, diarrhoea, polyuria and polydipsia lead to an initial suspicion. ^1–3 The tentative diagnosis is supported by a palpable thyroid gland and typical routine clinicopathological changes, such as increased liver enzymes, and decreased creatinine and fructosamine concentrations. ^1–3

Feline hyperthyroidism is mainly caused by benign nodular hyperplasia, adenomatous hyperplasia or a functional thyroid adenoma, involving one or both sides of the thyroid gland and inevitably leading to an increase in size of the gland. ⁴ Therefore, it is not surprising that a thorough clinical examination performed by an experienced clinician will reveal an enlargement of the thyroid gland on palpation in most affected cats. Previous studies reported a palpable thyroid gland in more than 90% of hyperthyroid cats. ^1,2 Nonetheless, there is increasing evidence that the presence of a goitre alone is a poor indicator of symptomatic hyperthyroidism. ^5–8 In a study by Chaitman and coworkers, ⁵ 38% of 26 euthyroid cats had a palpable thyroid gland, which could be confirmed by histopathology as adenomatous hyperplasia. While from this study no data on the size of the thyroid gland were available, Norsworthy and coworkers ⁸ proposed a semi-quantitative scoring system and reported a substantial enlargement in nearly 60% of clinically healthy and diseased euthyroid cats, although the size was still scored below the range of hyperthyroid cats.

The objective of the present study was to prospectively relate thyroid size to serum T₄ concentrations in a group of cats with suspected hyperthyroidism. We used a semi-quantitative scoring system initially proposed by Norsworthy and coworkers ⁸ and had two clinicians independently perform the palpation. In addition, we evaluated clinicopathological parameters and compared euthyroid cats with either a large or small score with hyperthyroid cats.

Materials and Methods

Animals

The study comprised 161 cats, 86 males (five intact) and 75 females (four intact), with a median body weight of 4.1 kg (range: 1.5–9.3) and a median age of 12 years (range: 2–23). They had been presented to the Clinic for Small Animal Internal Medicine of the University of Zurich between January 2003 and August 2005 for investigation of clinical signs consistent with hyperthyroidism (eg, polyuria, polydipsia, weight loss, polyphagia, vomiting, diarrhoea) or cats with a palpable thyroid gland. Ethylene Diamine Tetraacetic Acid (EDTA) anticoagulated blood and serum samples were drawn from the cephalic vein; urine was taken aseptically by cystocentesis. The study was conducted in compliance with institutional guidelines for research on animals and informed consent from pet owners was obtained.

Thyroid palpation technique and thyroid gland size

All thyroid palpations were undertaken independently by two clinicians experienced in thyroid palpation. Each cat was palpated with the classical palpation technique ⁹ and results were recorded on a prepared examination sheet (Fig 1). The thyroid gland size of each side was scored from 0 (non-palpable) to a maximum of 6 (nodule >25 mm), with score 1=1–3 mm, score 2=3–5 mm, score 3=5–8 mm, score 4=8–12 mm, score 5=12–25 mm. Allocation into scores was performed according to the largest detectable nodule. In addition, the total score sum of all palpable lobes for each cat was calculated. In cases in which there was a discrepancy between the two observers, mean score values of the largest lobe and mean score sum values were calculated, which resulted in score categories of 0.5 intervals.

Fig 1.

Thyroid gland palpation examination sheet: for each cat, results were recorded by two independent clinicians on a separate sheet.

