- Email: [email protected]
Heart rate and arrhythmia frequency of normal cats compared to cats with asymptomatic hypertrophic cardiomyopathy
Journal of Veterinary Cardiology (2014) 16, 215e225
www.elsevier.com/locate/jvc
Heart rate and arrhythmia frequency of normal cats compared to cats with asymptomatic hypertrophic cardiomyopathy Bethany L. Jackson, DVM*, Linda B. Lehmkuhl, DVM, MS , Darcy B. Adin, DVM MedVet Medical & Cancer Centers for Pets, 300 East Wilson Bridge Rd, Worthington, OH 43085, USA Received 4 May 2014; received in revised form 17 September 2014; accepted 6 October 2014
KEYWORDS Feline; Holter monitor; Ventricular; Supraventricular; HCM
Abstract Objectives: To compare heart rate and arrhythmia frequency and complexity in a normal population of cats to a population of cats with hypertrophic cardiomyopathy (HCM). Animals: 17 cats with HCM and 15 cats with normal echocardiograms. Methods: Results for echocardiography, electrocardiography, Doppler blood pressure, and 24-h Holter monitoring were compared between groups. Results: There was no difference in heart rate between HCM cats and normal cats regardless of modality used. All (17/17) HCM cats had ventricular arrhythmias (geometric mean 124 complexes/24 h) with 82% (14/17) exhibiting complex arrhythmias (couplets, triplets, or ventricular tachycardia). Most (14/15) normal cats had ventricular arrhythmias (geometric mean 4 complexes/24 h), but only 20% (3/15) exhibited complexity. HCM cats had significantly more total ventricular complexes, ventricular premature complexes and accelerated idioventricular rhythm than normal cats (P < 0.0001, P < 0.0001, and P ¼ 0.01, respectively). Eighty eight percent (15/17) of HCM cats had supraventricular arrhythmias (geometric mean 9 complexes/24 h) with 23% (4/17) exhibiting complexity. Sixty percent (9/15) of normal cats had supraventricular arrhythmias (geometric mean 1 complex/24 h) with 13% (2/15) exhibiting complexity. Cats with hypertrophic cardiomyopathy had significantly more supraventricular complexes than normal cats (P ¼ 0.0148).
* Corresponding author. E-mail address: [email protected] (B.L. Jackson). http://dx.doi.org/10.1016/j.jvc.2014.10.001 1760-2734/ª 2014 Elsevier B.V. All rights reserved.
216
B.L. Jackson et al. Conclusion: Cats with asymptomatic HCM have more frequent and complex ventricular and supraventricular arrhythmias than normal cats but do not have different overall heart rates compared to normal cats. Further studies are needed to determine if these arrhythmias are associated with an increased risk of sudden cardiac death or influence long-term survival. ª 2014 Elsevier B.V. All rights reserved.
Abbreviations 2-D AIVR 24 AVE HR AVE HR AUSC HR BP ECG ECG HR HR IVSd IVSs LA LA:AO LAX LVPWd LVPWS LVIDd LVIDs LVOT 24 MIN HR MIN HR M-mode 24 MAX HR MAX HR SAM SAX SCD TotSV TotVent VPCs
two-dimensional mode of echocardiography accelerated idioventricular rhythm 24-h average heart rate obtained by Holter monitor average heart rate obtained during a single hour of Holter monitoring heart rate obtained by auscultation blood pressure electrocardiogram heart rate obtained by ECG heart rate interventricular septum measured in diastole interventricular septum measured in systole left atrium left atrium-to-aorta ratio measured from the long axis left ventricular posterior wall measured in diastole left ventricular posterior wall measured in systole left ventricular internal dimension in diastole left ventricular internal dimension in systole left ventricular outflow tract 24-h minimum heart rate obtained by Holter monitor minimum heart rate obtained during a single hour of Holter monitoring motion mode echocardiography 24 h maximum heart rate obtained by Holter monitor maximum heart rate obtained during a single hour of Holter monitoring systolic anterior motion measured from the short axis sudden cardiac death total number of supraventricular complexes (premature, accelerated, and escape) total number of ventricular complexes (premature, accelerated, and escape) ventricular premature complexes
Introduction Hypertrophic cardiomyopathy (HCM) is the most common cardiac disease in cats. Many cats remain asymptomatic for this disease for several years; however, potential clinical outcomes include congestive heart failure, thromboembolism, or sudden cardiac death (SCD). The frequency of arrhythmias in normal cats has been described.1,2 Additionally, there is one report describing Holter data and arrhythmias in cats with
various cardiomyopathies and congestive heart failure.a In-hospital electrocardiograms (ECG) are commonly performed in clinical practice if an arrhythmia is ausculted. One study of cats with HCM showed an arrhythmia prevalence of 13% (8/ 61 cats) using this modality.3 Holter monitoring is a Fox PR, Moise NS, Price RA, Petrie JP, Bond BR. Analysis of continuous ECG (Holter) in normal cats and cardiomyopathic cats in congestive heart failure (abstract). J Vet Intern Med 1998; 12:199.
Arrhythmias in HCM cats expected to be more sensitive for arrhythmia detection4; however, Holter data in asymptomatic cats with HCM is not available. Therefore, the importance of arrhythmias in this population may be underestimated. In studies in humans, 88% of people with HCM had ventricular arrhythmias detected by Holter monitoring and 31% had ventricular tachycardia.5 Associations have been found in people between the presence of nonsustained ventricular tachycardia and risk for SCD.6 Similar arrhythmia frequencies and risk for SCD may be present in cats with HCM. The purpose of this study was to compare the 24-h arrhythmia frequency and complexity in cats with asymptomatic HCM and normal cats. A secondary aim of the study was to compare in-clinic auscultation and ECG to 24-h Holter monitor measures of heart rate and arrhythmia frequency. We hypothesized that cats with HCM have more frequent and complex arrhythmias than normal cats, and that 24-h Holter monitoring will detect more arrhythmias in cats than in-clinic ECG.
Materials and methods Animals Clinically normal cats owned by hospital employees of MedVet Medical & Cancer Centers for Pets and client-owned cats with HCM were screened for enrollment in the study between July 2009 and October 2011. HCM cats were obtained from a population of client-owned cats referred for cardiac ultrasound from their primary care veterinarians. All owners and employees provided informed consent prior to participation. Cats were included in the normal group if they were greater than 1 year of age, had a normal physical exam (a cardiac murmur was permitted if determined to be functional on echocardiographic examination), had systolic blood pressure obtained by Doppler (BP) < 180 mmHg, were euthyroid (total T4 < 2.2 mg/dL if >7 years old), and had a normal echocardiogram (left ventricular diastolic wall thickness <5.5 mm on 2-D and Mmode). Cats were included in the HCM group if they had echocardiographic evidence of diffuse or segmental hypertrophy (left ventricular diastolic wall thickness >6 mm on M-mode or 2-D)10 in the absence of congenital cardiac disease. Cats in the HCM group had a BP <180 mmHg and were euthyroid (T4 < 2.2 mg/dL). Cats that were less than 1 year of age, required sedation, had clinical signs attributable to cardiac disease, or that
217 received steroids or cardiac medications within the previous 30 days were excluded.
Diagnostic evaluation Heart rate (HR) was obtained by a trained technician who ausculted each cat prior to removal from the carrier. The HR was obtained by counting heart beats for 15 s and multiplying by 4 to obtain a per minute HR (AUSC HR). A 6-lead ECG was recorded at 50 mm/s for standard interval measurements. A rhythm strip (leads I, II, III) was recorded from cats in right lateral recumbency at 25 mm/s for 2 min to obtain in-clinic heart rhythm. The HR calculated from the ECG (ECG HR) was averaged over 6 s during sinus rhythm. Blood pressure measurementsb were performed by a trained technician on a front limb with cats in lateral recumbency as recommended in the ACVIM consensus statement.8 Transthoracic two-dimensional (2-D), M-mode, and Doppler (color flow, pulsed, and continuous wave) echocardiography was performed in right and left lateral recumbency using a 12-MHz or 8MHz probe.c Measurements included left ventricular dimensions and wall thicknesses in diastole and systole from the left ventricular short axis directly below the level of the mitral valve (interventricular septum in diastole (IVSd) and systole (IVSs), left ventricular posterior wall in diastole (LVPWd) and systole (LVPWs), left ventricular internal diameter in diastole (LVIDd) and systole (LVIDs) measured from M-mode and 2D),7,9,10 left atrial measurements and left atriumto-aorta ratio (LA:Ao) in short axis by 2-D and Mmode.10 The mitral valve was examined for systolic anterior motion (SAM) using M-mode and 2-D.7,11 The left ventricular outflow tract (LVOT) velocity profile and aortic velocity profile were obtained with pulsed and continuous wave Doppler from the left apical 5-chamber view. Left ventricular outflow tract obstruction was defined as a late systolic accelerating Doppler velocity profile with a velocity >2.5 m/s in conjunction with the presence of mitral valve SAM.7,11
Holter monitor A Holter monitord was placed by shaving the sternum and right and left ventral aspects of the
b Ultrasonic Doppler Flow Detector, Parks Medical, Unit 811-B, Aloha, Oregon. c Phillips iE-33, Software level 5.2.0.289.
