The study was performed at The College of Veterinary Medicine, Ohio State University, Columbus, OH.
Comparison of the Effects of Ivabradine and Atenolol on Heart Rate and Echocardiographic Variables of Left Heart Function in Healthy Cats
Article first published online: 21 MAR 2011
Copyright © 2011 by the American College of Veterinary Internal Medicine
Journal of Veterinary Internal Medicine
Volume 25, Issue 3, pages 469–476, May/June 2011
How to Cite
Riesen, S.C., Schober, K.E., Cervenec, R.M. and Bonagura, J.D. (2011), Comparison of the Effects of Ivabradine and Atenolol on Heart Rate and Echocardiographic Variables of Left Heart Function in Healthy Cats. Journal of Veterinary Internal Medicine, 25: 469–476. doi: 10.1111/j.1939-1676.2011.0705.x
- Issue published online: 3 MAY 2011
- Article first published online: 21 MAR 2011
- Submitted June 30, 2010; Revised January 12, 2011; Accepted January 27, 2011.
- Heart rate;
- Hypertrophic cardiomyopathy;
Background: Ivabradine is a novel negative chronotropic drug used for treatment of ischemic heart disease in people. Little is known about its effects and safety in cats.
Hypothesis/Objectives: Ivabradine is not inferior to atenolol with regard to clinical tolerance, heart rate (HR) reduction, and effects on cardiac function in healthy, lightly sedated cats.
Animals: Ten healthy laboratory cats.
Methods: Physical examination, systolic blood pressure measurement, and transthoracic echocardiography were performed in all cats at baseline and after oral administration (4 weeks each) of ivabradine (0.3 mg/kg q12h) and atenolol (6.25 mg/cat q12h; 1.0–1.7 mg/kg) in a prospective, double-blind, randomized, active-control, fully crossed study. A priori noninferiority margins for the effects of ivabradine compared with atenolol were set at 50% (f= 0.5) based on predicted clinical relevance, observer measurement variability, and in agreement with FDA guidelines. Variables were compared by use of 2-way repeated measures ANOVA.
Results: Ivabradine was clinically well tolerated with no adverse events observed. HR (ivabradine, P < .001; atenolol, P < .001; ivabradine versus atenolol, P= .721) and rate-pressure product (RPP) (ivabradine, P < .001; atenolol, P= .001; ivabradine versus atenolol, P= .847) were not different between treatments. At the dosages used, ivabradine demonstrated more favorable effects than atenolol on echocardiographic indices of left ventricular (LV) systolic and diastolic function and left atrial performance.
Conclusions and Clinical Importance: Ivabradine is not inferior to atenolol with regard to effects on HR, RPP, LV function, left atrial performance, and clinical tolerance. Clinical studies in cats with hypertrophic cardiomyopathy are needed to validate these findings.
- Ao Vmax
maximum aortic velocity
- AR duration
duration of the late diastolic pulmonary vein atrial reversal flow wave
- E : A
ratio between peak velocity of early diastolic transmitral flow (E wave) to peak velocity of late diastolic transmitral flow (A wave)
- E : IVRT
ratio between peak early transmitral flow velocity (E wave) to isovolumic relaxation time (IVRT)
fractional area change
heart rate obtained at clinical exam
heart rate obtained at echocardiographic exam
isovolumic relaxation time
area d left atrial area in diastole
area s left atrial area in systole
- LA SF
left atrial shortening fraction
minimum left artrial antero-posterior dimension
maximum left atrial antero-posterior dimension
left ventricular internal dimension in diastole
left ventricular internal dimension in systole
- Peak AR
peak velocity of late diastolic pulmonary vein atrial reversal flow wave
- Peak D
peak velocity of diastolic pulmonary vein flow
- Peak E
peak velocity of early diastolic transmitral flow
- Peak Ea lat
peak velocity of early diastolic motion of the lateral mitral annulus
- Peak LAA flow
peak velocity of left atrial appendage flow
- Peak S
peak velocity of systolic pulmonary vein flow
- Radial SR Peak S
peak radial systolic strain
- Radial SrR Peak A
peak radial late diastolic strain rate
- Radial SrR Peak E
peak radial early diastolic strain rate
- Radial SrR Peak S
peak radial systolic strain rate
- S : D
ratio between peak systolic (S wave) to peak diastolic (D wave) pulmonary vein flow velocity
systolic blood pressure
