Abstract
Objectives
The aim of this study was to estimate the prevalence of hypersomatotropism (HST) and hyperthyroidism in cats with diabetes mellitus (DM) from referral centers in Buenos Aires, Argentina.
Methods
This was a prospective study. Systematic screening of serum insulin-like growth factor 1 (IGF-1) and total thyroxine was performed in all cats diagnosed with DM at referral centers in Buenos Aires between February 2020 and February 2022.
Results
In total, 154 diabetic cats were evaluated (99 males and 55 females; median age 12 years [range 3–21]; mean body weight 5 kg [range 2–12]). Altogether, there were 115 (75%) domestic shorthairs and one domestic longhair; the remaining 38 cats were purebred (mainly Siamese, n = 25 [16%]). Twenty (12.9%) cats had IGF-1 concentrations >1000 ng/ml, and three (1.9%) had IGF-1 concentrations between 800 and 1000 ng/ml along with pituitary enlargement on CT, resulting in a 14.9% HST prevalence rate in diabetic cats. Intracranial imaging was performed in all cats with HST; median pituitary dorsoventral height was 5.8 mm (range 3.1–9.5). Fourteen of 23 (61%) cats had phenotypic changes consistent with acromegaly at the time of diagnosis of HST. Four of 154 (2.5%) cats had concurrent hyperthyroidism.
Conclusions and relevance
To date, this is the first study outside of Europe to have evaluated the prevalence of HST and hyperthyroidism in cats with DM. In Buenos Aires referral centers, feline HST is the most common concurrent endocrinopathy in cats with DM but with a lower prevalence than has previously been reported. Hyperthyroidism is a rare concurrent endocrinopathy in diabetic cats from referral centers in Buenos Aires.
Keywords
Introduction
Diabetes mellitus (DM) is a metabolic disorder of multiple etiologies characterized by chronic hyperglycemia with disturbances in the metabolism of carbohydrates, fats and proteins, resulting from defects in insulin secretion or insulin sensitivity in target tissues, or both.1–3 According to the American Diabetes Association, DM in humans can be classified into type 1 and type 2 DM, gestational DM and ‘specific types of DM’. 1 Within the ‘specific types of DM’, a subtype of DM related to endocrinopathies is described (acromegaly, Cushing syndrome, pheochromocytoma, primary hyperaldosteronism, hyperthyroidism, glucagonoma and other neuroendocrine tumors).1,4 According to the definitions of the Agreeing Language in Veterinary Endocrinology (ALIVE) project, DM in dogs and cats can be classified as insulin-deficient DM (beta [β] cell-related disorders) and insulin-resistant DM (target-organ disorders). 5
In domestic cats (Felis catus), it has been estimated that around 80% of cases present with type 2 DM, which is a heterogeneous disease attributable to a combination of insulin resistance and β cell failure.2,5–7 In contrast, type 1 DM (autoimmune destruction of pancreatic β cells) is considered extremely rare in cats.2,8,9 Within the ‘insulin-resistant DM’ category, the following endocrinopathies have been reported in cats: hypersomatotropism (HST), hyperthyroidism, hypercortisolism (HC) and primary hyperaldosteronism with hyperprogesteronism.3,10–12
HST, or acromegaly, is an emerging disease in diabetic cats. Although for many years it was considered a rare disease, studies carried out in Europe have indicated that between 17.8% (the Netherlands and Switzerland) and 24.8% (the UK) of diabetic cats have an excess of growth hormone (GH).13,14 Studies of the prevalence of other endocrine diseases in cats with DM are scarce; it has been reported that around 5% have hyperthyroidism and <3% possibly have HC.11,13 No studies have evaluated the prevalence of primary hyperaldosteronism–hyperprogesteronism in cats with DM.
Recognition of these types of ‘insulin-resistant DM’ is important in choosing the best treatment options and providing of the most accurate prognosis. 11 The prevalence of concurrent endocrine diseases in cats with DM from Argentina, or Latin America, is unknown. The main objective of this study was to evaluate the prevalence of HST and hyperthyroidism in diabetic cats from referral centers in Buenos Aires.