218 thorax, and attaching 5 electrodes to the skin to obtain two ECG channels for 24 h. A light wrap was placed over the thorax, and the cat returned to the home environment. The Holter monitor was placed at the time of the clinical appointment between the hours of 9 AM and 6 PM. Because the Holter monitor softwared did not accurately determine HR in cats, HR analysis was performed using the emka software systeme. The lead with the least artifact was exported from the Holter system as a .txt file and imported to the emka software system. Holter recordings were included if at least 85% of the complexes could be analyzed. Hourly HR analysis was performed to determine an average (AVE HR), minimum (MIN HR), and maximum HR (MAX HR) for each hour during Holter monitoring. Average HR was calculated as the average of the ReR intervals, MIN HR as the lowest 10% of the ReR intervals, and MAX HR as the highest 10% of the ReR intervals for each hour. A 24-h average HR (24 AVE HR) was calculated as the average of the 24 hourly heart rate averages. The 24-h minimum HR (24 MIN HR) was the minimum HR that occurred during the hour with the lowest MIN HR; whereas, the 24-h maximum HR (24 MAX HR) was the maximum heart rate that occurred during the hour with the highest MAX HR. Rhythm analysis was performed by the Holter system softwaree and confirmed by manual review of the full disclosure. All abnormal complexes that were not detected by the Holter system were hand coded. Ventricular complexes were classified as premature (instantaneous HR 160 bpm), idioventricular (instantaneous HR 100e159 bpm), or ventricular escapes (instantaneous HR 99 bpm). Supraventricular complexes were classified as premature (instantaneous HR 160 bpm), idiojunctional (instantaneous HR 120e159 bpm), or junctional escapes (instantaneous HR 119 bpm). Ventricular arrhythmia complexity was compared using a complexity scale: 0 ¼ no arrhythmias, 1 ¼ only single premature complexes, 2 ¼ couplets, triplets, or ventricular tachycardia. Owners were asked two questions regarding their cats experience wearing the Holter monitor. Question 1 was “How much did wearing the Holter monitor affect your cat’s day-to-day activity?” with answer options of not affected, mildly affected, moderately affected, or severely affected. Question 2 was “If your cardiologist d Trillium Platinum Holter Analysis Software, version 02.12/ 04.25, Forest Medical, LLC, East Syracuse, NY. e emka ECG auto software, version 2.5.1.31, emka Technologies, Paris, France.
B.L. Jackson et al. recommended another Holter monitor for this cat or another cat, would you agree to it?” with answer options of yes, no, or maybe.
Statistical analysis Statistical analysis was performed by standard computer statistical software.f Statistical significance was set at P < 0.05. Data were tested for normality using the ShapiroeWilk test. Outcomes meeting the assumption of normality were presented as mean and standard deviation, and comparisons between groups were made using a non-paired Student’s t-test. Where values were deemed to follow a log-normal distribution, results are presented as geometric means. Results not following a normal distribution are summarized using medians and ranges, and between groups comparisons were made using Wilcoxon’s rank sum test. Group age and weight were compared using a t-test. The proportion of genders within group was compared using Fisher’s exact test. M-mode measurements for Ao, LA, IVSs, LVIDd, LVIDs, LVPWd, LVPWs, and 2-D measurements for aorta in short-axis (Ao SAX), left atrium in short-axis (LA SAX), left atrium in long-axis (LA LAX), IVSd, IVSs, LVIDd, LVIDs, and LVPWs were compared by t-test, and are presented as mean and standard deviation. The M-mode measurements for LA:Ao, IVSd, and the 2-D measurements for LA:Ao SAX, LVPWd, and LVOT Vmax were compared with a rank sum test, and are presented as median and range. The QT interval and R wave height were compared using a t-test, and are presented as mean and standard deviation. The P wave duration, PR and QRS intervals, and P wave height were compared using a rank sum test, and are presented as median and range. The AUSC HR, ECG HR, and Holterderived 24 AVE HR, 24 MIN HR, and 24 MAX HR were compared between normal and affected cats using a t-test. HR data were presented as mean and standard deviation. The pooled HR values for all cats obtained by the different modalities (AUSC HR, ECG HR, and the 24 AVE HR) were compared using repeated measures ANOVA, given the results from the different modalities were obtained from the same animal. Bias between AUSC HR and 24 AVE HR and between ECG HR and 24 AVE HR was assessed by Bland Altman analysis.12 Hourly HR data obtained by Holter monitoring of normal and affected cats were compared between groups and over time relative to time of day using repeated measures ANOVA with 8 AM as the reference point.
f
SAS/STAT 12.1, Carey, NC.
Arrhythmias in HCM cats Group, time and the interaction between group and time were included in the statistical model as fixed effects. Where the interaction was significant, within time group differences were evaluated; otherwise the main effect of group was assessed. Hourly HR data obtained by Holter monitoring of normal and affected cats was also compared between groups and over time relative to time post Holter monitor placement using repeated measures ANOVA with hour 1 as the reference point. The model described above was used in this comparison. The total ventricular complexes (TotVent), number of ventricular premature complexes (VPCs), and total supraventricular complexes (TotSV) detected during the 24-h Holter monitor were compared between normal and affected cats by t-test on log transformed data. Accelerated idioventricular rhythm (AIVR) and arrhythmia complexity between normal and affected were assessed by a rank sum test. The frequency of arrhythmias (TotVent, VPCs, and AIVR) in HCM cats with LVOT obstruction was compared to HCM cats without LVOT obstruction by t-test. Correlation between LVOT velocity and arrhythmia frequency was evaluated with linear regression on log-transformed data. Arrhythmia data are presented as geometric means when log normally distributed and as medians when not normally distributed.
Results The normal population consisted of 15 cats (13 castrated males, 2 spayed females) with a mean age of 6.0 years (2.8 years) and a mean weight of 5.9 kg (1.19 kg). The HCM population consisted of 17 cats (14 castrated males, 3 spayed females) with a mean age of 7.8 years (3.5 years) and a mean weight of 5.4 kg (1.32 kg). Eleven of 17 HCM cats had mitral valve SAM and secondary LVOT obstruction. No significant differences were noted between normal and HCM cats for age, sex, or weight (P ¼ 0.12, 0.27, 0.35, respectively). HCM cats had thicker ventricular septal and posterior wall measurements, and smaller internal left ventricular dimensions than HCM cats (both by M-mode and 2-D measurements). HCM cats had higher LVOT velocities than normal cats. There was no difference in left atrial or aortic size between HCM and normal cats (Table 1). Overall, the HCM cats represent a mildly to moderately affected population. The P wave, PR, QRS, and QT interval durations and P wave amplitude were not different between normal and HCM groups. The R wave amplitude
219 was greater in HCM cats compared to normal cats (Table 2). There was no difference in HR between normal and HCM cats by AUSC HR, ECG HR, 24 AVE HR, 24 MIN HR, or 24 MAX HR (Table 3). Heart rate obtained by AUSC and HR obtained by ECG was higher than 24 AVE HR obtained by Holter monitoring but the HR measured by AUSC and by ECG was not different from each other (P ¼ 0.0532). The in-clinic ECG HR exceeded the 24 AVE HR obtained by Holter monitoring by 30 bpm (bias) with limits of agreement of 49 bpm (Fig. 1). The in-clinic AUSC HR exceeded the 24 AVE HR obtained by Holter monitoring by 21 bpm (bias) with limits of agreement 54 bpm (Fig. 2). Hourly HR data of normal and HCM cats were compared between groups and over time with 8 AM being the reference point. There was no significant group by time interaction for AVE HR, MIN HR, or MAX HR (P ¼ 0.06, 0.14, 0.22, respectively). Additionally, the main effect of group was not significant when evaluated across time for AVE HR, MIN HR, or MAX HR (P ¼ 0.96, 0.84, 0.56, respectively). There were small statistical differences for HR at certain times of the day compared to 8 AM in both normal and HCM cats (P < 0.0001 for AVE HR, MIN HR, and MAX HR) as detailed in Fig. 3. There was no discernible diurnal pattern to these differences, although HRs tended to be higher midday. Hourly HR data of normal and affected cats were compared between groups and over time to the 1st hour post-Holter monitor placement (Fig. 4). The group by time interaction was significant for the AVE HR (P ¼ 0.03); however, within a given hour, there were no significant group effects. There was no significant group by time interaction for MIN HR or MAX HR (P ¼ 0.58, 0.23 respectively). Additionally, the main effect of group was not significant for MIN HR or MAX HR (P ¼ 0.84, 0.56 respectively). There were statistical differences in HR at certain times postplacement (P < 0.0001 for AVE HR, MIN HR, MAX HR) compared to hour 1 for both normal and HCM cats as detailed in Fig. 4. For both groups, the heart rate tended to be higher during the first 1e3 h post placement compared to subsequent hours. All HCM cats had ventricular arrhythmias and 14/15 normal cats had ventricular arrhythmias. HCM cats had significantly more TotVent, VPCs, and AIVR than normal cats (P < 0.0001, P < 0.0001, and P ¼ 0.01, respectively) (Table 4). There was no difference in TotVent, VPCs, or AIVR between obstructive and non-obstructive HCM cats (P ¼ 0.88, P ¼ 0.13, P ¼ 0.67, respectively) (Table 5). There was no significant correlation
220
B.L. Jackson et al.