Hypertrophic cardiomyopathy (HCM) is the most commonly observed myocardial disease in cats.1,2 The progression of HCM is variable with congestive heart failure being an important clinical outcome. Although the mechanisms leading to clinical decompensation are not yet fully understood, several risk factors have been suggested including, but not limited to, unwanted tachycardia with bursts of increased heart rate (HR).1–5β adrenergic blockers and calcium channel blockers are the most frequently used drugs in cats with preclinical HCM. However, the use of these drugs remains controversial because a beneficial effect on disease progression or survival has not been demonstrated with either drug.6–8 The use of bradycardic agents is predicated on their anti-ischemic properties and potential to enhance left ventricular diastolic function, control sinus tachycardia, and decrease dynamic outflow tract obstruction. However, concerns related to adverse effects of atenolol include weakness, lethargy, hypotension, and decreased LA function. These effects may limit its clinical utility.6–8
HR, a major determinant of myocardial oxygen consumption and cardiac work load, has become a novel treatment target in people with ischemic heart disease. Ivabradine is a highly selective funny current (If) inhibitor that acts directly on the sinoatrial node to induce a use- and dose-dependent decrease in HR without significant effects on inotropy, lusitropy, or dromotropy.9,10 Ivabradine has been shown to have favorable effects on cardiac output, coronary blood flow, and to decrease myocardial oxygen consumption in experimental animals.11,12
Preliminary studies in cats with IV and PO ivabradine have demonstrated hemodynamic,a pharmacokinetic,b and pharmacodynamicc effects that warrant consideration of this drug for PO use in the clinical setting. These studies indicated that ivabradine at the dose used in this study is safe, has to be administered twice daily for sufficient control of HR, and does not suppress central hemodynamics and cardiac function in healthy cats. To the authors' knowledge, no data on the clinical effects of If—inhibitors in cats are available. Accordingly, the objectives of the study reported herein were to (1) evaluate the clinical tolerability of ivabradine and (2) study the short-term effects of ivabradine on HR, blood pressure, LV function, and LA performance in healthy cats after repeated PO dosing. We hypothesized that ivabradine would not be inferior to atenolol with regard to decrease in HR, rate-pressure product (RPP), and systolic wall stress and effects on LV and LA function.
Materials and Methods
All cats were acquired from a commercial vendor.d The study protocol was reviewed and approved by the Animal Care and Use Committee and the Institutional Review Board of the Department of Veterinary Clinical Sciences, College of Veterinary Medicine, The Ohio State University.
This was a prospective noninferiority study of 2 months' duration. The design was further classified as double-blinded, randomized,e active-control, and fully crossed. Each cat was assigned to either ivabradinef (0.3 mg/kg q12h PO; group 1; n = 5) or atenololg (6.25 mg/cat q12h PO; group 2; n = 5) for the first 4 weeks and then switched to the alternative treatment for another 4 weeks without a wash-out period between treatments. The dose of ivabradine selected was based on prior pharmacologic and pharmacokinetic studies in our laboratory,c,d and the dose of atenolol evaluated was based on common clinical usage.13 Before randomization, exact doses of both drugs were prepared for each cat and subsequently filled in opaque capsulesh assuring blinding. The capsules were administered manually twice daily by the investigators (S.C.R. and R.M.C.) to assure proper drug administration. No other drug was administered during the study period.
At baseline, each cat was subjected to clinical evaluation, including a physical examination, a CBC and blood biochemistry panel, plasma T4 concentration (in cats >7 years of age), an indirect (Doppler) measurement of systolic blood pressure (SBP),i and a transthoracic echocardiogram.j Follow-up evaluations were performed at 4 and 8 weeks after inclusion. Physical examination, BP measurement, and echocardiography were performed 3 hours after drug administration when the maximal negative chronotropic effect of each drug was anticipated.c Echocardiographic recordings were labeled with random numbers selected by a third party not involved in the study, allowing subsequent offline measurements in a blinded fashion. All measurements were performed by a board-certified cardiologist (S.C.R.) within a 3-week time period after conclusion of the study using stored images assessed in random order and with the observer blinded to the cat's identification and results of previous measurements. In addition, the overall well-being of the cats including clinical variables such as appetite, behavior, activity, interaction, defecation, urination, vomiting, grooming, and respiration was monitored daily during the entire study period and quantitatively assessed.