Materials and methods
This study was approved by the Ethics Committee of the Veterinary Science Center (Institutional Committee on the Care and Use of Experimental Animals [CICUAE]), in accordance with the laws on animal testing in Argentina and the recommendations of the World Health Organization. Informed written consent was obtained from all study cat owners before enrolment. This was a prospective study. Client-owned cats with newly diagnosed DM attending 10 private referral centers for endocrinology in Buenos Aires (metropolitan area and province of Buenos Aires) were enrolled in the study, which was conducted in the Veterinary Science Center (Maimonides University) between February 2020 and February 2022. To be included in the trial, insulin therapy had to have been initiated at least 6 weeks prior to enrolment. Likewise, cats on long-term (>14 months) insulin therapy were excluded. 15
A diagnosis of DM was made according to clinical signs, persistent fasting hyperglycemia, glycosuria and increased serum fructosamine concentrations. All cats were treated with insulin and received a commercial diet for diabetic cats.
Serum insulin-like growth factor 1 (IGF-1) and total thyroxine (TT4) concentrations were systematically evaluated in all diabetic cats that were diagnosed in these referral centers during the study period, regardless of the quality of glycemic control and whether or not they presented with insulin resistance.
In parallel, measurements of urinary cortisol:creatinine ratio (UCCR), low-dose dexamethasone suppression test (LDDST), aldosterone:renin ratio (ARR) and progesterone concentration were only evaluated in diabetic cats with clinical signs of hypercortisolism or hyperaldosteronism with hyperprogesteronism, or cats with an adrenal mass visible on ultrasound.
Diagnosis of HST was made according to clinical signs (acromegalic features), serum IGF-1 concentration and pituitary images obtained by CT. Cats were considered compatible with a diagnosis of HST based on a serum IGF-1 concentration >1000 ng/ml. 14 Likewise, in cases in which serum IGF-1 concentrations were between 800 and 1000 ng/ml, diagnosis of HST was further supported by the presence of pituitary enlargement on CT (>4 mm dorsoventral height). 16 Serum IGF-1 was measured by a commercially available radioimmunoassay (RIA; Immuno-Biological Laboratories). 17 CT studies with a Siemens SOMATOM Perspective 128 slice were performed in all cats with an IGF-1 concentration >800 ng/ml.
A diagnosis of hyperthyroidism was made according to clinical signs (eg, weight loss despite a good appetite and palpable goiter) and increased TT4 concentration (>4 µg/dl; reference interval [RI] 1–4 µg/dl). 18 TT4 measurements were carried out by chemiluminescence (Immulite 1000 TT4; Siemens).
In diabetic cats that presented with compatible clinical signs of HC (eg, abdominal distension, alopecia, unkempt haircoat, and fragile, thin, dry and inelastic skin) or cats that had an adrenal mass visible on ultrasound, specific hormonal tests were performed to rule out HC, hyperaldosteronism and hyperprogesteronism. A diagnosis of HC was made according to clinical signs, increased urinary UCCR (>36 ×10–6; RI <36 ×10–6) and lack of suppression on the LDDST (0.1 mg/kg dexamethasone IV; cortisol at 8 h >1.4 µg/dl). Cortisol measurements were made by RIA using a commercial kit (DPC).
A diagnosis of primary hyperaldosteronism was made according to clinical signs (eg, muscle weakness and ocular signs of hypertension), ultrasonographic identification of an adrenal mass in conjunction with hypokalemia and increased ARR (ratio of plasma aldosterone concentration to plasma renin activity [>4; RI <4]). 19 Aldosterone measurements were carried out with a commercially available RIA (Immunotech; Bioanalytical), and renin activity was calculated by measuring angiotensin I with a commercially available RIA (Immunotech; Bioanalytical). Diagnosis of hyperprogesteronism was made according to ultrasonographic identification of an adrenal mass in conjunction with serum progesterone concentrations ⩾5 nmol/l (RI ⩽2.0 for neutered cats). Progesterone measurements were made by chemiluminescence (Immulite 1000 Progesterone; Siemens). Abdominal ultrasonography was performed in all cats at the time of enrollment at the Veterinary Science Center (Maimonides University) with a Mindray M8 ultrasound scanner.