Table 1
Echocardiographic measurements.
Echo Variable (mm)
Normal
HCM
P-value
Ao MMa LA MMa LA:Ao MMb IVSd MMb IVSs MMa LVIDd MMa LVIDs MMa LVPWd MMa LVPWs MMa Ao SAX 2Da LA SAX 2Da LA:Ao SAX 2-Db LA LAX 2-Da IVSd 2-Da IVSs 2-Da LVIDd 2-Da LVIDs 2-Da LVPWd 2-Db LVPWs 2-Da LVOT Vmax (m/s)b
9.8 0.8 13.1 1.3 1.3 (1.1e1.37) 4.0 (3.4e5.0) 7.56 0.74 15.64 1.98 7.58 2.14 4.22 0.53 7.63 1.21 9.87 1.02 12.72 1.10 1.3 (1.08e1.48) 15.07 1.13 4.35 0.65 6.41 0.84 15.21 2.46 8.33 1.70 4.5 (3.5e5.5) 7.1 1.07 0.78 (0.62e1.13)
10.2 1.1 13.4 1.0 1.3 (1.1e1.8) 6.3 (4.8e7.7) 9.14 1.09 14.08 1.6 6.0 1.36 5.74 1.0 9.37 1.17 10.3 1.08 13.3 1.1 1.26 (1.08e2.0) 15.3 1.27 6.17 1.09 8.03 1.02 13.49 1.96 7.20 1.13 5.7 (4.4e6.9) 8.1 1.11 3.48 (0.89e5.71)
0.27 0.80 0.63 <0.001 <0.001 0.02 0.02 <0.001 <0.001 0.25 0.15 0.61 0.59 <0.001 <0.001 0.04 0.03 0.002 0.01 <0.001
mm e millimeter, MM e measured from M-mode, Ao e aorta, LA e left atrium, LA:Ao e ratio between the left atrium and aorta, IVSd e interventricular septum in diastole, IVSs e interventricular septum in systole, LVIDd e left ventricular internal dimension in diastole, LVIDs e left ventricular internal dimension in systole, LVPWd e left ventricular posterior wall in diastole, LVPWs e left ventricular posterior wall in systole, 2-D e measured from 2D, SAX e short axis, LAX e long axis, LVOT Vmax e maximum velocity measured in the left ventricular outflow tract. Values in bold are statistically significant. a mean standard deviation. b median (interquartile range).
detected between left ventricular outflow tract velocity and frequency of TotVent or VPCs (P ¼ 0.6, P ¼ 0.4, respectively). An accurate assessment of the effect of time of day or effect of time post-Holter monitor placement on arrhythmias could not be made because of the low hourly number of arrhythmias. HCM cats had significantly more complex arrhythmias (median complexity grade 2) than normal cats (median complexity grade 1) (P ¼ 0.0005). Of the normal cats, 3/15 had grade 2
Table 2
Table 3
ECG measurements.
Measurement
Normal
HCM
P-value
a
30 (20e40) 70 (60e80) 20 (20e30) 157 15.8 0.1 (0.1e0.2) 0.4 0.2
30 (20e40) 60 (60e80) 20 (20e40) 156 19.1 0.2 (0.1e0.3) 0.6 0.3
0.84 0.58 0.57 0.82 0.54 0.02
P (ms) PR (ms)a QRS (ms)a QT (ms)b P (mV)a R (mV)b
complexity, 11/15 had grade 1 complexity, and 1/15 had grade 0 complexity. Of the HCM cats, 14/17 had grade 2 complexity, and 3/17 had grade 1 complexity. Eighty-eight percent (15/17) of HCM cats had supraventricular arrhythmias (geometric mean 9 complexes/24 h). Sixty percent (9/15) of normal cats had supraventricular arrhythmias (geometric mean 1 complex/24 h). HCM cats had significantly more supraventricular complexes than normal cats (P ¼ 0.0148).
ms e milliseconds; P e P duration; PR e PR interval; QRS e QRS duration; QT e QT interval, mV e millivolts; R e R amplitude. Values in bold are statistically significant. a median (interquartile range). b mean standard deviation.
Average heart rate measurements.
Modality AUSC HR ECG HR 24 AVE HR 24 MIN HR 24 MAX HR
HR normal (bpm) 192 194 170 134 233
22 26 25 28 21
HR HCM (bpm)
P-value
0.82 0.26 0.96 0.99 0.16
191 206 170 133 246
23 34 25 33 28
Data is presented as mean standard deviation. AUSC HR e heart rate by auscultation, ECG HR e heart rate by in-clinic ECG, 24 AVE HR e average heart rate over 24 h by Holter, 24 MIN HR e minimum heart rate over 24 h by Holter, 24 MAX HR e maximum heart rate over 24 h by Holter.
Arrhythmias in HCM cats
221
Discussion
Figure 1 Bland-Altman graph depicting bias between ECG HR and 24 AVE HR. The data are pooled for all cats (n ¼ 32).
All HCM cats (17/17) had arrhythmias detected during Holter monitoring, while only 18% (3/17) had arrhythmias detected during in-clinic ECG recording. All normal cats (15/15) also had arrhythmias detected during Holter monitoring while none had arrhythmias detected during inclinic ECG recording. Over one-half of owners reported that wearing a Holter monitor moderately (8/31) or severely (11/ 31) affected their cat’s daily activity. Eleven out of 31 owners reported that wearing the Holter monitor only mildly affected their cat’s daily activity and one owner did not feel the Holter monitor affected her cat’s activity. Most owners (26/31) said they were willing to have a Holter monitor placed on their cat again; whereas, 2 owners would not and 3 were undecided.
Figure 2 Bland-Altman graph depicting bias between AUSC HR and 24 AVE HR. The data are pooled for all cats (n ¼ 32).
