Electrocardiographic and echocardiographic evaluation in dogs with hypothyroidism before and after levothyroxine supplementation: A prospective controlled study

Abstract Background Improvement in cardiac function has been demonstrated after thyroxine treatment in humans with hypothyroidism using the myocardial performance index (MPI). Cardiac changes after thyroxine supplementation are poorly documented in dogs with spontaneous hypothyroidism and comparison with clinically healthy dogs is lacking. Objectives To evaluate the electrical activity and mechanical function of the heart in dogs with primary hypothyroidism at baseline (T0) and after thyroxine supplementation (T60). Animals Forty client‐owned dogs with hypothyroidism and 20 clinically healthy dogs. Methods Prospective cohort study. Selected electrocardiographic and echocardiographic variables, including the MPI, were measured in all dogs at T0 and in 30 hypothyroid dogs at T60. Results Hypothyroid dogs had significantly decreased median or mean heart rate (HR), P wave amplitude, and R wave amplitude (P = .04, P = .002, and P = .003, respectively) and E‐point‐to‐septal separation normalized to body weight (EPSSn) and trans‐mitral E wave velocity (E max; P < .001 and P = .025, respectively) at T0 compared to control dogs. At T60, significantly increased median or mean HR, P wave amplitude, fractional shortening, and E max (P < .001, P = .004, P = .002, and P = .009, respectively) and significantly decreased left ventricular end‐diastolic volume index, and normalized systolic diameter and EPSSn (P = .03, P = .03, and P = .001, respectively) were found. Conclusions and Clinical Importance Hypothyroidism in dogs induces mild and reversible changes of electromechanical cardiac function. The MPI does not have clinical importance in identifying cardiac dysfunction in affected dogs.

and induced thyroid hormone deficiency. [3][4][5][6][7][8][9][10][11][12][13] Reported ECG abnormalities in hypothyroid dogs include cardiac arrhythmias, such as sinus bradycardia, 7,8 atrial fibrillation, [4][5][6]10 and atrioventricular blocks, 7,9,12,14 decreased R wave amplitude, 9,10,12,14 and decreased amplitude or inverted T wave. 10 Partial response to thyroxine treatment with normalization of ECG changes was reported in some dogs, [6][7][8][9][10] and conversion of atrial fibrillation to sinus rhythm was observed in a dog treated solely with thyroxine. 3 Previous echocardiographic studies in dogs with hypothyroidism have been predominantly focused on evaluation of left ventricular (LV) systolic cardiac function. Echocardiographic changes observed in dogs with spontaneous or experimentally induced hypothyroidism include increased left ventricular diameter in systole (LVDS) and diastole (LVDD), 5,9 decreased LV fractional shortening (FS), 3,5 prolonged pre-ejection period, and decreased velocity of circumferential fiber shortening. 9 All of these alterations might mimic a familiar form of primary cardiomyopathy such as dilated cardiomyopathy (DCM). 15 Thus, it was previously thought that hypothyroidism-induced cardiomyopathy should be considered a differential diagnosis for DCM in dogs. 15 However, LV systolic function and time intervals should improve in dogs with hypothyroid-induced cardiac changes when appropriate hormone replacement treatment has been established. 16 In humans, both systolic and diastolic cardiac function can be influenced by thyroid hormones, 17 and the myocardial performance index (MPI), defined as the sum of isovolumetric contraction time plus relaxation time divided by ejection time, has been proposed as a useful marker of combined systolic and diastolic cardiac function. 18,19 Measurement of MPI in human patients with hypothyroidism showed deterioration of cardiac function that was reversible after treatment with thyroxine. 17,20 No study has prospectively investigated the effects of naturally occurring hypothyroidism on both systolic and diastolic function using the MPI in dogs. Furthermore, previous studies on cardiac dysfunction in dogs with hypothyroidism were carried out without the use of a control group of clinically healthy dogs. Thus, we evaluated cardiac electrical activity and mechanical function, including measurement of MPI, in euthyroid dogs and dogs with primary hypothyroidism before and after thyroxine supplementation.

