Blood pressure in hyperthyroid cats before and after radioiodine treatment

Authors: Lisa Stammeleer, Pilar Xifra, Sara I. Serrano, Mark Rishniw, Sylvie Daminet, Mark E. Peterson

Abstract

Background

Hyperthyroid cats commonly have systemic hypertension, with a reported prevalence of 7% to 48%. Although hypertension might be expected to resolve once treatment restores euthyroidism, it can persist or only first develop after treatment.

Objectives

To determine the proportion of hyperthyroid cats with hypertension (systolic blood pressure [SBP] ≥160 mm Hg), persistence or first development of hypertension after successful radioiodine treatment, and correlation of post‐treatment hypertension with azotemia or hypothyroidism.

Animals

Four hundred one hyperthyroid nonazotemic cats were included in the study.

Methods

Prospective, cross‐sectional and before‐and‐after studies. All hyperthyroid cats had SBP measured by Doppler; 255 had SBP rechecked 6 months after successful radioiodine (^131^I) treatment.

Results

Of untreated hyperthyroid cats, 108/401 (27%) were hypertensive. A higher proportion of hypertensive cats were nervous/excited compared with normotensive cats (47% vs 12%; P < .001). Of the initially hypertensive cats, 87/108 cats were reexamined after ^131^I treatment; 43/87 (49%) cats normalized SBP, whereas 44/87 (51%) remained hypertensive. Of the initially normotensive cats, 16/168 (9.5%) first developed hypertension after successful ^131^I treatment. 7/60 (12%) of the ^131^I‐treated hypertensive cats were azotemic and 9/60 (15%) were hypothyroid. A higher proportion of cats remaining hypertensive had nervous/excited demeanor than did normotensive cats (50% vs 17%; P < .001).

Conclusions/Clinical Importance

Hypertension, when present, resolves in many hyperthyroid cats after successful treatment. Hyperthyroid cats uncommonly develop new hypertension after treatment. Persistent or newly detected hypertension was unrelated to azotemia or iatrogenic hypothyroidism. More frequently perceived nervousness/anxiety in radioiodine‐treated hypertensive cats suggests that many of these cats might have “situational” hypertension, as hyperthyroid‐induced hypertension should resolve after treatment.

INTRODUCTION

In most cats diagnosed with systemic hypertension, an underlying cause can be identified, with chronic kidney disease (CKD) and hyperthyroidism being most commonly reported. ^1^ , ^2^ , ^3^ The prevalence of hypertension in untreated hyperthyroid cats varies, ranging from 6.7% to 47.6%. ^4^ , ^5^ , ^6^ , ^7^ , ^8^ , ^9^ , ^10^ , ^11^ , ^12^ Similarly, in large groups of hypertensive cats, the prevalence of hyperthyroidism ranges from 7% to 24%. ^13^ , ^14^ , ^15^ , ^16^ , ^17^ , ^18^ , ^19^ Despite the fact that hyperthyroid cats only rarely develop any evidence of target organ damage associated with sustained hypertension, such as retinopathy or encephalopathy, ^13^ , ^20^ , ^21^ most authorities recommend routine measurement of blood pressure as part of the initial workup in cats with hyperthyroidism. ^1^ , ^2^ , ^3^

Like cats, hyperthyroid human patients develop secondary systolic hypertension, with a prevalence that ranges from 20% to 30%. ^22^ , ^23^ Hyperthyroidism mainly leads to increases in systolic blood pressure (SBP), with diastolic pressures remaining normal. ^24^ , ^25^ Successful reestablishment of euthyroidism in these patients usually resolves the hypertension. ^26^ , ^27^ , ^28^

After successful treatment of hyperthyroidism in cats, one might similarly expect hypertension to resolve, but multiple investigators report only small or undetectable decreases in SBP after treatment. ^6^ , ^10^ , ^29^ , ^30^ In almost all of these studies, blood pressure was measured before and after treatment with methimazole or carbimazole, sometimes in combination with surgical thyroidectomy. ^6^ , ^8^ , ^30^ In only 2 studies was blood pressure measured in cats before and after treatment with radioiodine. ^10^ , ^29^ In contrast to reports in which hyperthyroid cats were hypertensive before treatment, up to 23% of cats first developed hypertension only after becoming euthyroid. ^6^ , ^9^ These cats had been treated with antithyroid drugs alone, or in combination with thyroidectomy. We are unaware of any studies that have examined the effect of iatrogenic hypothyroidism or the “unmasking” of CKD on either the resolution or development of hypertension.