All blood tests were performed after 12 h of solid fasting. Blood samples were collected and centrifuged in private veterinary clinics. Urine samples were collected by the owners at home before they went to the clinic. All serum and urine samples were stored at 4°C until they were collected by the research team within 12 h of the samples being taken. Hormonal measurements were performed at Diagnotest Laboratory and at Veterinary Science Center (Maimonides University).
Statistical analysis
Statistical analysis was performed with GraphPad Prism 6. Data were analyzed for normal distribution (Shapiro–Wilk tests). The Mann–Whitney U-test was used to compare insulin doses in different groups of diabetic cats. Values were expressed as mean ± SD or median (range), as applicable, with a significance level set at P <0.05.
Results
In total, 154 diabetic cats were recruited during the study period. There were 115 domestic shorthair (DSH) cats (75%) and one domestic longhair; the remaining 38 cats were purebred: Siamese (n = 25; 16%), Oriental (n = 3; 1.9%), Burmese (n = 2; 1.3%), Ragdoll (n = 2; 1.3%), Persian (n = 2; 1.3%); Bengal (n = 1; 0.6%), Norwegian Forest Cat (n = 1; 0.6%), American Shorthair (n = 1; 0.6%) and Maine Coon (n = 1; 0.6%). Ninety-nine were male and 55 were female (all neutered); the median age was 12 years (range 3–21); mean body weight was 5 kg (range 2–12 kg).
All cats had at least 6 weeks of insulin therapy and none of them exceeded 6 months of insulin treatment at the time of enrolment (median 2 months [range 1.5–6]). All cats received insulin (143/154 [93%] insulin glargine; 6/154 [3.8%] human neutral protamine hagedorn [NPH] insulin; 3/154 [1.9%] detemir insulin; and 2/154 [1.3%] porcine insulin zinc suspension) and a commercial diet for diabetic cats. Insulin dosage ranged from 0.14 IU/kg to 5.7 IU/kg q12h (median 0.53). None of the cats received non-insulin antidiabetic drugs.
Serum IGF-1 concentration
Median serum IGF-1 concentration was 507 ng/ml (range 77–1980) (Table 1). Twenty (12.9%) cats had IGF-1concentrations >1000 ng/ml, and three (1.9%) had IGF-1 concentrations between 800 and 1000 ng/ml along with pituitary enlargement on CT, resulting in a 14.9% prevalence rate of HST in diabetic cats from Buenos Aires referral centers.
Serum insulin-like growth factor 1 (IGF-1) and total thyroxine (TT4) concentrations in diabetic cats with or without concurrent endocrinopathy
Data are presented as median (range)
HST = hypersomatotropism
Sixteen of these 23 cats were DSHs, six were Siamese and one was Burmese. Sixteen were male and seven were female (all neutered); the median age was 11 years (range 6–17); mean body weight was 5.5 kg (range 3.4–12).
Fourteen of 23 (61%) cats had phenotypic changes consistent with acromegaly at the time of HST diagnosis: 15 had prognathia inferior (65%), 10 had broad facial features (43%), five respiratory stridor (22%), four a clubbed paw appearance (17%) and four abdominal enlargement (17%). Of the cats with IGF-1 concentrations between 800 and 1000 ng/ml and pituitary enlargement, all three had mild phenotypic changes consistent with acromegaly (prognathia inferior and broad facial features).
Of the 154 cats, four had serum IGF-1 concentrations between 800 and 1000 ng/ml but no evidence of pituitary enlargement on CT, or acromegalic features.
Intracranial imaging was performed in all cats with HST and the median pituitary dorsoventral height was 5.8 mm (range 3.1–9.5). Of the 20 cats with an IGF-1 >1000 ng/ml, five had no consistent evidence of pituitary enlargement on CT.
Fifteen of 23 cats with HST (65%) showed difficulties in glycemic control, requiring increasing doses of insulin. Comparing insulin doses between diabetic cats (without concurrent endocrine disease) and diabetic cats with HST, significant differences were observed (median 0.31 IU/kg q12h [range 0.14–1.3] vs median 1.5 IU/kg q12h [0.15–5.7]; P <0.001). Likewise, significant differences were observed when comparing insulin doses between diabetic cats with HST and cats with HC (median 1.5 IU/kg q12h [range 0.15–5.7] vs median 0.5 IU/kg q12h [range 0.18–1.7]; P = 0.007).