HR, RPP, and Systolic Wall Stress
At clinical examination, heart rate (HRCE) by auscultation and SBP were determined, and the RPP calculated (RPP = SBP × HRCE).14 Left ventricular end-systolic wall stress was estimated using a cylindrical model12 as stress = 1.36 × (SBP ×D/2h), where D is the internal short-axis diameter and h is wall thickness of the LV at end-systole.
After sedation with acepromazinek (0.1 mg/kg IM) and butorphanoll (0.25 mg/kg IM), each cat underwent a transthoracic 2-dimensional (2D), M-mode, spectral and color Doppler, pulsed wave tissue Doppler imaging (TDI), and 2D-strain echocardiographic examinations. In 5 cats, the sedative effects were insufficient for quality examinations and ketaminem (1 mg/kg IM) was added. All cats received the same type and dose of the tranquilizers at each examination. Echocardiographic examinations were performed with a transducer array of 5.0–7.0 MHz nominal frequency.j A simultaneous 1-lead ECG was recorded and used for heart rate (HREcho) measurement. Data were stored digitally and echocardiographic data analysis was performed off-line by use of a commercial analysis software package.n The mean of 3 cardiac cycles was calculated for each variable measured.
Standard right and left parasternal imaging views were acquired, and 2D, M-mode, tissue Doppler, and 2D strain variables were obtained.3,15–17 LV ejection fraction (EF) was measured by use of the modified single plane Simpson's method from left apical long-axis images.18 Left atrial dimensions and area were assessed from the right parasternal long-axis view3,16 and fractional area change (FAC) and shortening fraction (LA SF) were calculated.3,16 Left atrial appendage (LAA) flow, pulmonary venous flow, and transmitral flow were recorded and quantified as recently described.3,15,16 Pulsed wave Doppler-derived velocities of myocardial motion (Peak Sa, Peak Ea, and Peak Aa) were recorded from the left apical imaging view with a sample volume of 5 mm placed at the lateral mitral annulus.19 The right parasternal midventricular short-axis plane of the LV was imaged for quantification of the peak systolic radial strain (radial SR Peak S), and peak systolic and diastolic radial strain rate (radial SrR Peak S, radial SrR Peak E, and radial SrR Peak A). Optimal frame rate was obtained by adjusting the sector width and image depth to achieve frame rates of 86–251 per second by the formula: 0.8 × HR (A. Stoylen, personal communication). Off-line analysis of 2D speckle imaging was performed as previously described in dogs.20 The LV radial strain and strain rate values reported were determined as the average of 6 corresponding myocardial segments.20 Fused or partially fused diastolic filling and wall motion waves (PW Doppler, tissue Doppler, and 2D strain) were eliminated from statistical analyses.
For comparison of the effects of ivabradine and atenolol a noninferiority study design was chosen. Application of this method offers the advantage that drug effects are not only compared statistically but also with regard to their clinical importance. By this approach, the effects of a new drug (eg, ivabradine) can only be defined as inferior or as at least equivalent to atenolol, however, not as superior. To determine that treatment with ivabradine was at least as effective (noninferior) as treatment with atenolol with regard to HRCE, RPP, systolic wall stress, and echocardiographic indices of LV and LA function, a noninferiority margin was set a priori at 50% (f= 0.5). That is, noninferiority of ivabradine compared with atenolol was defined when reduction of HRCE, RPP, and systolic wall stress was equally achieved with either drug, echocardiographic indices of the systolic LV function (fractional shortening [FS], EF, maximum aortic velocity [Ao Vmax], Radial SrR Peak S, and Radial SR Peak S) were similarly or less decreased, and indices of diastolic LV function (isovolumic relaxation time [IVRT], ratio between peak early transmitral flow velocity [E wave] to IVRT [E : IVRT], Peak S, Peak D, ratio between peak systolic [S wave] to peak diastolic [D wave] pulmonary vein flow velocity [S : D], and Radial SrR Peak E) and indices of LA function (FAC, LA SF, peak velocity of late diastolic pulmonary vein atrial reversal flow wave [Peak AR], duration of the late diastolic pulmonary vein atrial reversal flow wave [ARduration], Peak LAA flow, and Radial SrR Peak A) were equally or less affected by ivabradine. The noninferiority margin used in this study was based on predicted clinical relevance, statistical reasoning related to estimated random effects, and observer measurement variation of measurements of HR and echocardiographic indices obtained by other laboratories and experience by our own laboratory (data acquired but not shown),21–24 and is in agreement with FDA guidelines.25–27 The noninferiority margin is defined in terms of some fraction (f) of the treatment effect observed with the alternative treatment (ie, atenolol).25–27 That is, noninferiority of ivabradine compared with atenolol was present if the effects of ivabradine were at least 50% of the treatment effects observed with atenolol.