Cats with hyperthyroidism are frequently noted to be hyperactive, irritable, anxious or restless, clinical signs that appear to worsen as the disease progresses in severity. ^21^ , ^31^ Therefore, it is possible that hyperthyroid cats develop situational hypertension (white coat hypertension) ^32^ , ^33^ more frequently than clinically normal cats or cats with nonthyroidal disease (eg, CKD). In one study, hyperthyroid cats had higher SBP values when measured by different operators in a typical clinical setting than when measured by highly experienced veterinary technicians in a secluded setting. ^5^ In addition, few hyperthyroid cats with hypertension develop signs of hypertensive retinopathy/choroidopathy or encephalopathy, ^21^ , ^34^ , ^35^ suggesting that the prevalence of true (rather than situational) hypertension might be overestimated in these cats.

In our study, we sought to determine the prevalence of systemic systolic hypertension (SBP ≥160 mm Hg) in a large cohort of nonazotemic cats with untreated hyperthyroidism. Furthermore, we sought to determine how the cats' perceived demeanor affects SBP, both before and after successful radioiodine (^131^I) treatment; if and how blood pressure changes in these cats after treatment; if any differences in blood pressure exist in ^131^I‐treated cats that become azotemic versus remaining nonazotemic; and if blood pressure differs in ^131^I‐treated cats that become euthyroid versus hypothyroid.

MATERIALS AND METHODS

Study cohort and study design

Our study was conducted in 2 phases. The first phase was a prospective, multicenter, cross‐sectional study ^36^ of hyperthyroid cats referred to 1 of 3 treatment sites for evaluation before radioiodine treatment (Figure 1). The second phase was a before‐after study ^37^ , ^38^ involving a subset of the cats from the initial study that were reevaluated approximately 6 months after successful treatment with radioiodine.

Approval was obtained from each institution's Animal Ethics Committee before the study commenced. All owners provided informed consent.

Initial cross‐sectional study

All hyperthyroid cats underwent an evaluation that included a review of the medical history, complete physical examination, routine laboratory testing (CBC, serum biochemical profile, complete urinalysis), and determination of serum thyroid hormone concentrations (total thyroxine [T4], triiodothyronine [T3], and thyroid‐stimulating hormone [TSH]). ^39^ , ^40^ All thyroid and TSH testing was done by assays validated for use in cats, as previously described. ^41^ To maintain consistency among the 3 study sites, all laboratory samples were submitted to a designated reference veterinary diagnostic laboratory (IDEXX Laboratories North America/Europe). Cats also underwent qualitative and quantitative thyroid scintigraphy to confirm hyperthyroidism. ^42^ , ^43^ We excluded cats if they had received methimazole within 7 days of evaluation, if they had concurrent nonthyroidal disease (eg, azotemia), or if they were receiving any drugs that could affect blood pressure.

Once hyperthyroid cats were deemed eligible for inclusion, SBP was measured on a separate day according to the American College of Veterinary Internal Medicine (ACVIM) consensus guidelines ^1^ by experienced operators following a standardized protocol. All cats were first placed in a quiet room, away from all other animals, and allowed to acclimate for at least 5 to 10 min. Each cat was then subjectively scored based on their temperament or demeanor with the following a 4‐level scoring 1 = calm, 2 = anxious, 3 = nervous, and 4 = excited/aggressive, ^33^ , ^44^ , ^45^ with cats removed from the study if they were deemed too nervous or aggressive to obtain accurate blood pressure readings. Eligible cats next had indirect SBP measured by a Doppler ultrasound technique. Cuff size was selected so that the cuff width was 30% to 40% of the circumference of the limb at the cuff site (most cats had a cuff size of 2‐3 cm). Using gentle restraint, cats were allowed to settle into a comfortable position in sternal recumbency once the cuff had been attached to the right forelimb. The hair over the radial artery was clipped and moistened with ultrasound gel, and the probe then placed over the region of the first palmar common digital artery. Throughout the assessment, the cuff was kept as level as possible with the level of the right atrium. At least 6 measurements were recorded, with the first being automatically discarded; once a plateau was reached, 5 to 7 consecutive measurements were obtained and averaged. Cats with high blood pressure (≥160 mm Hg) underwent indirect fundoscopy to determine if the cat had any evidence of retinal hemorrhage or detachment, subretinal edema, or tortuous vessels secondary to hypertension. ^46^ , ^47^