Serum TT4 concentration
The median serum TT4 concentration was 1.7 µg/dl (range 0.8–6.6) (Table 1). Four of 154 cats had a serum TT4 concentration >4 µg/dl, resulting in a hyperthyroidism prevalence rate of 2.5% in diabetic cats from Buenos Aires referral centers.
All four were neutered male DSH cats; median age was 12.5 years (range 8–15) and mean body weight was 4.7 kg (range 3.6–5.1 kg). All four cats with TT4 concentration >4 µg/dl had clinical signs consistent with hyperthyroidism and the presence of palpable thyroid nodules (3/4 bilateral, 1/4 unilateral).
No significant differences in insulin dosages were observed between diabetic cats without concurrent endocrine disease (median 0.31 IU/kg q12h [range 0.14–1.3]), diabetic hyperthyroid cats (median 0.5 IU/kg q12h [range 0.2–0.8]) and diabetic cats with HC (median 0.5 IU/kg q12h [range 0.18–1.7]; P = 0.79).
UCCR, LDDST, ARR and progesterone concentrations
UCCR and LDDST were measured in 43 cats. Median UCCR was 9.2 × 10–6 (range 1.8–290 ×10–6) (Table 1). Median serum cortisol level 8 h post-dexamethasone was 0.2 μg/dl (range 0.1–3.6). Sixteen cats had a UCCR >36 ×10–6. Ten cats had a serum cortisol level >1.4 μg/dl 8 h post-dexamethasone. Therefore, 10/154 cats (6.5%) had clinical signs of HC, increased urinary UCCR and lack of suppression on LDDST.
All 10 cats had clinical signs consistent with HC. Nine cases had pituitary-dependent HC and one had adrenal-dependent HC.
No significant differences in insulin dosages were observed between diabetic cats (without concurrent endocrine disease) and diabetic cats with HC (median 0.31 IU/kg q12h [range 0.14–1.3] vs median 0.5 IU/kg q12h [range 0.18–1.7]; P = 0.15).
ARR and serum progesterone concentration were measured in two cats with adrenal tumors. One of 154 cats (0.6%) had an ARR >4 with serum progesterone concentrations ⩾5 nmol/l.
Discussion
This is the first study to date, carried out in Latin America, to have evaluated the prevalence of concurrent endocrinopathies in cats with DM. In this study, serum concentrations of IGF-1 and TT4 were systematically evaluated in diabetic cats that were diagnosed in 10 veterinary referral centers distributed around Buenos Aires. Likewise, UCCR and LDDST were measured when diabetic cats were suspected of having HC, and ARR and serum progesterone concentration was measured when diabetic cats were suspected of having primary hyperaldosteronism with hyperprogesteronism.
Studies on the prevalence of HST and hyperthyroidism in the diabetic cat population are lacking. After the first report on the prevalence of HST by Niessen et al, 20 HST/acromegaly was no longer considered a ‘rare disease’. In that study, serum IGF-1 concentrations were measured in 184 diabetic cats in the UK, and the prevalence of HST was estimated at 32%. 20 Subsequently, the same research team increased the size of the sample analyzed (n = 1221) and concluded that the overall prevalence of HST in diabetic cats in the UK was 24.8%. 14 A study conducted in the Netherlands and Switzerland estimated a HST prevalence rate of 17.8% in their population of diabetic cats. 13 The results of the present study suggest that HST is the most common concurrent endocrinopathy in diabetic cats, which is in agreement with previous prevalence studies carried out in Europe. However, the prevalence of feline HST in referral centers in Buenos Aires (14.9%) seems to be much lower than that observed in the UK (24.8–32%) and the Netherlands (21.6%), and similar to that observed in Switzerland (12.5%).13,14,20 It is a curious finding that the prevalence of HST in Buenos Aires is lower than in the UK and the Netherlands, since a higher prevalence could be expected in the present study owing to the potential bias of including cats from referral centers. It could be hypothesized that these differences are due to variations in geographic area (ie, regional variants of HST, with different receptor expressions, agonist affinity or signal transduction), cat breeds or endocrine disruptors. Nevertheless, further studies are necessary to explain these differences properly.