Statistical analysis was performed by use of commercially available software.o Descriptive statistics were calculated and presented as mean and standard deviation (SD) unless stated otherwise. To compare the effects of ivabradine and atenolol, repeated measures linear models (2-way repeated measures ANOVA) were used with sequence effect (group 1: ivabradine followed by atenolol; group 2: atenolol followed by ivabradine) with cat as a within-subject factor and treatment as a between-subjects factor (to identify differences between groups and treatments). Differences between treatments and baseline were compared by use of the Holm-Sidak posthoc test. If more than 4/10 observations per treatment were missing, statistical analyses were not performed and data were only reported descriptively. Bonferroni's correction was used for multiple comparisons. For all analyses, P≤ .05 was considered significant.
Animals and Clinical Tolerance
Ten healthy, female spayed, Domestic Shorthair cats were studied. The cats were 2–7 years old (mean, 4.6; SD, 1.7) and weighed 2.8–6.4 kg (mean, 4.3; SD, 1.0). In this study, the mean dose of atenolol was 1.3 mg/kg (range, 1.0–1.7 mg/kg). Both ivabradine and atenolol were well tolerated and adverse effects were not observed. Neither clinically relevant changes in body weight nor hematologic or serum biochemical abnormalities were detected with either treatment.
At baseline, cats of both groups were not different with regard to age, body weight, HR, SBP, or any other variable determined. No significant sequence effect was observed for any of the variables.
HR, RPP, and Systolic Wall Stress
Results are summarized in Table 1 and Figure 1. Ivabradine decreased HRCE and RPP significantly (P < .001), with no statistical difference between treatments (HRCE, P= .721; RPP, P= .847). Ivabradine had no significant effect on systolic wall stress compared with baseline; however, the difference between treatments was significant (P= .009). Systolic BP was not changed by ivabradine, with no difference between treatments (P= .083).
|HRCE (min−1)||204 (32)||144 (20)a||−27 (10)||135 (30)a||−34 (15)b|
|SBP (mmHg)||148 (18)||156 (16)||+5 (11)||164 (20)||+11 (14)|
|RPP (mmHg/min)||30,048 (5229)||22,616 (4859)a||−25 (16)||22,245 (6063)a||−26 (20)b|
|Systolic wall stress (g/cm2)||94.1 (17.7)||103.6 (10.6)c||+10 (11)||84.0 (15.3)c||−11 (16)|
Results are summarized in Tables 2–5.