Before‐after study of hyperthyroid cats treated with radioiodine

A subset of hyperthyroid cats evaluated in the initial cross‐sectional study were reevaluated approximately 6 months after treatment with radioiodine. These cats again underwent a thorough evaluation that included a review of the medical history since time of ^131^I treatment, complete physical examination, routine laboratory testing (CBC, serum biochemical profile, urinalysis), and determination of serum T4 and TSH concentrations. Based on these results, we classified the cats as nonazotemic versus azotemic (serum creatinine >2.0 mg/dL). We also classified the cats as being euthyroid versus hypothyroid, based on the following euthyroid (T4, 1.0‐3.8 μg/dL; TSH ≤0.30 ng/mL), overtly hypothyroid (T4 < 1.0 μg/dL; TSH >0.30 ng/mL), and subclinically hypothyroid (T4, 1.0‐2.5 μg/dL; TSH >0.30 ng/mL), as previously described. ^41^ , ^48^ , ^49^ , ^50^ Cats that remained persistently hyperthyroid were excluded (Figure 1).

All these ^131^I‐treated cats had SBP measured again, by the protocol described above. For each cat, the temperament or demeanor at time of blood pressure measurement was again scored and recorded.

Data and statistical analysis

Data were assessed for normality by the D'Agostino‐Pearson test and by visual inspection of graphical plots. ^51^ Data were not normally distributed; therefore, all analyses used were performed by nonparametric tests.

For analysis, hyperthyroid cats were divided into hypertensive (defined as cutoff ≥160 mm Hg) and normotensive groups. The cats were also categorized into 2 groups of disease severity based on total T4 and T3 concentrations. Cats with mild‐moderate hyperthyroidism had serum T4 and T3 concentrations <9.4 μg/dL and <126 ng/dL, respectively, whereas cats with moderate‐to‐severe hyperthyroidism had T4 and T3 concentrations ≥9.4 μg/dL and ≥ 126 ng/dL. We also divided cats into 2 categories based on their demeanor score ^44^ (ie, calm or anxious cats [score 1 or 2] vs nervous, excited, or aggressive cats [score 3 or 4]).

Results for continuous data (eg, blood pressure, serum T4, T3, TSH, and creatinine concentrations) are expressed as median (interquartile range [25th‐75th percentile]) and represented graphically as box‐and‐whisker plots (Tukey method). ^52^ Comparisons between 2 continuous variables between groups or within groups (before‐after) were analyzed with the Mann‐Whitney U test and Wilcoxon signed ranks test, respectively. Comparisons among 3 continuous variables (ie, among the 3 different centers) were analyzed with the Kruskal‐Wallis test, followed by Conover's multiple comparisons test.

Results for qualitative (categorical) data are expressed as ratio (eg, breed, sex) or percent of cats (eg, prevalence of underweight, demeanor). Categorical variables were compared between 2 groups by the chi‐square test or Fisher's exact test, where appropriate. For comparison of 3 categorical variables (ie, among the 3 different centers), data were again analyzed with the chi‐square test or Fisher's exact test, followed by the Holm‐Bonferroni correction procedure for within group comparison.

For all the data on the supplemental tables, we also used Holm‐Bonferroni method to account for multiple comparisons to reduce the family‐wise error rate (type 1 errors). ^53^ , ^54^

For all analyses, statistical significance was defined as P ≤ .05. Statistical analyses were performed by proprietary statistical software (GraphPad Prism, version 10.2; GraphPad Software, La Jolla, California, USA; MedCalc, version 22.0, MedCalc Statistical Software, Ltd, Ostend, Belgium).