Haematology, serum biochemistry and urine analysis, and T₄ concentrations

Complete haemograms were performed using a Cell Dyn 3500 (Abbott, Baar, Switzerland) including packed cell volume (PCV), haemoglobin concentration, red blood cell (RBC) counts, mean corpuscular haemoglobin, mean corpuscular haemoglobin concentration, mean corpuscular volume and white blood cell (WBC) counts. Blood smears were made from fresh EDTA anticoagulated blood and Giemsa stained using an AMES Hema Tek slide stainer (Bayer, Zürich, Switzerland). They were evaluated for WBC differentials and RBC morphological characteristics. Platelets were counted manually if requested by the attending veterinarian as they could not reliably be counted using the Cell Dyn 3500. Serum biochemistry was performed using a Cobas Integra 700 (Roche Diagnostics, Rotkreuz, Switzerland) by standard procedures recommended by the International Federation of Clinical Chemistry. Serum biochemistry included blood bilirubin, glucose, blood urea nitrogen (BUN), creatinine, protein, albumin, cholesterol, alkaline phosphatase (ALP), aspartate aminotransferase (AST), alanine aminotransferase (ALT), lipase and electrolytes (sodium, chloride, potassium and phosphate). Urine specific gravity was measured using a refractometer. Reference ranges are stated as 5% and 95% quantiles and were determined from 58 healthy adult cats using identical methods. T₄ was measured by use of a commercially available radioimmunoassay (Coat-A-Count, Diagnostic Products Corp, Los Angeles, CA, USA) validated for use in cats. Hyperthyroidism was confirmed if T₄ ≥3.5 μg/dl.

Statistical analyses

Data were analysed using non-parametric statistical methods (SPSS, Statistical Package for the Social Science, Software Packets for Windows, Version 14.0 and GraphPad PRISM for Windows, Version 4.0).

Correlations were calculated using the Spearman rank correlation test. The Mann–Whitney U-test was used to determine differences between groups. χ² and Fisher's exact tests were used for categorical variables. Values of P<0.05 were considered significant.

Results

In 127 out of 161 (78.9%) cats at least one thyroid gland with a score between 0.5 and 4 was palpable, and 55 of them had bilateral involvement determined by at least one clinician.

Correlation between score, total score sum and T₄

Evaluating all 127 cats with a palpable thyroid gland revealed a statistically significant although weak positive correlation between the mean score of the largest lobe and the T₄ concentration (P=0.006, R=0.244; Fig 2) as well as between the mean total score sum and the T₄ concentration (P=0.007, R=0.238).

Fig 2.

Correlation of the T4 concentration with the mean score of the largest lobe. R=0.244, P=0.006. The grey shaded area represents the range of the T4 concentration considered normal in a cat

Evaluation of the hyperthyroid cats

Based on the results of the T₄ concentrations, 17 cats with a median age of 14 years (range: 10–17) and a median body weight of 3.5 kg (range: 1.9–7.2) were considered hyperthyroid with a median T₄ concentration of 7.9 μg/dl (range 3.6–14.8). All of the hyperthyroid cats had a palpable thyroid gland, 10 of them with bilateral involvement. The median score of the largest lobe and the median of the total score sum were 3.5 (range: 0.5–4) and 4 (range: 1–7), respectively. Nine of the hyperthyroid cats had a palpation score of the largest lobe >3.

Evaluation of the euthyroid cats

In 144 cats T₄ concentrations were between 0.1 and 3.0 μg/dl (median 1.6 μg/dl). Two of the animals had a T₄ of 0.1 and 0.2 μg/dl, respectively, while nine had a T₄ concentration between 2.3 and 3.0 μg/dl.

They had a median age of 12 years (range: 2–23), and a median body weight of 4.2 kg (range: 1.5–9.3).

Among the 144 euthyroid cats, 110 (76.4%) had a palpable nodule, 45 of them with bilateral involvement. The median score of the largest lobe and the median of the total score sum were 1 (range: 0.5–4) and 1.5 (range: 0.5–6), respectively. Eight cats had been included only because they had a palpable lesion without relevant clinical signs.

Three of the euthyroid cats had a palpation score of the largest lobe >3. In one cat with a palpation score of 4, a parathyroid adenoma was diagnosed. In the second, with a palpation score of 3.5, essential hypertension (systolic blood pressure 240 mmHg) with secondary retinopathy was diagnosed and treated with amlodipine. Repeated palpation 6 and 12 months later revealed a score of only 1. Hypertension was well controlled and repeated T₄ concentrations were still in the normal range (1.6 and 1.8 μg/dl, respectively). The third cat with a palpation score of 3.5 and a T₄ concentration of 2.7 μg/dl, presented in poor general condition with pleural effusion. Decompensated hypertrophic cardiomyopathy was diagnosed and the cat was euthanased 2 days later by a private veterinarian without necropsy being performed. Six of the euthyroid cats had a palpation score of the largest lobe of 3, however, limited or no follow-up information was available and hyperthyroidism could neither be confirmed nor excluded.