|HREcho (min−1)||247 (33)||165 (25)a||−33 (10)||161 (28)a||−35 (11)b|
|LADd (mm)||10.4 (1.7)||10.8 (0.1)||+4 (1)||9.8 (0.1)||−6 (1)|
|LADs (mm)||13.9 (1.4)||13. 9 (0.1)||0 (1)||14.0 (0.2)||+1 (1)b|
|LA area d (cm2)||1.02 (0.31)||1.10 (0.23)||+8 (23)||1.00 (0.21)||−1 (20)|
|LA area s (cm2)||1.76 (0.32)||1.91 (0.33)||+9 (19)||1.98 (0.46)||+13(26)b|
|LVIDd (mm)||11.0 (0.9)||11.4 (1.6)||+4 (15)||12.3 (2.2)||+12 (20)b|
|LVIDs (mm)||6.8 (1.7)||7.0 (1.1)c||+3 (15)||5.2 (1.0)c||−24 (15)|
|EDV (mL)||1.89 (0.68)||2.24 (0.71)a||+19 (37)||2.80 (1.02)a||+49 (54)b|
|ESV (mL)||0.63 (0.26)||0.86 (0.28)a||+37 (45)||0.84 (0.29)a||+33 (46)b|
|FS (%)||39 (14)||38 (10)c||−3 (10)||57 (10)a,c||+46 (25)|
|EF (%)||67 (6)||61 (6)||−9 (8)||68 (11)||+1 (16)|
|Ao Vmax (m/s)||1.31 (0.65)||0.74 (0.12)a,c||−44 (9)||1.16 (0.33)c||−11 (25)|
|Radial SrR Peak S (1/s)||4.81 (1.73)||3.55 (0.84)a,c||−26 (17)||4.98 (0.97)c||+4 (20)|
|Radial SR Peak S (%)||38.5 (11.1)||48.1 (13.9)c||+25 (36)||60.2 (10.7)a,c||+56 (28)b|
|E : Ad||0.8; 0.7||2||1.2 (0.1)||+59 (16)||4||0.9 (0.2)||+23 (27)||10|
|IVRT (ms)||55 (12)||10||66 (7)a||+ 20 (12)||10||62 (7)||+13 (13)b||10|
|E : IVRT (m/s2)||15.9 (6.5)||10||9.6 (2.5)a||−40 (16)||10||10.0 (2.0)a||−37 (13)b||10|
|Peak Ea (cm/s)d||5.1; 4.6||2||4.85 (0.66)||0 (14)||4||5.79 (1.41)||+19 (29)||10|
|E : Ea latd||11.2; 9.8||2||11.5 (1.8)||+10 (17)||4||11.1 (3.3)||+6(31)||10|
|Peak S (m/s)||0.72 (0.16)||10||0.48 (0.25)a||−33 (35)||10||0.62 (0.14)||−14 (19)||10|
|Peak D (m/s)||0.35 (0.11)||10||0.44 (0.36)||+26 (100)||10||0.45 (0.07)||+29 (20)b||10|
|S/D||2.18 (0.46)||10||1.60 (0.49)a||−27 (9)||10||1.37 (0.23)a||−37 (11)b||10|
|Radial SrR Peak E (1/s)||ND||0||4.54 (1.54)||ND||7||4.74 (1.20)||ND||10|
|FAC (%)||43 (8)||10||42 (6)||−2 (14)||10||49 (6)||+14 (15)||10|
|LA SF (%)||25 (6)||10||22 (3)c||−12 (10)||10||29 (5)c||+16 (19)||10|
|Peak AR (m/s)||0.35 (0.09)||10||0.18 (0.04)a,c||−49 (11)||10||0.29 (0.09)c||−17 (26)||10|
|ARduration (ms)||59 (10)||10||58 (9)||−2 (15)||10||50 (8)||−15 (14)b||10|
|Peak LAA flow (m/s)||0.78 (0.22)||10||0.51 (0.10)a,c||−35 (13)||10||0.71 (0.14)c||−9 (18)||10|
|Radial SrR Peak A (1/s)||ND||0||2.34 (0.37)c||ND||7||3.49 (0.78)c||ND||10|
Variables of LA and LV Size. Ivabradine did not change variables of LA size significantly, and no differences between treatments were observed (Table 2). M-mode-derived indices of the LV size (left ventricular internal dimension in diastole [LVIDd] and left ventricular internal dimension in systole [LVIDs]) were not changed by ivabradine from baseline, but a significant difference for LVIDs (P= .013) between treatments was evident (Fig 1 and Table 2). Ivabradine increased estimated LV end-systolic volume (ESV; P= .005) and LV end-diastolic volume (EDV; P < .001), with no differences between treatments (ESV, P= .773; EDV, P= .097).
Variables of LV Systolic and Diastolic Function. Ivabradine increased FS (P < .001) and radial SR Peak S (P < .001) from baseline, with significant differences between treatments (FS, P < .001; radial SR Peak S, P= .025). The remaining variables of LV systolic function were not significantly altered by ivabradine (Fig 1 and Table 3); however, differences were observed between treatments for Ao Vmax (P < .008) and radial SrR Peak S (P < .003).