RESULTS

Initial cross‐sectional study (untreated hyperthyroid cats)

Characteristics of hyperthyroid cats

During the study period, 401 hyperthyroid cats were enrolled in the initial cross‐sectional 124 cats from clinic 1 (Ghent), 138 cats from clinic 2 (Madrid), and 139 cats from clinic 3 (New York; Figure 1). The 401 cats ranged in age from 3 to 20 years (median, 13 years; IQR, 11‐14 years; Supplemental Table 1). Breeds included Domestic Shorthair and Longhair (n = 358; 89.2%), Siamese (13), Ragdoll (9), Norwegian Forest Cat (4), Persian (4), British Shorthair (3), Eastern Shorthair (2), Maine Coon (2), Russian Blue (2), Siberian Forest Cat (2), and Abyssinian, Burmese, Bengal, Devon Rex, and Scottish Fold (1 each). Of the 401 cats, 224 (56%) were female, and 177 (44%) were male; 389 cats (97%) had been neutered, and 12 cats were intact.

The time from diagnosis of hyperthyroidism to ^131^I treatment ranged from 7 days to 6 years (median, 71 days; IQR, 38‐186 days Supplemental Table 1). Of the 401 cats, 160 (40%) had been treated with methimazole, but the drug was discontinued at least 7 days before evaluation in all cats.

Body weight ranged from 1.5 to 7.95 kg (median, 4.0 kg; IQR, 3.3‐4.8); 138 (34.4%) cats were considered underweight, 199 (49.6%) had an ideal body condition score (BCS), and 64 (16%) were considered overweight (Supplemental Table 1). Of the 401 cats with untreated hyperthyroidism, 79 (19.7%) cats had a normal muscle condition score (MCS), whereas mild, moderate, and severe muscle loss was recorded in 132 (32.9%), 131 (32.7%), and 59 (14.7%), respectively (Supplemental Table 1).

Of the hyperthyroid cats, 394/401 (98%) had high serum T4 concentrations (Supplemental Table 1). The 7 cats with high‐normal serum T4 concentrations had their hyperthyroidism confirmed by finding high free T4 concentrations (by equilibrium dialysis), ^39^ as well as scintigraphic evidence of hyperthyroidism (ie, “hot” thyroid nodules, with high thyroid‐to‐salivary gland ratios). ^42^ , ^43^ Of the 401 hyperthyroid cats, serum TSH concentrations were below the limit of detection (<0.03 ng/mL) in 391 cats (97.5%). Of the 401 hyperthyroid cats, 200 had moderate‐to‐severe disease, whereas 201 had mild‐to‐moderate disease.

When the hyperthyroid cats from each of the 3 treatment sites were compared, the New York cats were younger (median, 12 years; IQR, 11‐14 years) than the cats from Ghent and Madrid (median, 13 years; IQR, 11‐14 years; Supplemental Table 2). The New York cats also weighed more (median, 4.3 kg; IQR, 3.48‐5.09 kg) than the cats from Ghent (median, 3.3 kg; IQR, 3.3‐4.6 kg) and Madrid (median, 3.7 kg; IQR, 3.1‐4.7 kg). The New York cats also had higher BCSs and MCSs than the cats from Ghent and Madrid. The New York cats also had lower serum T3 concentrations, higher serum creatinine concentrations, and lower urine specific gravity values than did the cats from Ghent and Madrid (Supplemental Table 2).

Pretreatment systolic blood pressure in 401 hyperthyroid comparing hypertensive to normotensive cats

In the 401 cats with untreated hyperthyroidism, the median SBP was 147 mm Hg, with a range from 100 to 240 mm Hg (IQR, 130‐160 mm Hg; Supplemental Table 1). Of these 401 cats, 108 (27%) had SBP ≥160 mm Hg, and 50 (12.5%) had SBP >180 mm Hg (Supplemental Table 1). The remaining 293 cats (73%) were normotensive (<160 mm Hg).

A higher proportion of hypertensive cats had low BCSs than did normotensive cats (48% vs 29%; P = .001; Supplemental Table 1). Hypertensive cats also had higher serum T3 concentrations than did normotensive cats (171 vs 117 ng/dL; P = .001; Supplemental Table 1). Cats with moderate‐to‐severe hyperthyroidism had higher SBP than did cats with mild‐to‐moderate hyperthyroidism (150 vs 140 mm Hg; P < .001; Figure 2A,B).

Finally, a higher proportion of the hypertensive cats were nervous/aggressive compared with normotensive cats (47% vs 12%; P < .001; Supplemental Table 1). The 87 hyperthyroid cats with a nervous, excited, or fractious demeanor also had higher SBP (P = .001) than did the 314 cats that were judged to be calm or anxious (Figure 3).