This study evaluated the heart rate, arrhythmia frequency and arrhythmia complexity of normal cats compared to a population of asymptomatic HCM cats. To the authors’ knowledge, there are no previous published studies evaluating heart rate and arrhythmia frequency by Holter analysis in a group of mild to moderately affected asymptomatic HCM cats. The average HR during Holter monitor recordings for normal cats was 170 25 bpm, which is higher than that reported by Ware (157 3.7 bpm)1 and Fox (151 bpm),a but is similar to a report by Hana ˚s et al. (168 bpm).2 One of the limitations encountered in our study was that the Holter monitor software programe did not consistently identify the QRS complexes for many cats, and therefore the Holter-derived calculated HR was inaccurate, and artifactually low. This issue was addressed by exporting the data into a second systeme that was able to accurately calculate the HR by identifying the QRS complexes. In the study by Hana ˚s,2 the authors calculated a semi-manual HR over 7 s every 30 min, and this may explain why the values in our study are similar to those reported by Hana ˚s,2 but higher than other studies. We did not detect a difference in Holter monitor derived HR between normal and affected cats. This is similar to what was found by Fox et al,a who compared Holter recordings of normal cats to cats with different cardiomyopathies and congestive heart failure in the hospital. That study found no significant difference between average and minimum heart rates, but a significantly higher maximum heart rate in the cats with congestive heart failure. This difference could be due to the asymptomatic stage of disease and the at-home environment for Holter monitoring in our study, both of which could result in lower sympathetic tone and lower heart rates. Our finding that cats with mild to moderate asymptomatic HCM do not have higher heart rates than normal cats could suggest that this population of cats does not have an elevation in sympathetic tone; although, this was not specifically investigated in this study. Alternatively, these results could suggest that Holter monitoring creates the same amount of stress and sympathetic tone elevation in both normal and mildly to moderately affected asymptomatic HCM cats. There was no difference in hourly HRs between normal and HCM cats based on either the time of day or the time of the Holter placement. The time
222
B.L. Jackson et al.
Figure 3 Hourly HR by Holter, shown by time of day. There were no differences within hour between groups or between groups across time for AVE HR, MIN HR, or MAX HR (P ¼ 0.96, 0.84, 0.56, respectively). There were differences within a group at some hours compared to 8 AM. * ¼ time when HR was different than 8 AM for HCM cats. ^¼ time when HR was different from 8 AM for normal cats. a) MIN HRs e Times 12:00, 13:00, 16:00, and 18:00 were different for HCM cats. No times were different from 8 AM for normal cats. b) AVE HRs -Times 12:00, 13:00, 16:00, and 18:00 were different for HCM cats. Times 15:00, 19:00, 20:00, and 21:00 were different for normal cats. c) MAX HRs e Time 12:00 was different for HCM cats. Times 15:00, 17:00, 18:00, 19:00, 20:00, and 21:00 were different for normal cats.
of day that the Holter was placed was not standardized in this study to accommodate clinic flow and patient wait times. However, as there was no difference between the groups based on the time of day, this likely did not impact the results significantly. Additionally, no significant diurnal variation in heart rate was noted in this study. Finally, the HR was higher the first 1e3 h after placement of the Holter monitor, which likely represents the time the cat was still in the clinic or driving home and in an acclimation period to the device. This is similar to findings in a previous study that found that HR in clinic by ambulatory ECG was higher than HR at home by ambulatory ECG13 in normal cats. This study detected differences in HR obtained by auscultation and in-clinic ECG compared to HR obtained by 24-h Holter monitoring. The in-clinic ECG HR overestimated the average HR obtained
from the Holter monitor by 30 bpm, and the AUSC HR overestimated the average Holter HR by 21 bpm. This most likely can be explained by the stress of the hospital visit and the stress of having the ECG performed. We tried to minimize this effect as much as possible by performing auscultation before the cat was removed from the carrier. However, there was no statistical difference between HR obtained by auscultation and in-clinic ECG, so the stress of the hospital visit likely played a large role in the higher HR obtained compared to HR obtained during the Holter monitor recording, which was performed in the home environment for the majority of the recording. This is similar to what was found by Abbott,13 who showed that the stress of echocardiography produced a higher HR than unrestrained ambulatory ECG monitoring in the clinic, and unrestrained ambulatory ECG in the clinic showed higher HRs
Arrhythmias in HCM cats
223
Figure 4 Hourly HR by Holter, shown by time post placement. There were no differences between groups within hours or between groups across time for AVE HR, MIN HR, or MAX HR (P ¼ 0.98, 0.84, 0.56, respectively). There were differences within a group at some hours compared to hour 1 (placement hour). y ¼ time when HR was not different than hour 1 for HCM cats. # ¼ time when HR was not different than hour 1 for normal cats. Time that HRs were not different from hour 1 represent times of higher HRs and presumed stress of Holter monitor after placement. a) MIN HRs e During every hour post placement MIN HR was different from hour 1 for HCM cats. Hours 3, 4 were not different from hour 1 for normal cats. b) AVE HRs e During every hour post placement AVE HR was different from hour 1 for HCM cats. Hours 3, 4 were not different from hour 1 for normal cats. c) MAX HRs e Hours 2, 3,4,5,8 were not different from hour 1 for HCM cats. Hour 2 was not different from hour 1 for normal cats.
than HRs obtained at home by ambulatory ECG monitoring. Similar to previous reports, this study found that normal cats have relatively low numbers of ventricular arrhythmias over a 24-h period,1,2 and HCM cats have more frequent and severe arrhythmias.a The numbers of abnormal beats and level of complexity is much lower in our study compared to a previous study.a This is not unexpected, because
Table 4
the cats in our study had mild to moderate disease compared to severe disease associated with congestive heart failure in the prior study. The in-clinic ECG did not accurately predict the arrhythmia frequency when compared to a 24-h Holter. All cats had some abnormal complexes noted on their Holter recording, although few cats had evidence of arrhythmias on the in-clinic ECG. This is similar to what has been found in studies in
Ventricular arrhythmia frequency.
Rhythm a
TotVent (# complexes) VPCsa (# complexes) AIVRb (# complexes)
Normal
HCM
P-value
4 4 0 (range 0e0)
124 61 0 (range 0e2254)
<0.0001 <0.0001 0.0126
TotVent e total ventricular complexes, VPCs e ventricular premature complexes, AIVR e accelerated idioventricular rhythm. TotVent and VPCs presented as geometric means, AIVR is presented as median and range. a Values were log transformed prior to the statistical analysis. Back transformed values (geometic means) are presented. b Median values (interquartile range) are presented.
224 Table 5
B.L. Jackson et al. Arrhythmia frequency in obstructive vs. non-obstructive HCM cats.
Rhythm TotVent (# complexes) VPCs (# complexes) AIVR (# complexes)
Obstructive HCM
Non-obstructive HCM
P-value
364 140 225
425 49 376
0.88 0.13 0.67
TotVent e total ventricular complexes, VPCs e ventricular premature complexes, AIVR e accelerated idioventricular rhythm.
dogs.4,14 The small number of cats enrolled in this study and low overall arrhythmia counts did not allow accurate evaluation of predictive values or specificity and sensitivity of arrhythmia frequency of in-clinic ECG compared to 24-h Holter. Eleven of 17 (64%) cats in our study had mitral valve SAM and secondary LVOT obstruction, which is similar to the prevalence of 67% previously reported.7 We found that having an obstructive form of HCM did not influence the number of ventricular arrhythmias. Additionally, there was no correlation between severity of obstruction and the presence of arrhythmias. Although the overall arrhythmia numbers were low, given the level of obstruction observed, we may have seen a difference if obstruction was a significant factor in frequency or complexity of arrhythmias. Most owners perceived that wearing the Holter monitor negatively affected their cat’s activity level. This may indicate that some degree of stress persisted during home HR/rhythm assessment. Nevertheless, based on lower at home Holter HRs, stress levels at home appear lower than in the clinic. This study has several limitations. First, the inclusion of only asymptomatic, mildly to moderately affected HCM cats makes extrapolation of these findings to cats with more severe disease or those with clinical signs not possible. Second, the time of day for Holter application was not standardized, making assessment of diurnal variation of heart rate and arrhythmias difficult. Finally, the study included a small population of cats, and different results may have been obtained with a larger population.
Conclusion In this study, cats with mild to moderate asymptomatic HCM had more frequent and complex ventricular and supraventricular arrhythmias than normal cats, but no difference in overall HR, compared to normal cats. Further studies are needed to determine if these arrhythmias are associated with an increased risk of sudden cardiac death or influence long-term survival.
Conflict of interest None.
Acknowledgments Funding for this study was provided by the ACVIM Cardiology Resident Research Grant and MedVet Medical & Cancer Centers for Pets. The authors would like to thank QTest Labs in Columbus, OH for use and training with the emka ECG auto software system. We would like to thank Dr. Thaibinh Nguyenba for submission of cases. We would like to thank Dr. Steven Radecki for assistance with statistical analysis.