Among the euthyroid cats with a palpable thyroid gland, cats with a score <3 (further referred to as group 1) were significantly younger (median age 11 years; range: 2–23) than cats with a score ≥3 (further referred to as group 2; median age 16 years; range: 10–20) (P=0.001). There were no differences in body weight and T₄ concentration between the two groups. Interestingly, cats of group 2 had significantly lower specific gravities of urine (P=0.008) and significantly higher serum lipase concentration (P=0.011) than cats of group 1. There was a trend toward higher AST concentrations in cats of group 2 compared to cats of group 1 (P=0.06). When euthyroid cats without a palpable gland were compared to those with a palpable gland, no significant differences were found for any of the parameters tested.

Among the 110 euthyroid cats with a palpable lesion, 39 had a creatinine concentration above the reference range, 32 of them had a concurrent urine specific gravity below 1035. Three azotaemic cats tested as statistical outliers with creatinine concentrations of 711, 849 and 960 μmol/l. Azotaemic cats were significantly older than non-azotaemic euthyroid cats (P=0.004) but there was no difference in T₄ concentrations (P=0.59). However, in five of the azotaemic animals (one cat without palpable gland), T₄ concentrations were between 2.3 and 2.7 μg/dl; in two of them mild chronic kidney disease had been diagnosed before hyperthyroidism was suspected. Excluding azotaemic cats from statistical analyses revealed that cats of group 1 were significantly younger compared to animals of group 2 (P=0.004) with no differences in body weight and T₄ concentrations. There was a trend toward lower urine specific gravities (P=0.064) and higher serum lipase concentrations (P=0.072) in cats of group 2.

Comparison of hyperthyroid with euthyroid cats

Hyperthyroid cats were significantly older (P=0.044) and had a significantly lower body weight (P=0.018) compared to the euthyroid cats. There was no difference in sex between the two groups. The median score of the largest lobe and the median of the total score sum were significantly higher in the hyperthyroid compared to all euthyroid cats (P<0.001) and compared to euthyroid cats with a palpable gland (P<0.001).

Increasing lobe size was not associated with increased frequency of bilateral involvement either in the hyperthyroid cats or in the euthyroid cats with a palpable node.

PCV, bilirubin, ALP, ALT, AST, phosphate concentrations and urine specific gravities were significantly higher, while creatinine, fructosamine and total protein concentrations were significantly lower in hyperthyroid compared to all euthyroid cats and compared to euthyroid cats with a palpable thyroid gland (Table 1).

Table 1.

Selected blood parameters of hyperthyroid cats and euthyroid cats with a palpable thyroid gland

Variable	Hyperthyroid cats		Euthyroid cats (score >0)		Reference range	P
	Median	Range	Median	Range
Packed cell volume (%)	37	27–50	34	13–48	33–45	0.012
Protein (g/l)	62.0	54.0–75.0	70.5	29–92	64.0–80.0	0.007
Bilirubin (μmol/l)	4.4	2.0–13.0	2.7	0.7–51.7	1.7–7.2	0.004
ALP (U/l)	105	40–256	34	1–165	16–43	<0.001
ALT (U/l)	213	74–502	64	21–442	34–98	<0.001
AST (U/l)	72.5	24–173	36.5	15–372	19–44	0.003
Creatinine (μmol/l)	101	47–388	142	51–960	98.0–163	<0.001
Phosphate (mmol/l)	1.5	1.1–4.5	1.3	0.5–4.6	0.9–1.8	0.041
Fructosamine (μmol/l)	267	234–341	303	164–742	202–340	0.020
Urine specific gravity	1046	1018–1050	1030	1006–1050		0.034

Interestingly, among all cats with a palpation score ≥3, euthyroid cats were significantly older (P=0.028) and had a higher body weight than hyperthyroid cats (P=0.036). Moreover, euthyroid cats had lower ALP, ALT, phosphate concentrations and urine specific gravities but higher serum creatinine and lipase concentrations compared to the hyperthyroid cats (Table 2).

Table 2.