At baseline, 80% of the cats had fusion of transmitral flow waves, and all cats had fused diastolic waves derived by 2D strain imaging (Tables 4 and 5). Ivabradine led to separation of diastolic waves assessed by PW spectral Doppler, PW TDI, and 2D strain imaging in all cats. Atenolol decreased the frequency of fusion of E and A from 80 to 60% for both PW Doppler and PW TDI, and from 100 to 30% for 2D strain imaging.
Variables of LV relaxation. IVRT was not changed by ivabradine, with no difference between treatments. However, IVRT was significantly prolonged by atenolol (P= .006, Fig 1).
Variables of LV compliance, filling, and filling pressures. Ivabradine decreased E : IVRT (P < .003) and S : D (P < .001), with no difference between treatments (E : IVRT, P= .328; S : D, P= .234).
Variables of LA Function. Ivabradine did not change Peak AR and Peak LAA from baseline; however, significant differences were observed between treatments (Peak AR, P= .005; Peak LAA, P < .009; Table 5 and Fig 1). A significant difference (P= .012) was observed for radial SrR Peak A between treatments, with higher values observed after ivabradine compared with atenolol. Ivabradine did not change LA SF and LA FAC; however, a significant difference was observed between treatments for LA SF (P= .001, Table 5).
Other echocardiographic estimates of LV diastolic and LA systolic function (Peak E, Peak A, E : A, Peak Ea, ratio between peak velocity of early diastolic transmitral flow (E wave) to peak velocity of early diastolic motion (Ea wave) of the lateral mitral annulus [E : Ea] lat, Radial SrR Peak E, and Radial SrR Peak A) could not be statistically evaluated due to low numbers of observations owing to fusion of diastolic waves at baseline.
Ivabradine was not inferior (f≥ 0.5) to atenolol with regard to: HR, RPP, maximum left atrial antero-posterior dimension (LADs), LA area s, LVIDd, EDV, ESV, IVRT, E : IVRT, Peak D, S : D, ARduration, and radial SR Peak S, whereas other echocardiographic variables could not be evaluated because of the low number of observations.
In this study of healthy laboratory cats, noninferiority of ivabradine was observed compared with atenolol and was evident with regard to HRCE, RPP, and numerous variables of LA and LV function. The drug was clinically well tolerated and did not affect SBP. Moreover, ivabradine demonstrated more favorable effects on several echocardiographic variables, including a decrease in LV systolic wall stress and a neutral effect on indices of LV relaxation, LA function, and left auricular flow compared with baseline. This makes ivabradine potentially attractive in the treatment of feline HCM, although its effects in cats with HCM and in particular with dynamic LV outflow obstruction require investigation. Additionally, summated filling waves, a common obstacle in the assessment of LV diastolic function in cats during echocardiographic examination, was uncommon in cats treated with ivabradine making ivabradine potentially attractive as a diagnostic aid.
Active-control, noninferiority trials are performed with increasing frequency in people. Placebo-controlled trials, thought to generate high level of evidence, may be considered unethical under certain circumstances. Moreover, new drugs with similar effectiveness may offer advantages to established treatments with regard to adverse effects, dosing frequency, or costs.25–27 The critical step in determining therapeutic noninferiority is the selection of the marginal difference. Statistical reasoning and clinical judgment are commonly used to choose this margin.25–27 The selection of this margin (f= 0.5) in this study was, however, somehow arbitrary, and selection of a different value could have altered the interpretation of our findings. Future studies are needed to determine the clinically most useful noninferiority margin, and it is anticipated that this margin will not be uniform among different cardiovascular drugs and treatments.