References 1. Ware W. Twenty-four-hour ambulatory ECG in normal cats. J Vet Intern Med 1999;13:175e180. 2. Hanas S, Tidholm A, Egenvall A, Holst BS. Twenty-four hour Holter monitoring of unsedated healthy cats in the home environment. J Vet Cardiol 2009;11:17e22. 3. Ferasin L, Sturgess CP, Cannon MJ, Caney SM, GruffyddJones TJ, Wotton PR. Feline idiopathic cardiomyopathy: a retrospective study of 106 cats (1994-2001). J Feline Med Surg 2003;5:151e159. 4. Miller RH, Lehmkuhl LB, Bonagura JD, Beall MJ. Retrospective analysis of the clinical utility of ambulatory electrocardiographic (Holter) recordings in syncopal dogs: 44 cases (1991e1995). J Vet Intern Med 1999;2:111e122. 5. Adabag AS, Casey SA, Kuskowski MA, Zenovich AG, Maron BJ. Spectrum and prognostic significance of arrhythmias on ambulatory Holter electrocardiogram in hypertrophic cardiomyopathy. J Am Coll Cardiol 2005;45: 697e704. 6. Monserrat L, Elliott PM, Gimeno JR, Sharma S, PenasLado M, McKenna WJ. Non-sustained ventricular tachycardia in hypertrophic cardiomyopathy: an independent marker of sudden death risk in young patients. J Am Coll Cardiol 2003;42:873e879. 7. Fox PR, Liu SK, Maron BJ. Echocardiographic assessment of spontaneously occurring feline hypertrophic cardiomyopathy: an animal model of human disease. Circulation 1995; 92:2645e2651. 8. Brown S, Atkins C, Bagley R, Carr A, Cowgill L, Davidson M, Egner B, Elliott J, Henik R, Labato M, Littman M, Polzin D, Ross L, Snyder P, Stepien R. Guidelines for the identification, evaluation, and management
Arrhythmias in HCM cats
225
of systemic hypertension in dogs and cats. J Vet Intern Med 2007;21:542e558. 9. Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072e1083. 10. Abbott JA, MacLean HN. Two dimensional echocardiographic assessment of the feline left atrium. J Vet Intern Med 2006;20:111e119. 11. Schober K, Todd A. Echocardiographic assessment of left ventricular geometry and the mitral valve apparatus in cats with hypertrophic cardiomyopathy. J Vet Cardiol 2010;12: 1e16.
12. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1985;1:307e310. 13. Abbott JA. Heart rate and heart rate variability of healthy cats in home and hospital environments. J Feline Med Surg 2005;7:195e202. 14. Wess G, Schulze A, Geraghty N, Hartman K. Ability of 5minute electrocardiography (ECG) for predicting arrhythmias in Doberman Pinschers with cardiomyopathy in comparison with 24-hour ambulatory ECG. J Vet Intern Med 2010;24:367e371.
Available online at www.sciencedirect.com
ScienceDirect
www.elsevier.com/locate/jvc
Heart rate and arrhythmia frequency of normal cats compared to cats with asymptomatic hypertrophic cardiomyopathy Bethany L. Jackson, DVM*, Linda B. Lehmkuhl, DVM, MS , Darcy B. Adin, DVM MedVet Medical & Cancer Centers for Pets, 300 East Wilson Bridge Rd, Worthington, OH 43085, USA Received 4 May 2014; received in revised form 17 September 2014; accepted 6 October 2014
KEYWORDS Feline; Holter monitor; Ventricular; Supraventricular; HCM
Abstract Objectives: To compare heart rate and arrhythmia frequency and complexity in a normal population of cats to a population of cats with hypertrophic cardiomyopathy (HCM). Animals: 17 cats with HCM and 15 cats with normal echocardiograms. Methods: Results for echocardiography, electrocardiography, Doppler blood pressure, and 24-h Holter monitoring were compared between groups. Results: There was no difference in heart rate between HCM cats and normal cats regardless of modality used. All (17/17) HCM cats had ventricular arrhythmias (geometric mean 124 complexes/24 h) with 82% (14/17) exhibiting complex arrhythmias (couplets, triplets, or ventricular tachycardia). Most (14/15) normal cats had ventricular arrhythmias (geometric mean 4 complexes/24 h), but only 20% (3/15) exhibited complexity. HCM cats had significantly more total ventricular complexes, ventricular premature complexes and accelerated idioventricular rhythm than normal cats (P < 0.0001, P < 0.0001, and P ¼ 0.01, respectively). Eighty eight percent (15/17) of HCM cats had supraventricular arrhythmias (geometric mean 9 complexes/24 h) with 23% (4/17) exhibiting complexity. Sixty percent (9/15) of normal cats had supraventricular arrhythmias (geometric mean 1 complex/24 h) with 13% (2/15) exhibiting complexity. Cats with hypertrophic cardiomyopathy had significantly more supraventricular complexes than normal cats (P ¼ 0.0148).
* Corresponding author. E-mail address: [email protected] (B.L. Jackson). http://dx.doi.org/10.1016/j.jvc.2014.10.001 1760-2734/ª 2014 Elsevier B.V. All rights reserved.
216
B.L. Jackson et al. Conclusion: Cats with asymptomatic HCM have more frequent and complex ventricular and supraventricular arrhythmias than normal cats but do not have different overall heart rates compared to normal cats. Further studies are needed to determine if these arrhythmias are associated with an increased risk of sudden cardiac death or influence long-term survival. ª 2014 Elsevier B.V. All rights reserved.
Abbreviations 2-D AIVR 24 AVE HR AVE HR AUSC HR BP ECG ECG HR HR IVSd IVSs LA LA:AO LAX LVPWd LVPWS LVIDd LVIDs LVOT 24 MIN HR MIN HR M-mode 24 MAX HR MAX HR SAM SAX SCD TotSV TotVent VPCs
two-dimensional mode of echocardiography accelerated idioventricular rhythm 24-h average heart rate obtained by Holter monitor average heart rate obtained during a single hour of Holter monitoring heart rate obtained by auscultation blood pressure electrocardiogram heart rate obtained by ECG heart rate interventricular septum measured in diastole interventricular septum measured in systole left atrium left atrium-to-aorta ratio measured from the long axis left ventricular posterior wall measured in diastole left ventricular posterior wall measured in systole left ventricular internal dimension in diastole left ventricular internal dimension in systole left ventricular outflow tract 24-h minimum heart rate obtained by Holter monitor minimum heart rate obtained during a single hour of Holter monitoring motion mode echocardiography 24 h maximum heart rate obtained by Holter monitor maximum heart rate obtained during a single hour of Holter monitoring systolic anterior motion measured from the short axis sudden cardiac death total number of supraventricular complexes (premature, accelerated, and escape) total number of ventricular complexes (premature, accelerated, and escape) ventricular premature complexes
Introduction Hypertrophic cardiomyopathy (HCM) is the most common cardiac disease in cats. Many cats remain asymptomatic for this disease for several years; however, potential clinical outcomes include congestive heart failure, thromboembolism, or sudden cardiac death (SCD). The frequency of arrhythmias in normal cats has been described.1,2 Additionally, there is one report describing Holter data and arrhythmias in cats with
various cardiomyopathies and congestive heart failure.a In-hospital electrocardiograms (ECG) are commonly performed in clinical practice if an arrhythmia is ausculted. One study of cats with HCM showed an arrhythmia prevalence of 13% (8/ 61 cats) using this modality.3 Holter monitoring is a Fox PR, Moise NS, Price RA, Petrie JP, Bond BR. Analysis of continuous ECG (Holter) in normal cats and cardiomyopathic cats in congestive heart failure (abstract). J Vet Intern Med 1998; 12:199.
Arrhythmias in HCM cats expected to be more sensitive for arrhythmia detection4; however, Holter data in asymptomatic cats with HCM is not available. Therefore, the importance of arrhythmias in this population may be underestimated. In studies in humans, 88% of people with HCM had ventricular arrhythmias detected by Holter monitoring and 31% had ventricular tachycardia.5 Associations have been found in people between the presence of nonsustained ventricular tachycardia and risk for SCD.6 Similar arrhythmia frequencies and risk for SCD may be present in cats with HCM. The purpose of this study was to compare the 24-h arrhythmia frequency and complexity in cats with asymptomatic HCM and normal cats. A secondary aim of the study was to compare in-clinic auscultation and ECG to 24-h Holter monitor measures of heart rate and arrhythmia frequency. We hypothesized that cats with HCM have more frequent and complex arrhythmias than normal cats, and that 24-h Holter monitoring will detect more arrhythmias in cats than in-clinic ECG.