Selected blood parameters of cats with a palpation score ≤3 (largest lobe): hyperthyroid compared to euthyroid cats

Variable	Hyperthyroid cats (palpation score ≥3)		Euthyroid cats (palpation score ≥3)		Reference range	P
	Median	Range	Median	Range
ALP (U/l)	105	42–211	34	14–68	16–43	<0.001
ALT (U/l)	213	74–455	68.5	22–420	34–98	0.05
Creatinine (μmol/l)	101	47–388	145	115–370	98.0–163	0.001
Phosphate (mmol/l)	1.84	1.1–4.54	1.22	0.71–2.02	0.9–1.8	0.009
Lipase (U/l)	17.5	11–23	40.5	20–131	8.0–26	<0.001
Urine specific gravity	1034	1018–1050	1015	1007–1048		0.009

Exclusion of azotaemic cats from statistical analyses revealed that ALP, ALT, AST, phosphate concentrations were significantly higher, while creatinine and fructosamine concentrations were lower in hyperthyroid compared to all euthyroid cats and compared to euthyroid cats with a palpable thyroid gland. Comparing all non-azotaemic cats with a palpation score ≥3 revealed significantly higher ALP and phosphate concentrations, but lower serum creatinine and lipase concentrations in hyperthyroid compared to euthyroid cats.

Discussion

In the present study we investigated the relationship between thyroid gland size and serum T₄ concentration as well as haematological, serum biochemical parameters and urine specific gravity in cats with clinical signs consistent with hyperthyroidism. In line with the results reported by Norsworthy and coworkers, ⁸ we observed higher T₄ concentrations as the maximal lobe size increased and as the total score sum increased. Calculation of the total score sum did not offer an advantage compared to the determination of the largest lobe size because the correlation between T₄ and mean score of the largest lobe and that of T₄ and total score sum did not differ. Therefore, it can be concluded that the larger a palpable thyroid gland is in a cat with clinical signs consistent with hyperthyroidism, the higher the likelihood that the cat is truly hyperthyroid. However, 76.4% of the euthyroid cats of our study also had a palpable thyroid gland; this percentage surpasses the 38% and 59% reported previously. ^5,8 Remarkably, we also observed cats with a palpation score >3 and a T₄ concentration in the normal range. Both observations might be due to the population under investigation: we evaluated cats in which hyperthyroidism was clinically suspected. In addition, a palpable thyroid gland was one of our inclusion criteria. Thus, a high prevalence of thyroid lesions was expected and the examiners might have performed a more thorough palpation.

In our study, having two experienced clinicians perform a thorough thyroid palpation, revealed a high percentage of small palpable glands in euthyroid cats. Therefore, unlike previously assumed, a palpable thyroid gland is not rare and does not necessarily have to be associated with hyperthyroidism. Interestingly, euthyroid cats with a palpation score ≥3 were older than euthyroid cats with a score <3 as well as older than the hyperthyroid cats. This may indicate that non-functional thyroid changes are more common in older animals, a phenomenon that has been shown for humans, where the prevalence of thyroid nodules increases with the advanced age of the patient. ^10–12 However, whether a silent or non-functional goitre is truly more common in older cats cannot be concluded from our data as we only evaluated cats in which hyperthyroidism was suspected. We did not assess healthy age-matched cats, which would be an interesting population to screen for the presence of asymptomatic goitre.

Remarkably, in one of our cats with a score of 3.5, the lobe size decreased after 6 months. A palpation error would theoretically be possible, however, because palpation was performed independently by two clinicians this is an unlikely explanation. Cystic thyroid and parathyroid lesions have not only been reported in humans but also in feline patients. ^7,13,14 Therefore, a cyst without hormonal activity can be suspected in our case. Primary hyperparathyroidism was diagnosed in a cat with a palpation score of 4, a disease of feline patients in which a large palpable mass lesion in the ventral cervical region has frequently been described. ^15–17 Hence, based solely on our palpation results other differential diagnoses like cystic lesions, parathyroid adenoma, salivary mucocele or abscess cannot be excluded. Unfortunately, only limited or no follow-up information was available from our other euthyroid cats with palpable thyroid gland, which would have been interesting.