When the hyperthyroid cats from each of the 3 treatment sites were compared, the New York cats had lower SBP (median, 131 mm Hg; IQR, 120‐145 mm Hg) than did the cats from Ghent (median, 155 mm Hg; IQR, 140‐165 mm Hg) or Madrid (median, 154 mm Hg; IQR, 140‐167 mm Hg; Supplemental Table 2). New York cats also had a lower prevalence of hypertension (12.1%) than did the cats from Ghent (31.5%) or Madrid (38.2%). The New York cats also had a much lower prevalence of nervous/excited demeanor (3.6%) than did the cats from Ghent (52.1%) or Madrid (23.2%; Supplemental Table 2).

Before‐after study of hyperthyroid cats treated with radioiodine

Characteristics of the cats studied before‐after

^131^I‐treatment

We reevaluated 255/401 cats 6.5 months (IQR, 5.6‐7.5 months) after ^131^I treatment (Figure 1). These 255 cats ranged in age from 3 to 20 years (median, 12 years; IQR, 11‐14 years; Supplemental Table 3). Breeds included domestic longhair and shorthair (n = 222; 87.1%), Siamese (11), Ragdoll (5), Norwegian Forest Cat (3), Persian (3), British Shorthair (2), Russian Blue (2), Siberian Forest Cat (2), Abyssinian (1), Bengal (1). Burmese (1), Eastern Shorthair (1), and Maine Coon (1). One hundred thirty‐two cats (51.8%) were female, and 123 were male; 247 cats (97%) had been neutered. The signalment and clinical features of these 255 cats were similar to that of the entire cohort of 401 cats (Supplemental Table 1).

Posttreatment body weight, body and muscle condition, and serum concentrations of T4

, TSH, and creatinine in 255 ^131^I‐treated cats

After ^131^I treatment, the 255 cats had increases in body weight, BCS, and MCS (P < .001; Supplemental Table 4). Cats' serum T4 concentrations decreased after ^131^I treatment, whereas the suppressed serum TSH concentrations increased (P < .001; Supplemental Table 4). Of the 255 cats, 193 (76%) became euthyroid (normal serum T4 and TSH concentrations), whereas 62 (24%) developed either subclinical (47/62) or overt hypothyroidism (15/62; low to low‐normal T4 with high TSH concentrations; Figure 1, Supplemental Table 4).

Serum creatinine concentrations increased after ^131^I treatment (P < .001; Supplemental Table 4), with 44 (17%) of the 255 cats becoming azotemic (serum creatinine >2.0 mg/dL). When azotemic cats were separated into 2 groups based on their thyroid status, the 62 hypothyroid cats were more likely to become azotemic than were the 193 euthyroid cats (30.7% vs 13%, P = .003). The proportion of cats having a nervous/aggressive demeanor did not change when reevaluated after ^131^I treatment (before, 27.8% vs after, 25%; P = .53; Supplemental Table 4). Of the 255 cats, 241 (94.5%) did not change their demeanor score after treatment, 3 (1.3%) went up in demeanor score (from 1‐2 to 3‐4), and 11 (4.2%) showed a decrease in demeanor score (from 3‐4 to 1‐2).

When the hyperthyroid cats from each of the 3 treatment sites were compared after ^131^I treatment, the cats from Ghent weighed less (median, 4.2 kg; IQR, 3.6‐5.1 kg) than the cats from Madrid (median, 4.7 kg; IQR, 3.7‐5.6 kg) or New York (median, 4.7 kg; IQR, 4‐5.6 kg; P = .04; Supplemental Table 5). The New York cats had higher posttreatment serum TSH concentrations (median, 0.14 ng/mL; IQR, 0.05‐0.47 ng/mL) than did cats from Ghent (median, 0.08 ng/mL; IQR, 0.03‐0.26 ng/mL) or Madrid (median, 0.07 ng/mL; IQR, 0.03‐0.25 ng/mL; P = .02; Supplemental Table 5), but there was no difference (P = .56) in the prevalence of iatrogenic hypothyroidism among the 3 treatment sites. The prevalence of posttreatment azotemia was higher in the cats from Madrid (19%) and New York (27%) than in the cats from Ghent (5.2%; P = .01; Supplemental Table 5).