Materials and methods Animals Clinically normal cats owned by hospital employees of MedVet Medical & Cancer Centers for Pets and client-owned cats with HCM were screened for enrollment in the study between July 2009 and October 2011. HCM cats were obtained from a population of client-owned cats referred for cardiac ultrasound from their primary care veterinarians. All owners and employees provided informed consent prior to participation. Cats were included in the normal group if they were greater than 1 year of age, had a normal physical exam (a cardiac murmur was permitted if determined to be functional on echocardiographic examination), had systolic blood pressure obtained by Doppler (BP) < 180 mmHg, were euthyroid (total T4 < 2.2 mg/dL if >7 years old), and had a normal echocardiogram (left ventricular diastolic wall thickness <5.5 mm on 2-D and Mmode). Cats were included in the HCM group if they had echocardiographic evidence of diffuse or segmental hypertrophy (left ventricular diastolic wall thickness >6 mm on M-mode or 2-D)10 in the absence of congenital cardiac disease. Cats in the HCM group had a BP <180 mmHg and were euthyroid (T4 < 2.2 mg/dL). Cats that were less than 1 year of age, required sedation, had clinical signs attributable to cardiac disease, or that
217 received steroids or cardiac medications within the previous 30 days were excluded.
Diagnostic evaluation Heart rate (HR) was obtained by a trained technician who ausculted each cat prior to removal from the carrier. The HR was obtained by counting heart beats for 15 s and multiplying by 4 to obtain a per minute HR (AUSC HR). A 6-lead ECG was recorded at 50 mm/s for standard interval measurements. A rhythm strip (leads I, II, III) was recorded from cats in right lateral recumbency at 25 mm/s for 2 min to obtain in-clinic heart rhythm. The HR calculated from the ECG (ECG HR) was averaged over 6 s during sinus rhythm. Blood pressure measurementsb were performed by a trained technician on a front limb with cats in lateral recumbency as recommended in the ACVIM consensus statement.8 Transthoracic two-dimensional (2-D), M-mode, and Doppler (color flow, pulsed, and continuous wave) echocardiography was performed in right and left lateral recumbency using a 12-MHz or 8MHz probe.c Measurements included left ventricular dimensions and wall thicknesses in diastole and systole from the left ventricular short axis directly below the level of the mitral valve (interventricular septum in diastole (IVSd) and systole (IVSs), left ventricular posterior wall in diastole (LVPWd) and systole (LVPWs), left ventricular internal diameter in diastole (LVIDd) and systole (LVIDs) measured from M-mode and 2D),7,9,10 left atrial measurements and left atriumto-aorta ratio (LA:Ao) in short axis by 2-D and Mmode.10 The mitral valve was examined for systolic anterior motion (SAM) using M-mode and 2-D.7,11 The left ventricular outflow tract (LVOT) velocity profile and aortic velocity profile were obtained with pulsed and continuous wave Doppler from the left apical 5-chamber view. Left ventricular outflow tract obstruction was defined as a late systolic accelerating Doppler velocity profile with a velocity >2.5 m/s in conjunction with the presence of mitral valve SAM.7,11
Holter monitor A Holter monitord was placed by shaving the sternum and right and left ventral aspects of the
b Ultrasonic Doppler Flow Detector, Parks Medical, Unit 811-B, Aloha, Oregon. c Phillips iE-33, Software level 5.2.0.289.
218 thorax, and attaching 5 electrodes to the skin to obtain two ECG channels for 24 h. A light wrap was placed over the thorax, and the cat returned to the home environment. The Holter monitor was placed at the time of the clinical appointment between the hours of 9 AM and 6 PM. Because the Holter monitor softwared did not accurately determine HR in cats, HR analysis was performed using the emka software systeme. The lead with the least artifact was exported from the Holter system as a .txt file and imported to the emka software system. Holter recordings were included if at least 85% of the complexes could be analyzed. Hourly HR analysis was performed to determine an average (AVE HR), minimum (MIN HR), and maximum HR (MAX HR) for each hour during Holter monitoring. Average HR was calculated as the average of the ReR intervals, MIN HR as the lowest 10% of the ReR intervals, and MAX HR as the highest 10% of the ReR intervals for each hour. A 24-h average HR (24 AVE HR) was calculated as the average of the 24 hourly heart rate averages. The 24-h minimum HR (24 MIN HR) was the minimum HR that occurred during the hour with the lowest MIN HR; whereas, the 24-h maximum HR (24 MAX HR) was the maximum heart rate that occurred during the hour with the highest MAX HR. Rhythm analysis was performed by the Holter system softwaree and confirmed by manual review of the full disclosure. All abnormal complexes that were not detected by the Holter system were hand coded. Ventricular complexes were classified as premature (instantaneous HR 160 bpm), idioventricular (instantaneous HR 100e159 bpm), or ventricular escapes (instantaneous HR 99 bpm). Supraventricular complexes were classified as premature (instantaneous HR 160 bpm), idiojunctional (instantaneous HR 120e159 bpm), or junctional escapes (instantaneous HR 119 bpm). Ventricular arrhythmia complexity was compared using a complexity scale: 0 ¼ no arrhythmias, 1 ¼ only single premature complexes, 2 ¼ couplets, triplets, or ventricular tachycardia. Owners were asked two questions regarding their cats experience wearing the Holter monitor. Question 1 was “How much did wearing the Holter monitor affect your cat’s day-to-day activity?” with answer options of not affected, mildly affected, moderately affected, or severely affected. Question 2 was “If your cardiologist d Trillium Platinum Holter Analysis Software, version 02.12/ 04.25, Forest Medical, LLC, East Syracuse, NY. e emka ECG auto software, version 2.5.1.31, emka Technologies, Paris, France.
B.L. Jackson et al. recommended another Holter monitor for this cat or another cat, would you agree to it?” with answer options of yes, no, or maybe.
Statistical analysis Statistical analysis was performed by standard computer statistical software.f Statistical significance was set at P < 0.05. Data were tested for normality using the ShapiroeWilk test. Outcomes meeting the assumption of normality were presented as mean and standard deviation, and comparisons between groups were made using a non-paired Student’s t-test. Where values were deemed to follow a log-normal distribution, results are presented as geometric means. Results not following a normal distribution are summarized using medians and ranges, and between groups comparisons were made using Wilcoxon’s rank sum test. Group age and weight were compared using a t-test. The proportion of genders within group was compared using Fisher’s exact test. M-mode measurements for Ao, LA, IVSs, LVIDd, LVIDs, LVPWd, LVPWs, and 2-D measurements for aorta in short-axis (Ao SAX), left atrium in short-axis (LA SAX), left atrium in long-axis (LA LAX), IVSd, IVSs, LVIDd, LVIDs, and LVPWs were compared by t-test, and are presented as mean and standard deviation. The M-mode measurements for LA:Ao, IVSd, and the 2-D measurements for LA:Ao SAX, LVPWd, and LVOT Vmax were compared with a rank sum test, and are presented as median and range. The QT interval and R wave height were compared using a t-test, and are presented as mean and standard deviation. The P wave duration, PR and QRS intervals, and P wave height were compared using a rank sum test, and are presented as median and range. The AUSC HR, ECG HR, and Holterderived 24 AVE HR, 24 MIN HR, and 24 MAX HR were compared between normal and affected cats using a t-test. HR data were presented as mean and standard deviation. The pooled HR values for all cats obtained by the different modalities (AUSC HR, ECG HR, and the 24 AVE HR) were compared using repeated measures ANOVA, given the results from the different modalities were obtained from the same animal. Bias between AUSC HR and 24 AVE HR and between ECG HR and 24 AVE HR was assessed by Bland Altman analysis.12 Hourly HR data obtained by Holter monitoring of normal and affected cats were compared between groups and over time relative to time of day using repeated measures ANOVA with 8 AM as the reference point.
f
SAS/STAT 12.1, Carey, NC.