| Animals
Client-owned dogs presented to 2 veterinary teaching hospitals and with a final diagnosis of primary hypothyroidism were prospectively included in the study. After the owner signed a written consent form, each dog underwent physical examination, ECG, echocardiographic examination, and blood sampling for CBC and serum biochemical profile. Dogs were considered hypothyroid if they showed consistent clinical signs of hypothyroidism (eg, dermatological abnormalities, lethargy, asthenia, exercise intolerance, cold intolerance, obesity, weight gain), serum total thyroxine (TT4), and canine thyrotropin hormone (cTSH) concentrations below and above the reference range, respectively (ie, TT4 <0.93 μg/dL and cTSH >0.5 ng/mL). In dogs with serum TT4 concentration below the reference range and normal serum cTSH concen- A control population of 20 clinically healthy dogs also was enrolled. They were recruited from the patients of the 2 hospitals presented for periodic routine visits or for evaluation before an elective procedure. Dogs were considered healthy on the basis of clinical history and normal results of physical examination, CBC, and serum biochemical profile, including serum TT4 and cTSH concentrations (ie, serum TT4 and cTSH >0.93 μg/dL and <0.5 mg/mL, respectively).
Moreover, these animals were considered to have normal cardiovascular health if results of echocardiographic examination were within normal limits and no rhythm disturbance (other than sinus arrhythmia or sinus tachycardia) was observed on ECG.

| Analytical procedures and thyroid evaluation
Serum cTSH concentrations were measured by use of a solid-part, 2-site chemiluminescent enzyme immunometric assay (IMMULITE 1000 Canine TSH; Diagnostic Products Corporation, Los Angeles, California). 21 The intra-assay coefficients of variation were 5.0%, 4%, and 3.8% at TSH concentrations of 0.20, 0.50, and 2.6 ng/mL, respectively. The interassay coefficients of variation were 6.3% and 8.2% at TSH concentrations of 0.16 and 2.8 ng/mL, respectively. The sensitivity of the assay was 0.03 ng/mL. The upper limit of the reference range was 0.5 ng/mL. Serum TT4 concentrations were determined using a homologous solid-phase, chemiluminescent enzyme immunoassay (IMMULITE 1000 Canine total T4, Diagnostic Products Corporation). The intra-assay and interassay coefficients of variation were 3.9%-10.8% and 5.2%-13.8%, respectively. The sensitivity of the assay was 0.5 μg/dL. The reference range for TT4 was 0.93-2.87 μg/dL. To perform the rhTSH-stimulation test, serum TT4 was measured before and 6 hours after the IV administration of recombinant human TSHc at a dose of 75 μg per dog. 22

| Cardiac evaluation
A complete cardiac evaluation including physical examination, 6 lead standard ECG, and trans-thoracic 2-dimensional (2D), M-mode, and echo-Doppler echocardiogram was performed at T0 in all dogs and at T60 in hypothyroid dogs.
Each ECG examination was performed with the dogs placed in the right lateral recumbency using dedicated devices (Archiwin, Esaote S.p. A., Firenze, Italy; Cube ECG, Cardioline S.p.A., Caverano, Italy; TouchECG HD, Cardioline S.p.A). Two-minute recordings were acquired to detect the presence of rhythm disturbances. Electrocardiographic measurements were performed by analyzing a 10-second strip, using a ruler.
Intervals and wave amplitude were obtained as multiple of 10 ms and 0.1 mV, respectively. Variables analyzed included cardiac rhythm (eg, normal sinus rhythm, sinus arrhythmia, pathological arrhythmias), heart rate (HR) in beats per minute (bpm) calculated by determining the number of QRS complexes in a 3-second interval and multiplying this number by 20 (reference range, 60-160 bpm), 23 amplitude and duration of the P wave, PQ-interval duration, R wave amplitude and duration of the QRS complex, duration of the QT interval corrected for HR according to the logarithmic formula (QT interval corrected = log600 × QT/logRR), 24 and mean electrical axis (MEA) of the QRS complex calculated using the following equation: MEA = arctan (Iamp, aVFamp) × 180/π. 25 Echocardiographic images were obtained in awake animals using ultrasound units (CX50, Philips, Eindhoven, The Netherlands; iE33 ultrasound system, Philips Healthcare, Monza, Italy) equipped with 1-5 or 3-8 MHz phased-array transducers and simultaneous ECG trace recording, according to previously published standards. 26 An M-mode interrogation of the LV was performed from the right parasternal short-axis view at the level of the chordae tendineae, and measurements were obtained using the leading edge-to-leading edge method. 27 The LVDD and LVDS were indexed to the body weight (BW) according to Cornell's method 28  M-mode images obtained from the short-axis view at the level of the mitral valve, as previously described in dogs. 29 Because EPSS is a linear dimension, mildly influenced by the weight of the animal, the absolute value then was indexed to BW (EPSSn) using the formula: EPSSn = (EPSS/BW 1/3 ). 28,30 Left atrial (LA) and aortic root (Ao) diameters were measured using a 2D method from the right parasternal short-axis view at the heart base level in early diastole, and their ratio (LA/Ao) was calculated. 31

| Statistical analysis
All data were tested for their distribution using a D'Agostino-Pearson normality test. Normally distributed data are reported as mean ± standard deviation, and nonnormally distributed data are reported as median and range (minimum and maximum).