Before‐after systolic blood pressure in 255 hyperthyroid cats

The cats that underwent follow‐up evaluation had an initial median SBP of 150 mm Hg (IQR, 130‐165 mm Hg), 87 (34%) had SBP ≥160 mm Hg, and 43 (17%) had SBP ≥180 mm Hg (Supplemental Table 3). After ^131^I treatment, median blood pressure fell from 150 to 145 mm Hg (IQR, 130‐158 mm Hg; P = .001; Supplemental Table 4), with the prevalence of cats having SBP ≥160 mm Hg decreasing from 34% to 23%; (P = .01; Supplemental Table 4).

Systolic blood pressure in the initially normotensive cats did not change (P = .12) after ^131^I treatment (Figure 4A). One hundred fifty‐two (90.5%) of these 168 cats remained normotensive after ^131^I treatment, whereas 16 (9.5%) previously normotensive cats first developed hypertension only after treatment (Figures 4A and 5).

In contrast, SBP decreased in the initially hypertensive cats (P < .001) after ^131^I treatment (Figure 4B). Forty‐four (50.6%) of the 87 cats remained hypertensive after ^131^I treatment, whereas 43 cats became normotensive (Figures 4B and 5).

When the hyperthyroid cats from each of the 3 treatment sites were compared after treatment, the New York cats had a lower SBP (median, 132 mm Hg; IQR, 120‐145 mm Hg) than did the cats from Ghent (median, 150 mm Hg; IQR, 145‐165 mm Hg) or Madrid (median, 145 mm Hg, IQR, 130‐161 mm Hg; P < .001; Supplemental Table 5). New York cats also had a lower prevalence of posttreatment hypertension (11%) than did the cats from Ghent or Madrid (both 30%; P = .01; Supplemental Table 5). The New York cats also had a much lower prevalence of nervous/excited demeanor (4.1%) than did the cats from Ghent (47.5%) or Madrid (26.7%; P < .001; Supplemental Table 5).

Comparison of SBP in

^131^I‐treated cats that developed azotemia or remained nonazotemic

After ^131^I treatment, median serum creatinine concentration increased from 1.0 mg/dL (IQR, 0.7‐1.2 mg/dL) to 1.5 mg/dL (IQR, 1.3‐2.0 mg/dL; P < .001; Supplemental Table 4). However, when the 196 normotensive cats were compared with the 59 hypertensive cats at the follow‐up evaluations, the serum creatinine concentrations (median, 1.6 vs 1.5 mg/dL; IQR, 1.3‐2.0 vs 1.2‐1.9 mg/dL; P = .32) and prevalence of azotemia (19% vs 12%; P = .24) were similar (Supplemental Table 6). Finally, the 44 cats that became azotemic after treatment had SBP similar to that of the 211 cats that remained nonazotemic (P = .06; Figure 6).

Comparison of SBP in

^131^I‐treated cats that became euthyroid versus hypothyroid

After ^131^I treatment, we found no difference in median serum concentrations of T4 (median, 1.7 mg/dL; IQR, 1.3‐2.2 mg/dL vs median, 1.6 mg/dL; IQR, 1.0‐2.0 mg/dL; P = .25) or TSH (median, 0.10 ng/mL; IQR, 0.03‐0.33 ng/mL; vs median, 0.06 ng/mL; IQR, 0.03‐0.2 ng/mL; P = .11) between the 196 normotensive cats and the 59 cats with posttreatment hypertension (Supplemental Table 6). In addition, the 59 hypertensive cats had a prevalence of iatrogenic hypothyroidism that was similar to that of the normotensive cats (15% vs 27%; P = .08; Supplemental Table 6).

The 62 cats with iatrogenic hypothyroidism had similar SBP to the 193 euthyroid cats (median, 140 vs 145 mm Hg; IQR, 130‐152 vs 130‐160 mm Hg; P = .14; Figure 7). The 62 hypothyroid cats had a similar prevalence of hypertension (9/62; 14.5%) compared with that of the euthyroid cats (50/193; 25.9%; P = .08).