Arrhythmias in HCM cats Group, time and the interaction between group and time were included in the statistical model as fixed effects. Where the interaction was significant, within time group differences were evaluated; otherwise the main effect of group was assessed. Hourly HR data obtained by Holter monitoring of normal and affected cats was also compared between groups and over time relative to time post Holter monitor placement using repeated measures ANOVA with hour 1 as the reference point. The model described above was used in this comparison. The total ventricular complexes (TotVent), number of ventricular premature complexes (VPCs), and total supraventricular complexes (TotSV) detected during the 24-h Holter monitor were compared between normal and affected cats by t-test on log transformed data. Accelerated idioventricular rhythm (AIVR) and arrhythmia complexity between normal and affected were assessed by a rank sum test. The frequency of arrhythmias (TotVent, VPCs, and AIVR) in HCM cats with LVOT obstruction was compared to HCM cats without LVOT obstruction by t-test. Correlation between LVOT velocity and arrhythmia frequency was evaluated with linear regression on log-transformed data. Arrhythmia data are presented as geometric means when log normally distributed and as medians when not normally distributed.
Results The normal population consisted of 15 cats (13 castrated males, 2 spayed females) with a mean age of 6.0 years (2.8 years) and a mean weight of 5.9 kg (1.19 kg). The HCM population consisted of 17 cats (14 castrated males, 3 spayed females) with a mean age of 7.8 years (3.5 years) and a mean weight of 5.4 kg (1.32 kg). Eleven of 17 HCM cats had mitral valve SAM and secondary LVOT obstruction. No significant differences were noted between normal and HCM cats for age, sex, or weight (P ¼ 0.12, 0.27, 0.35, respectively). HCM cats had thicker ventricular septal and posterior wall measurements, and smaller internal left ventricular dimensions than HCM cats (both by M-mode and 2-D measurements). HCM cats had higher LVOT velocities than normal cats. There was no difference in left atrial or aortic size between HCM and normal cats (Table 1). Overall, the HCM cats represent a mildly to moderately affected population. The P wave, PR, QRS, and QT interval durations and P wave amplitude were not different between normal and HCM groups. The R wave amplitude
219 was greater in HCM cats compared to normal cats (Table 2). There was no difference in HR between normal and HCM cats by AUSC HR, ECG HR, 24 AVE HR, 24 MIN HR, or 24 MAX HR (Table 3). Heart rate obtained by AUSC and HR obtained by ECG was higher than 24 AVE HR obtained by Holter monitoring but the HR measured by AUSC and by ECG was not different from each other (P ¼ 0.0532). The in-clinic ECG HR exceeded the 24 AVE HR obtained by Holter monitoring by 30 bpm (bias) with limits of agreement of 49 bpm (Fig. 1). The in-clinic AUSC HR exceeded the 24 AVE HR obtained by Holter monitoring by 21 bpm (bias) with limits of agreement 54 bpm (Fig. 2). Hourly HR data of normal and HCM cats were compared between groups and over time with 8 AM being the reference point. There was no significant group by time interaction for AVE HR, MIN HR, or MAX HR (P ¼ 0.06, 0.14, 0.22, respectively). Additionally, the main effect of group was not significant when evaluated across time for AVE HR, MIN HR, or MAX HR (P ¼ 0.96, 0.84, 0.56, respectively). There were small statistical differences for HR at certain times of the day compared to 8 AM in both normal and HCM cats (P < 0.0001 for AVE HR, MIN HR, and MAX HR) as detailed in Fig. 3. There was no discernible diurnal pattern to these differences, although HRs tended to be higher midday. Hourly HR data of normal and affected cats were compared between groups and over time to the 1st hour post-Holter monitor placement (Fig. 4). The group by time interaction was significant for the AVE HR (P ¼ 0.03); however, within a given hour, there were no significant group effects. There was no significant group by time interaction for MIN HR or MAX HR (P ¼ 0.58, 0.23 respectively). Additionally, the main effect of group was not significant for MIN HR or MAX HR (P ¼ 0.84, 0.56 respectively). There were statistical differences in HR at certain times postplacement (P < 0.0001 for AVE HR, MIN HR, MAX HR) compared to hour 1 for both normal and HCM cats as detailed in Fig. 4. For both groups, the heart rate tended to be higher during the first 1e3 h post placement compared to subsequent hours. All HCM cats had ventricular arrhythmias and 14/15 normal cats had ventricular arrhythmias. HCM cats had significantly more TotVent, VPCs, and AIVR than normal cats (P < 0.0001, P < 0.0001, and P ¼ 0.01, respectively) (Table 4). There was no difference in TotVent, VPCs, or AIVR between obstructive and non-obstructive HCM cats (P ¼ 0.88, P ¼ 0.13, P ¼ 0.67, respectively) (Table 5). There was no significant correlation
220
B.L. Jackson et al.
Table 1
Echocardiographic measurements.
Echo Variable (mm)
Normal
HCM
P-value
Ao MMa LA MMa LA:Ao MMb IVSd MMb IVSs MMa LVIDd MMa LVIDs MMa LVPWd MMa LVPWs MMa Ao SAX 2Da LA SAX 2Da LA:Ao SAX 2-Db LA LAX 2-Da IVSd 2-Da IVSs 2-Da LVIDd 2-Da LVIDs 2-Da LVPWd 2-Db LVPWs 2-Da LVOT Vmax (m/s)b
9.8 0.8 13.1 1.3 1.3 (1.1e1.37) 4.0 (3.4e5.0) 7.56 0.74 15.64 1.98 7.58 2.14 4.22 0.53 7.63 1.21 9.87 1.02 12.72 1.10 1.3 (1.08e1.48) 15.07 1.13 4.35 0.65 6.41 0.84 15.21 2.46 8.33 1.70 4.5 (3.5e5.5) 7.1 1.07 0.78 (0.62e1.13)
10.2 1.1 13.4 1.0 1.3 (1.1e1.8) 6.3 (4.8e7.7) 9.14 1.09 14.08 1.6 6.0 1.36 5.74 1.0 9.37 1.17 10.3 1.08 13.3 1.1 1.26 (1.08e2.0) 15.3 1.27 6.17 1.09 8.03 1.02 13.49 1.96 7.20 1.13 5.7 (4.4e6.9) 8.1 1.11 3.48 (0.89e5.71)
0.27 0.80 0.63 <0.001 <0.001 0.02 0.02 <0.001 <0.001 0.25 0.15 0.61 0.59 <0.001 <0.001 0.04 0.03 0.002 0.01 <0.001
mm e millimeter, MM e measured from M-mode, Ao e aorta, LA e left atrium, LA:Ao e ratio between the left atrium and aorta, IVSd e interventricular septum in diastole, IVSs e interventricular septum in systole, LVIDd e left ventricular internal dimension in diastole, LVIDs e left ventricular internal dimension in systole, LVPWd e left ventricular posterior wall in diastole, LVPWs e left ventricular posterior wall in systole, 2-D e measured from 2D, SAX e short axis, LAX e long axis, LVOT Vmax e maximum velocity measured in the left ventricular outflow tract. Values in bold are statistically significant. a mean standard deviation. b median (interquartile range).
detected between left ventricular outflow tract velocity and frequency of TotVent or VPCs (P ¼ 0.6, P ¼ 0.4, respectively). An accurate assessment of the effect of time of day or effect of time post-Holter monitor placement on arrhythmias could not be made because of the low hourly number of arrhythmias. HCM cats had significantly more complex arrhythmias (median complexity grade 2) than normal cats (median complexity grade 1) (P ¼ 0.0005). Of the normal cats, 3/15 had grade 2
Table 2
Table 3
ECG measurements.
Measurement
Normal
HCM
P-value
a
30 (20e40) 70 (60e80) 20 (20e30) 157 15.8 0.1 (0.1e0.2) 0.4 0.2
30 (20e40) 60 (60e80) 20 (20e40) 156 19.1 0.2 (0.1e0.3) 0.6 0.3
0.84 0.58 0.57 0.82 0.54 0.02
P (ms) PR (ms)a QRS (ms)a QT (ms)b P (mV)a R (mV)b
complexity, 11/15 had grade 1 complexity, and 1/15 had grade 0 complexity. Of the HCM cats, 14/17 had grade 2 complexity, and 3/17 had grade 1 complexity. Eighty-eight percent (15/17) of HCM cats had supraventricular arrhythmias (geometric mean 9 complexes/24 h). Sixty percent (9/15) of normal cats had supraventricular arrhythmias (geometric mean 1 complex/24 h). HCM cats had significantly more supraventricular complexes than normal cats (P ¼ 0.0148).
ms e milliseconds; P e P duration; PR e PR interval; QRS e QRS duration; QT e QT interval, mV e millivolts; R e R amplitude. Values in bold are statistically significant. a median (interquartile range). b mean standard deviation.