One cat with a palpation score of 3.5 had a T₄ concentration in the upper end of the reference range. It had been presented in a poor general condition, therefore, suppression of the T₄ concentration due to concurrent non-thyroidal illness was suspected in this animal. Effect of non-thyroidal illness on serum T₄ concentration is a well known phenomenon not only in cats. Low or undetectable levels of T₄ would be expected in euthyroid (normal) animals with severe non-thyroidal illness, as we saw in two of our cats. High-normal values may be encountered in hyperthyroid cats that are evaluated at the time they are presented with concurrent illness. ^18,19 These cases then can display a diagnostic dilemma. Nine of our animals had a T₄ concentration in the upper third of the reference range. None of them had been presented in a bad general condition or had severe disease; however, five of them had a mild azotaemia and in two of them chronic kidney disease had been diagnosed before hyperthyroidism was suspected. A recent study by Wakeling and coworkers ²⁰ evaluated T₄ concentrations in cats with mild chronic kidney disease: although still within the normal range, animals that had additional clinical signs compatible with hyperthyroidism had significantly higher T₄ values compared to animals with chronic kidney disease only. Diagnosis of hyperthyroidism was confirmed at a later time point by repeated T₄ measurement or by T₃ suppression test. The authors assume that the azotaemic cats with clinical signs of hyperthyroidism but a within-reference T₄ concentration were indeed hyperthyroid at the time of initial clinical suspicion. They suggest that a T₄ concentration of 30 nmol/l (corresponding to 2.3 μg/dl) obtained in a cat with chronic kidney disease is highly suspicious for hyperthyroidism. Considering these results and taking into account that two of our five azotaemic cats had a palpation score of 3, hyperthyroidism is likely. However, unfortunately no other tests have been performed nor are follow-up information available from these animals to definitely confirm the diagnosis. In the four non-azotaemic cats with a T₄ concentration ≥2.3 μg/dl it is possible that some of them were in an early state of hyperthyroidism in which the T₄ was still normal or that they had a normal fluctuation of the T₄ concentration over time. The latter has been shown in cats with mild hyperthyroidism (ie, minimal clinical signs and T₄ values slightly above the reference range), which may periodically have T₄ values in the reference range. ²¹ The use of more stringent cut-off points in elderly cats has been proposed by some authors ⁶ to identify those ‘euthyroid’ cats as being biochemically hyperthyroid.

Excluding azotaemic cats from statistical analysis, euthyroid cats still had significantly higher creatinine and lipase concentrations compared to hyperthyroid cats. This could potentially be attributed to an increased glomerular filtration rate observed during the hyperthyroid state ²² leading to an increased creatinine and lipase excretion. Moreover, body weights of the hyperthyroid animals were significantly lower compared to euthyroid cats. Also muscle mass can have an influence on creatinine levels. ²³ Lower creatinine concentrations in hyperthyroid cats have also been shown in the study by Wakeling and coworkers. ²⁰ The authors documented a decline in the creatinine concentrations that occurred during the onset of hyperthyroidism in 14 of 16 of their cats.

Increased serum ALT-, AST-, LDH- and ALP-concentrations have been reported in >90% of hyperthyroid cats. ^2,3 In one study, a correlation between T₄ and ALP concentrations was found. ²⁴ The lower concentrations of these enzymes in our cats with a large palpable thyroid gland but a normal T₄ concentration compared to the hyperthyroid cats further supports the euthyroid status of these animals.

Of course it is abundantly clear that hyperthyroidism cannot be completely excluded based on a normal T₄ concentration. ²⁵ The development and application of a more sensitive feline TSH assay, analogous to that used in human medicine, has been proposed to differentiate euthyroid from hyperthyroid cats in an earlier state. ⁶ Unfortunately, so far there are no such optimised TSH tests on the veterinary market. Repeated T₄ testing or other diagnostic tests such as T₃ suppression tests must be performed to classify those cats as being hyperthyroid or not.

In conclusion, we were able to show that the likelihood of hyperthyroidism increases with increasing size of the thyroid gland. Conversely, because a considerable proportion of euthyroid cats had a palpable thyroid gland, the presence of a goitre alone, even in the presence of clinical signs, is a weak indicator of symptomatic hyperthyroidism. Determination of additional serum biochemical parameters like ALP might help to further differentiate these cats, but must be viewed with caution. Based on our data, we cannot exclude the development of clinical hyperthyroidism in presumed euthyroid cats at a later time point. Therefore, long-term follow-up of euthyroid cats with a palpable thyroid gland should be undertaken to track their fate and to exclude or confirm an early state of hyperthyroidism.