Diabetic cats included in this study had at least 6 weeks of insulin therapy, and no more than 6 months, for measurement of serum IGF-1 concentration. On the one hand, it is known that insulin deficiency in recently diagnosed diabetic humans and cats can influence serum IGF-1 concentrations, resulting in lower IGF-1 readings.10,21 On the other hand, prolonged insulin treatment (>14 months) can influence serum IGF-1 concentrations, leading to higher IGF-1 readings.15,22
Although numerous mechanisms have been described by which excess GH promotes the development of DM in humans, the most important mechanism is insulin resistance. GH promotes insulin resistance primarily through lipolysis, decreased glucose utilization in the muscle and the blockage of insulin-signaling mediators. 23 Considering that GH excess causes multiple systemic disorders, it is important to identify HST in order to control the condition itself and concurrent DM. 10 Therefore, since feline HST is the most common endocrine disease that promotes the development of insulin-resistant DM, it is essential to screen all diabetic cats, regardless of whether they have insulin resistance or not, and regardless of whether they have acromegalic features or not.10,14,20
Diagnostic criteria for HST may be relatively controversial for several reasons, including the chosen cut-off of 1000 ng/ml and the difficulty in identifying an enlarged pituitary on CT images. Serum IGF-1 concentration >1000 ng/ml have been associated with a 95% positive predictive value of the presence of HST. 20 This cut-off is probably too high, leading to an underestimation of presence of HST. In the present study, seven cats had serum IGF-1 concentrations in this ‘non-specific range’, of which four did not show evidence of pituitary enlargement and were classified as ‘normal’. It could be hypothesized that some cats with serum IGF-1 concentrations between 800 and 1000 ng/ml and without pituitary enlargement on CT are in an early stage of the disease, or with mild HST caused by hyperplasia of the pituitary somatotropic area or by a somatotropic microadenoma (difficult to identify on CT). Likewise, if 700 ng/ml (and not 800 ng/ml) is considered as the upper limit of the RI for serum IGF-1 concentration, as already evaluated in the UK, 24 the prevalence of HST in diabetic cats could be even higher. In addition, not all cases with HST have been described as having evident pituitary enlargement on CT,14,25 thus making diagnosis more difficult if cats have serum IGF-1 concentrations in the ‘non-specific range’. Fourteen of 23 (61%) cats in this study had different degrees of phenotypic changes consistent with acromegaly at the time of diagnosis of HST, while in other studies, physical changes were only observed in 5–24% of cases.13,14 Considering the results of this study, it is recommended that serum IGF-1 is measured in all diabetic cats, regardless of whether they have insulin resistance or not and regardless of whether they have an acromegalic phenotype or not.
It is important to highlight that not all cats with HST necessarily have concurrent DM. In acromegalic humans, most do not develop overt DM. 23 Reports in cats about this issue are lacking, and only five cases with HST without DM have been published to date.26–28 In a recent study on quality of life in cats with HST, another seven cases of acromegaly without DM were mentioned. 29
Two studies have indicated that the prevalence of hyperthyroidism in the population of diabetic cats is around 5–6%.11,13 One study concluded that cats with induced hyperthyroidism have impaired glucose tolerance, possibly due to an increase in insulin resistance. 30 Likewise, spontaneous hyperthyroidism may lead to long-lasting alterations of glucose tolerance and insulin secretion, which may not be reversed by treatment. 31 In the present study, 4/154 cats (2.6%) had a serum TT4 concentration >4 µg/dl, palpable thyroid nodules and clinical signs compatible with hyperthyroidism, showing a lower prevalence than previous reports.11–13 This prevalence may be underestimated, as serum TT4 concentrations may be lower with concurrent non-thyroidal disease. 32 To minimize this influence, TT4 evaluations were carried out at least 6 weeks after insulin therapy, and not at the time of diagnosis of DM. Although the ALIVE project mentions hyperthyroidism as a cause of insulin-resistant DM in cats, 5 it is essential to point out that establishing a causal relationship between both diseases is controversial. Hyperthyroidism is a very common disease in senior cats; 33 therefore, the concurrence with DM may be a coincidence due to the high prevalence of both diseases and the same age of presentation. Interestingly, the prevalence of hyperthyroidism in cats with DM from referral centers in Buenos Aires is relatively low (2.5%).