Of these 62 hypothyroid cats, 47 (75.8%) had subclinical hypothyroidism, whereas 15 (24.2%) had overt hypothyroidism (Figure 1). The 15 cats with overt hypothyroidism (median, 136 mm Hg; IQR, 126‐151 mm Hg), 47 cats with subclinical hypothyroidism (median, 140 mm Hg; IQR, 135‐155 mm Hg), and 193 euthyroid cats (median, 145 mm Hg; IQR, 130‐160 mm Hg) had similar SBP values (P = .2; Supplemental Figure 1).

Before‐after comparison of SBP in cats with calm/anxious demeanor versus nervous/aggressive demeanor

Before treatment, a higher proportion of hypertensive cats were scored as being nervous/aggressive than the normotensive cats (50% vs 16%; P < .001; Supplemental Table 3), similar to the findings in the entire group of 401 untreated cats (Supplemental Table 1). The 70 untreated hyperthyroid cats with a nervous/aggressive demeanor also had higher SBP (P < .001) than did the 185 cats that were calmer (Figure 8A), again similar to findings in the entire group of 401 cats (Figure 3; Supplemental Table 3).

After treatment, a higher proportion of the 59 hypertensive cats were scored as being nervous/aggressive than the normotensive cats (52% vs 16.4%; P < .001; Supplemental Table 6). The 63 treated cats with a nervous/aggressive demeanor continued to have higher SBP (P < .001) compared with the 192 cats that were calmer (Figure 8B).

DISCUSSION

Our multicenter, prospective study of over 400 hyperthyroid cats treated with ^131^I identified hypertension in approximately a quarter at presentation for ^131^I treatment. Of these, approximately half normalized their blood pressure after successful ^131^I treatment, and cats infrequently developed posttreatment hypertension. Persistent or newly diagnosed hypertension was not associated with posttreatment azotemia or iatrogenic hypothyroidism in these cats. Fifty percent of hypertensive cats, either at the initial presentation or after ^131^I treatment, exhibited higher levels of anxiety or stress during examination, suggesting that their hypertension might be situational rather than disease related.

The prevalence of hypertension in the untreated cats in our study falls midway between the ranges of previous studies ^4^ , ^5^ , ^6^ , ^7^ , ^8^ , ^9^ , ^10^ , ^11^ , ^12^ , ^55^ and is similar to prevalence reported in hyperthyroid people. ^22^ , ^23^ However, we observed an effect of treatment center on this prevalence, with centers in Europe having substantially higher proportions of cats with hypertension (32% and 38%) than the center in the United States (12%). The European centers also reported substantially higher proportions of nervous or anxious cats (47% and 22%) than the center in the United States (4%). Thus, prevalence of reported hypertension in previous studies might be site‐specific, rather than generalizable to populations of hyperthyroid cats at large, and could explain (at least in part) the large range of reported prevalence in those studies. ^4^ , ^5^ , ^6^ , ^7^ , ^8^ , ^9^ , ^10^ , ^11^ , ^12^ , ^55^

We found no association between either pretreatment or posttreatment hypertension and azotemia—measures of kidney function were similar between normotensive and hypertensive cats, and proportions of hypertensive cats with azotemia did not differ from proportions of nonhypertensive cats with azotemia. Our findings mirror those of previous studies, in which the plasma creatinine concentration did not differ between normotensive and hypertensive hyperthyroid cats ^55^ and pretreatment hypertension did not predict the development of posttreatment azotemia. ^11^ However, our findings differ from those of other investigators, who reported that a high proportion (29% and 36%) of cats developing hypertension after successful antithyroid treatment also became azotemic. ^9^ , ^55^ In these cats, the investigators attributed the development of hypertension to the unmasking of kidney disease.

Successful treatment of hyperthyroidism in human patients is usually accompanied with resolution of hypertension. ^26^ , ^27^ , ^28^ However, only half of the hypertensive hyperthyroid cats in our study normalized blood pressure after ^131^I treatment. This contrasts with 2 previous studies in cats that found no difference in SBP before and after radioiodine treatment. ^10^ , ^29^

In contrast to previous studies, which documented novel hypertension in up to 24% of cats after treatment, ^6^ , ^9^ , ^55^ only 9.5% of these cats became hypertensive after treatment. The investigators in these earlier studies defined hypertension as a blood pressure above 170 mm Hg in 2 studies ^9^ , ^55^ and above 175 mm Hg in the third study. ^6^ Therefore, by the same criteria as we did, their prevalence of posttreatment novel hypertension would likely have been even higher. Similar to the effect of study center on pretreatment prevalence of hypertension, we observed a study center effect on novel posttreatment prevalence. Nevertheless, there was no significant difference in development of new hypertension among the different study sites. In our study, cats were treated with radioiodine, which provides a more constant control of thyroid status. In one study following 133 apparently healthy cats aged ≥9 years, 7% of cats developed idiopathic hypertension over the study period. ^56^ This percentage is comparable with the development of new hypertension posttreatment in these cats. This is supportive of the relative low risk of development of new hypertension in ^131^I‐treated hyperthyroid cats.