The effects of a new drug also may fall outside the noninferiority margin; however, this does not necessarily mean that the new drug is clinically inferior to the alternative drug. The opposite effect (ie, superiority) may occur, as was seen for echocardiographic variables of LA performance after ivabradine in this study. Atenolol led to a decrease in mean LA SF of 12%, whereas ivabradine led to an increase of 16% (P= .001). Although the changes were outside the noninferiority margin, they most likely represent a favorable effect of ivabradine, as an improvement of LA performance may be of clinical benefit in cats with HCM. Similar superior effects of ivabradine compared with atenolol were observed for other variables, including LV systolic wall stress, radial SrR Peak S and SR Peak S, left atrial FAC, and Peak LAA. However, ivabradine was potentially less favorable with regard to LVIDd and LV EDV compared with atenolol, which is in accordance with previous observations from our group.a
Ivabradine and atenolol had similar negative chronotropic effects and decreased the RPP by approximately 25%. The HR-lowering effect of ivabradine observed is in accordance with previous studies in dogs and humans.11,12,28 By lowering HR, myocardial oxygen balance is potentially improved, which may prevent the development of myocardial ischemia. Evidence suggests that ischemia is a major contributor to clinical signs, disease progression, and fatal outcome and may be present even in asymptomatic human patients with HCM.4
Left ventricular systolic function was improved after ivabradine, as opposed to atenolol. The negative inotropic effect observed after atenolol is in agreement with anticipated effects of selective β receptor blockade and is a proposed beneficial mechanism in cats with HCM and obstruction of the LV outflow tract.6 Dynamic obstruction is common in cats with HCM,29 and negative inotropy is a well-known mechanism for relief of the obstruction.6–8 Whether or not ivabradine would favor obstruction in cats with HCM due to improvement of shortening fraction as observed in healthy cats cannot be answered by this study. However, because ivabradine also resulted in a significant increase of left ventricular volumes, we do not expect the improved systolic function to further deteriorate dynamic outflow obstruction as chamber enlargement may offset the effects of ivabradine on LV systolic function with regard to favoring the development of obstruction. Moreover, we demonstrated in a previous study in anesthetized cats with HCMa that ivabradine has mild negative inotropic effects as determined by direct measurement of the rate of LV pressure increase (+dP/dtmax). In the latter, echocardiographic indices such as LV SF and LV EF were not influenced by ivabradine or were even mildly increased, supporting the concept that echocardiographic variables of LV systolic function may not necessarily be good indicators of LV contractility.p,30
Diastolic function is routinely assessed in cats with cardiomyopathy.6,15,24,31 Unfortunately, fusion of diastolic filling waves may be observed in a majority of cats undergoing an echocardiographic study, making determination of LV diastolic function impossible.3,32 In this study, separated filling waves were only identified in 2 cats and were absent using 2D strain imaging at baseline in all cats. However, after treatment with ivabradine and less consistently with atenolol, E and A waves could be identified in all cats allowing for assessment of LV diastolic function. Therefore, we conclude that ivabradine may be of diagnostic benefit in the assessment of LV diastolic function in cats.
Abnormalities of LV relaxation are a hallmark of HCM, and drugs with negative lusitropic properties such as β blockers may further aggravate LV relaxation. Ivabradine did not change IVRT, an estimate of LV relaxation,33 possibly indicating no or lesser effects on lusitropy as compared with atenolol which prolonged the IVRT. The clinical relevance of this potential benefit deserves further study.
Left atrial size and function are known to be altered in cats with HCM and may contribute to the development of arterial thromboembolism (ATE).34,35 Cats with HCM have decreased LAA function as measured by LAA flow velocities and presence of ATE and spontaneous echocardiographic contrast, increased LA size, and decreased LAA flow velocities are interrelated.3,36 Therefore, any drug that increases LA size or decreases LAA flow velocity potentially may favor the development of ATE.37,38 In the present study, Peak LAA, Peak S, S : D, and Peak AR (all estimates of left atrial function39) were significantly decreased by atenolol, whereas only S : D was decreased by ivabradine. Significant differences between treatments were observed with regard to LA SF, Peak AR, and Peak LAA, indicating preserved systolic and global LA function only after ivabradine. We conclude that atenolol in contrast to ivabradine may cause deterioration of LA function. Therefore, caution may be advised if treatment with atenolol is considered in cats with significant LA pathology. Under these circumstances, ivabradine may be preferred over atenolol for HR control because it does not affect echocardiographic variables of LA function. Studies addressing the long-term effects of ivabradine on LA performance, thromboembolic risk, and overall mortality in cats with HCM are needed.