An excess of glucocorticoids have diabetogenic effects and may induce insulin resistance that predisposes to the development of DM in cats, dogs and humans.34–36 In cats, the concurrence of DM and HC is high, and it is estimated that around 80–90% of cats with HC have insulin-resistant DM.5,34 Studies on the prevalence of HC in populations of diabetic cats are lacking. In one study, UCCR levels were evaluated in diabetic cats and it was observed that 15% had increases in UCCR, but only 2.5% had clinical signs compatible with HC. 13 Evaluation of UCCR may be useful for the diagnosis of HC in cats, although it has several limitations: it requires urine sample collection by the owner at home and can be easily influenced by stress and non-adrenal diseases.37–39 In the present study, UCCR was evaluated in cats with clinical suspicion of HC and was complemented by hormonal diagnosis with LDDST, a more accurate test to arrive at a diagnosis of HC. 40 It has been reported that the results of the LDDST in cats with DM are normal. 41 Surprisingly, 10/154 cats (6.5%) had increased UCCR and a serum cortisol level 8 h post-dexamethasone of > 1.4 μg/dl, compatible with a diagnosis of HC. This finding is interesting because it indicates that the second concurrent endocrinopathy (after HST) in diabetic cats is probably HC and not hyperthyroidism, at least in referral centers in Buenos Aires during the study period.
It is known that multiple corticosteroid-secreting adrenocortical tumors (aldosterone and progesterone) can have concurrent DM. In a retrospective study of 10 cases of aldosterone and progesterone-secreting adrenal tumors, all cats had DM. 12 In the present study, 1/154 cats (0.6%) had primary hyperaldosteronism with hyperprogesteronism.
Diabetic cats with pheochromocytoma, glucagonoma or neuroendocrine tumors, as well as diabetic cats with primary hyperaldosteronism due to idiopathic hyperplasia without hyperprogesteronism, have not been reported in the literature and were not observed in this study.
This study has several limitations. The population of diabetic cats was relatively small compared with previous prevalence studies. However, this study had a selection bias, since although serum IGF-1 and TT4 were systematically evaluated in all diabetic cats, the referral of cats with DM is generally performed when metabolic control is poor and associated with concurrent diseases. In Argentina, although cats with DM are usually referred at the time of diagnosis, it is important to highlight that this study was conducted with diabetic cats that were referred to endocrinology referral centers and not primary care centers, implying a relevant bias. Therefore, the conclusions of this study should be interpreted considering the specific study population. The diagnosis of HC and primary hyperaldosteronism with hyperprogesteronism may have been underestimated in this study, since the systematic evaluation of UCCR, LDDST, ARR and progesterone was not performed in all the cats – only in those cases with a clinical suspicion of having any of these diseases.
Conclusions
This is the first study to date, outside of Europe, to have evaluated the prevalence of HST and hyperthyroidism in cats with DM. In Buenos Aires referral centers, feline HST is the most common concurrent endocrinopathy in cats with DM, but with a lower prevalence than previously reported. Hyperthyroidism is a rare concurrent endocrinopathy in diabetic cats from referral centers in Buenos Aires (2.5% of observed cases). HC is the second concurrent endocrinopathy (after HST) in diabetic cats, at least in referral centers in Buenos Aires, during the study period.
Footnotes
Acknowledgements
The authors would like to thank all veterinary surgeons around Buenos Aires, and all the diabetic cats and their owners, who participated in the study.
Correction (March 2023):
This article has been updated since its original publication to remove a duplicate word.
Conflict of interest
The authors declared no conflicts of interest with respect to the authorship, research and publication of this article.
Funding
The authors received no financial support for the research, authorship, and/or publication of this article.
Ethical approval
The work described in this manuscript involved the use of non-experimental (owned or unowned) animals. Established internationally recognised high standards (‘best practice’) of veterinary clinical care for the individual patient were always followed and/or this work involved the use of cadavers. Ethical approval from a committee was therefore not specifically required for publication in JFMS. Although not required, where ethical approval was still obtained, it is stated in the manuscript.
Informed consent
Informed consent (verbal or written) was obtained from the owner or legal custodian of all animal(s) described in this work (experimental or non-experimental animals, including cadavers) for all procedure(s) undertaken (prospective or retrospective studies). No animals or people are identifiable within this publication, and therefore additional informed consent for publication was not required.