We found that cats identified as nervous or anxious during the examination had a higher probability of being hypertensive than those with a more relaxed or calm demeanor. Although such perceived anxiety might not reflect the true emotional state of every cat, our data suggest that situational (“white coat”) hypertension might explain much of the apparent prevalence of hypertension both before and after successful resolution of hyperthyroidism. Importantly, we observed an effect of study center on both demeanor and hypertension, with cats in the United States being reported as much less frequently anxious or nervous and exhibiting a lower probability of hypertension than cats in the European centers. The clinical circumstances do differ substantially among the centers in Europe and the center in the United States. The centers in Europe are larger and busier clinics, whereas the center in the United States is a smaller feline‐only clinic (living room–like environment). These findings suggest that many of the previous studies might have also had a localized effect on apparent hypertension, rather than reflecting a true prevalence of pathological hypertension. This would help explain the large range of reported prevalence of hypertension by different investigators. It would also help explain the somewhat lower resolution following reestablishment of euthyroidism in cats than commonly occurs in human patients. Finally, like previous investigators, ^12^ we found only a weak association between hypertension and thyroid disease severity—if hyperthyroidism commonly caused hypertension, we might expect cats with more severe or long‐standing disease to have a higher probability of hypertension. Our suspicion that many “hypertensive” hyperthyroid cats are simply situationally hypertensive is further supported by the observations that other investigators consider severe hypertension uncommon ^5^ , ^32^ , ^57^ and that the prevalence of situational hypertension in elderly cats is high (at least 20%). ^33^ Finally, in agreement with the findings in our study in which none of these cats with a blood pressure ≥ 160 mm Hg had any evidence of retinopathy on fundus examination, hyperthyroid cats rarely develop hypertensive ocular lesions. ^4^ , ^13^ , ^34^

Our study includes some limitations. First, because of our 3 treatment sites, multiple operators performed blood pressure measurements and assigned a demeanor score to each cat. In addition, when cats were reevaluated 6 months after ^131^I treatment, operators were not always blinded to their initial demeanor score. These factors could have at least partially accounted for the observed effect of study center on both demeanor and hypertension. Second, we categorized these hyperthyroid cats as being hypertensive or normotensive based on only a single session of blood pressure measurements. Although it is true that cats with a single SBP ≥160 mm Hg have “high” blood pressure, that finding by itself does not equate to a diagnosis of hypertension. Rather, as per ACVIM guidelines, ^1^ a reliable diagnosis of hypertension (in the absence of target organ damage) is best confirmed by measuring a SBP ≥160 mm Hg on 2 or more occasions. Although hypertension was not confirmed in all of the cats, about 80% of hyperthyroid cats initially found to have a SBP ≥160 mm Hg did indeed have blood pressure again measured 6 months after treatment, thus fulling the diagnostic requirements outlined in the ACVIM guidelines. ^1^

Based on our results, we can conclude that the hyperthyroid cats seem to be predisposed to situational or stress and behavior‐induced hypertension rather than true pathological hypertension. Both severe hypertension (with associated risk of target organ damage) and development of new hypertension after ^131^I treatment are rare.

CONFLICT OF INTEREST DECLARATION

Authors declare no conflict of interest.

OFF‐LABEL ANTIMICROBIAL DECLARATION

Authors declare no off‐label use of antimicrobials.

INSTITUTIONAL ANIMAL CARE AND USE COMMITTEE (IACUC) OR OTHER APPROVAL DECLARATION

Authors declare that ethics approval and owner consent was obtained before the study commenced.

HUMAN ETHICS APPROVAL DECLARATION

Authors declare human ethics approval was not needed for this study.

Supporting information