| ECG and echocardiographic data
Electrocardiographic and echocardiographic data at T0 of control dogs and dogs with hypothyroidism are summarized in Table 2 Comparisons between ECG and echocardiographic variables obtained at T0 and T60 in 30 hypothyroid dogs are summarized in Table 3. After supplementation with levothyroxine, dogs with T A B L E 1 Descriptive data obtained from 60 dogs with (hypothyroid T0) and without (control g) hypothyroidism

| DISCUSSION
We evaluated the effects of hypothyroidism on cardiac electric activity and mechanical function before and after levothyroxine supplementa- Note: Normally and nonnormally distributed data are expressed as mean ± standard deviation and median and range (minimum-maximum), respectively. Values with statistically significant difference are indicated in bold. Abbreviations: EDVi, end diastolic volume indexed to body surface area; EF, ejection fraction; EPSSn, E-point-to-septal separation normalized to body weight; ESVi, end systolic volume indexed to body surface area; FS, fractional shortening; HR, heart rate; LVDDn, left ventricular diastolic diameter normalized to body weight; LVDSn, left ventricular systolic diameter normalized to body weight; MEA, mean electrical axis; MPI, myocardial performance index; QTc, QT corrected for heart rate, LA/Ao, left atrial to aortic root ratio.
index of the global myocardial function combining both systolic and diastolic cardiac performance. The principal findings were that hypothyroid dogs had mild ECG and echocardiographic changes compared to clinically healthy dogs, and these cardiac changes improved after levothyroxine supplementation. Furthermore, the MPI was not significantly different neither between hypothyroid dogs at the time of diagnosis and control dogs nor in the former dogs before and after approximately 2 months of treatment. Although the effect of thyroid hormones on cardiac function has been investigated previously in dogs, 8,9 ours is the first study comparing hypothyroid dogs with a control group of clinically healthy dogs.
The differential diagnosis of a dilated and hypokinetic LV is challenging in the dog, and the final diagnosis often can be achieved only by postmortem evaluation of the heart. 15 Therefore, evaluation of thyroid function has been suggested in the diagnostic evaluation to rule out hypothyroidism-induced cardiomyopathy. 15 Based on our results, this assumption likely should be reconsidered. Our findings are similar to those of a recent study focused on the evaluation of the relationship between hypothyroidism and DCM in Doberman Pinschers. 33 In that study, a cause-effect relationship between the 2 diseases could not be identified, and the authors stated that hypothyroidism does not play any role in the etiology and progression of DCM. 33 Similarly, results of our study do not justify considering hypothyroid-induced cardiac changes as identical to DCM, because the observed echocardiographic changes were too mild to reflect clinically relevant cardiac dysfunction. Some isolated reports of dogs with clinically relevant cardiac dysfunction associated with hypothyroidism have been published previously. [3][4][5][6] However, these cases were only sporadic compared to the overall prevalence of hypothyroidism in dogs. 34 It also must be considered that the diagnosis of hypothyroidism in dogs is often challenging, and the risk of misdiagnosing hypothyroidism in dogs with non-thyroidal illness syndrome is a major concern. 1,33,34 Dogs enrolled in our study represented a large cohort of hypothyroid dogs that were managed identically. In particular, all dogs underwent a standardized diagnostic protocol and all questionable cases received a rhTSH-stimulation test to obtain a definitive diagnosis. 22 Therefore, all dogs in our study had an ascertained diagnosis, but none of them showed clinical signs of cardiac disease nor had clinically relevant cardiac changes detected on ECG and echocardiography.
When systematically considering the measured ECG variables, mean HR was significantly decreased in the hypothyroid dogs at T0 compared to that of the control group, but only 1 hypothyroid dogs was bradycardic. Decreased HR in dogs with spontaneous and experimentally induced hypothyroidism has been reported, 7,9,11,12 and may be caused by a direct effect of thyroid hormones on the myocardium, decrease in tissue oxygen consumption, or downregulation of β-adrenergic receptors causing a decreased response to sympathetic stimulation. 35 The P-and R-wave amplitudes also were significantly decreased in the hypothyroid group compared to those of the control group. In 2 previous studies, low R-waves were reported in 11 of 19 (58%) and in 24 of 66 (36%) of hypothyroid dogs. 8,9 Although the exact cause of low R-wave amplitude is not known, obesity, decreased myocardial mass, or decreased circulating blood volume associated with hypothyroidism may be responsible for this ECG modification. 35