Certain limitations of this study require emphasis. Only spayed female cats were studied, all cats were healthy, and the number of cats was small. These aspects render the study underpowered to detect minor differences between groups or to evaluate drug effects in the putative clinical population. Repeatability of echocardiographic measurements was not specifically addressed, but previous data reported from our laboratory indicate acceptable reproducibility.16 Whether or not the dose of ivabradine is equipotent to the dose of atenolol administered is unknown. The effects of sedation on cardiac function, including drug interactions, could not be eliminated. A noninferiority study design was used. This approach is most useful if the comparator (ie, atenolol in this study) has been fully evaluated and is a generally accepted, evidence-based standard treatment. Finally, although no overt safety concerns were raised by this trial, drug efficacy and safety can only be assessed in an appropriately monitored clinical trial, in the target population of potential use, and with sufficient postapproval drug monitoring.
In summary, the results of this study demonstrate that ivabradine (0.3 mg/kg PO q12h) is not inferior to atenolol (6.25 mg PO q12h) in healthy cats with regard to negative chronotropy, reduction of RPP, and clinical tolerance after 4 weeks of treatment. Moreover, ivabradine seems to exert more beneficial effects on LV systolic wall stress, LV diastolic function, LA performance, and LAA velocity compared with atenolol. The increase in LV systolic function and LV EDV observed with ivabradine deserves further evaluation. Studies in cats with HCM, in particular those with dynamic LV outflow tract obstruction, are needed to further elucidate the potential value of ivabradine in clinical practice.
a Riesen SC, Schober KE, Smith DN, Otoni CC, Bonagura JD. Effects of ivabradine on invasive indices of LV function in anesthetized cats with hypertrophic cardiomyopathy. J Vet Intern Med 2010;24:692–693 (abstract)
b Riesen SC, Schober KE, Lindsey KJ, Carnes CA, Ni W, Phelps MA. Pharmacokinetics of oral ivabradine in healthy cats. J Vet Intern Med 2010;24:732 (abstract)
c Cober RE, Schober KE, Buffington CAT, Riesen SC, Bonagura JD. Effects of ivabradine, a selective If channel inhibitor, on heart rate in healthy cats. J Vet Intern Med 2010;24:695 (abstract)
d Liberty Research Inc, Waverly, NY
e ClinPro software, Clinical Systems, Garden City, NY
f Procoralan, Les Laboratoires Servier, 22 Rue Garnier, 92200 Neuilly-sur-Seine, France
g Atenolol, Mallinckrodt Inc, St Louis, MO
h Capsules 4 blue, Gallipot Inc, St Paul, MN
i Ultrasonic Doppler Flow Detector 811-AL, Parks Medical Electronics Inc, Aloha, OR
j Vivid 7 Vantage, GE Medical Systems, Milwaukee, WI
k Acepromazine maleate injection, Boehringer Ingelheim Vetmedica Inc, St Joseph, MO
l Torbugesic, Fort Dodge Laboratories, Fort Dodge, IA
m Ketaset, Fort Dodge Laboratories
n EchoPac software package, Version BT06, GE Medical Systems
o SigmaStat, Version 3.5, SPSS Inc, Chicago, IL
p Schober KE, Bonagura JD, Luis-Fuentes V, Hatfield D. Invasive validation of Doppler echocardiographic indices of left ventricular systolic function in healthy cats. J Vet Intern Med 2006;20:1534–1535 (abstract)
The authors gratefully acknowledge Sara Zaldivar, Danielle N. Smith, Kathleen J. Lindsey, Patricia T Mueller and Laura J Spayed for their contributions.
Financial disclosure or funding: This study was supported by a graduate student grant from Boehringer Ingelheim.
- 17Recommendations for standards in transthoracic two-dimensional echocardiography in the dog and cat. Echocardiography Committee of the Specialty of Cardiology, American College of Veterinary Internal Medicine. J Vet Intern Med 1993;7:247–252., , , et al.
- 25Food and Drug Administration, HHS. International Conference on Harmonisation. Choice of control group and related issues in clinical trials; availability notice. Fed Regist 2001;66:24390–24391.
- 34Arterial thromboembolism in cats: Acute crisis in 127 cases (1992–2001) and long-term management with low-dose aspirin in 24 cases. J Vet Intern Med 2003;17:73–83., , , et al.