Average heart rate measurements.
Modality AUSC HR ECG HR 24 AVE HR 24 MIN HR 24 MAX HR
HR normal (bpm) 192 194 170 134 233
22 26 25 28 21
HR HCM (bpm)
P-value
0.82 0.26 0.96 0.99 0.16
191 206 170 133 246
23 34 25 33 28
Data is presented as mean standard deviation. AUSC HR e heart rate by auscultation, ECG HR e heart rate by in-clinic ECG, 24 AVE HR e average heart rate over 24 h by Holter, 24 MIN HR e minimum heart rate over 24 h by Holter, 24 MAX HR e maximum heart rate over 24 h by Holter.
Arrhythmias in HCM cats
221
Discussion
Figure 1 Bland-Altman graph depicting bias between ECG HR and 24 AVE HR. The data are pooled for all cats (n ¼ 32).
All HCM cats (17/17) had arrhythmias detected during Holter monitoring, while only 18% (3/17) had arrhythmias detected during in-clinic ECG recording. All normal cats (15/15) also had arrhythmias detected during Holter monitoring while none had arrhythmias detected during inclinic ECG recording. Over one-half of owners reported that wearing a Holter monitor moderately (8/31) or severely (11/ 31) affected their cat’s daily activity. Eleven out of 31 owners reported that wearing the Holter monitor only mildly affected their cat’s daily activity and one owner did not feel the Holter monitor affected her cat’s activity. Most owners (26/31) said they were willing to have a Holter monitor placed on their cat again; whereas, 2 owners would not and 3 were undecided.
Figure 2 Bland-Altman graph depicting bias between AUSC HR and 24 AVE HR. The data are pooled for all cats (n ¼ 32).
This study evaluated the heart rate, arrhythmia frequency and arrhythmia complexity of normal cats compared to a population of asymptomatic HCM cats. To the authors’ knowledge, there are no previous published studies evaluating heart rate and arrhythmia frequency by Holter analysis in a group of mild to moderately affected asymptomatic HCM cats. The average HR during Holter monitor recordings for normal cats was 170 25 bpm, which is higher than that reported by Ware (157 3.7 bpm)1 and Fox (151 bpm),a but is similar to a report by Hana ˚s et al. (168 bpm).2 One of the limitations encountered in our study was that the Holter monitor software programe did not consistently identify the QRS complexes for many cats, and therefore the Holter-derived calculated HR was inaccurate, and artifactually low. This issue was addressed by exporting the data into a second systeme that was able to accurately calculate the HR by identifying the QRS complexes. In the study by Hana ˚s,2 the authors calculated a semi-manual HR over 7 s every 30 min, and this may explain why the values in our study are similar to those reported by Hana ˚s,2 but higher than other studies. We did not detect a difference in Holter monitor derived HR between normal and affected cats. This is similar to what was found by Fox et al,a who compared Holter recordings of normal cats to cats with different cardiomyopathies and congestive heart failure in the hospital. That study found no significant difference between average and minimum heart rates, but a significantly higher maximum heart rate in the cats with congestive heart failure. This difference could be due to the asymptomatic stage of disease and the at-home environment for Holter monitoring in our study, both of which could result in lower sympathetic tone and lower heart rates. Our finding that cats with mild to moderate asymptomatic HCM do not have higher heart rates than normal cats could suggest that this population of cats does not have an elevation in sympathetic tone; although, this was not specifically investigated in this study. Alternatively, these results could suggest that Holter monitoring creates the same amount of stress and sympathetic tone elevation in both normal and mildly to moderately affected asymptomatic HCM cats. There was no difference in hourly HRs between normal and HCM cats based on either the time of day or the time of the Holter placement. The time
222
B.L. Jackson et al.
Figure 3 Hourly HR by Holter, shown by time of day. There were no differences within hour between groups or between groups across time for AVE HR, MIN HR, or MAX HR (P ¼ 0.96, 0.84, 0.56, respectively). There were differences within a group at some hours compared to 8 AM. * ¼ time when HR was different than 8 AM for HCM cats. ^¼ time when HR was different from 8 AM for normal cats. a) MIN HRs e Times 12:00, 13:00, 16:00, and 18:00 were different for HCM cats. No times were different from 8 AM for normal cats. b) AVE HRs -Times 12:00, 13:00, 16:00, and 18:00 were different for HCM cats. Times 15:00, 19:00, 20:00, and 21:00 were different for normal cats. c) MAX HRs e Time 12:00 was different for HCM cats. Times 15:00, 17:00, 18:00, 19:00, 20:00, and 21:00 were different for normal cats.
of day that the Holter was placed was not standardized in this study to accommodate clinic flow and patient wait times. However, as there was no difference between the groups based on the time of day, this likely did not impact the results significantly. Additionally, no significant diurnal variation in heart rate was noted in this study. Finally, the HR was higher the first 1e3 h after placement of the Holter monitor, which likely represents the time the cat was still in the clinic or driving home and in an acclimation period to the device. This is similar to findings in a previous study that found that HR in clinic by ambulatory ECG was higher than HR at home by ambulatory ECG13 in normal cats. This study detected differences in HR obtained by auscultation and in-clinic ECG compared to HR obtained by 24-h Holter monitoring. The in-clinic ECG HR overestimated the average HR obtained
from the Holter monitor by 30 bpm, and the AUSC HR overestimated the average Holter HR by 21 bpm. This most likely can be explained by the stress of the hospital visit and the stress of having the ECG performed. We tried to minimize this effect as much as possible by performing auscultation before the cat was removed from the carrier. However, there was no statistical difference between HR obtained by auscultation and in-clinic ECG, so the stress of the hospital visit likely played a large role in the higher HR obtained compared to HR obtained during the Holter monitor recording, which was performed in the home environment for the majority of the recording. This is similar to what was found by Abbott,13 who showed that the stress of echocardiography produced a higher HR than unrestrained ambulatory ECG monitoring in the clinic, and unrestrained ambulatory ECG in the clinic showed higher HRs
Arrhythmias in HCM cats
223
Figure 4 Hourly HR by Holter, shown by time post placement. There were no differences between groups within hours or between groups across time for AVE HR, MIN HR, or MAX HR (P ¼ 0.98, 0.84, 0.56, respectively). There were differences within a group at some hours compared to hour 1 (placement hour). y ¼ time when HR was not different than hour 1 for HCM cats. # ¼ time when HR was not different than hour 1 for normal cats. Time that HRs were not different from hour 1 represent times of higher HRs and presumed stress of Holter monitor after placement. a) MIN HRs e During every hour post placement MIN HR was different from hour 1 for HCM cats. Hours 3, 4 were not different from hour 1 for normal cats. b) AVE HRs e During every hour post placement AVE HR was different from hour 1 for HCM cats. Hours 3, 4 were not different from hour 1 for normal cats. c) MAX HRs e Hours 2, 3,4,5,8 were not different from hour 1 for HCM cats. Hour 2 was not different from hour 1 for normal cats.
than HRs obtained at home by ambulatory ECG monitoring. Similar to previous reports, this study found that normal cats have relatively low numbers of ventricular arrhythmias over a 24-h period,1,2 and HCM cats have more frequent and severe arrhythmias.a The numbers of abnormal beats and level of complexity is much lower in our study compared to a previous study.a This is not unexpected, because
Table 4
the cats in our study had mild to moderate disease compared to severe disease associated with congestive heart failure in the prior study. The in-clinic ECG did not accurately predict the arrhythmia frequency when compared to a 24-h Holter. All cats had some abnormal complexes noted on their Holter recording, although few cats had evidence of arrhythmias on the in-clinic ECG. This is similar to what has been found in studies in
Ventricular arrhythmia frequency.
Rhythm a
TotVent (# complexes) VPCsa (# complexes) AIVRb (# complexes)
Normal
HCM
P-value
4 4 0 (range 0e0)
124 61 0 (range 0e2254)
<0.0001 <0.0001 0.0126
TotVent e total ventricular complexes, VPCs e ventricular premature complexes, AIVR e accelerated idioventricular rhythm. TotVent and VPCs presented as geometric means, AIVR is presented as median and range. a Values were log transformed prior to the statistical analysis. Back transformed values (geometic means) are presented. b Median values (interquartile range) are presented.
224 Table 5
B.L. Jackson et al. Arrhythmia frequency in obstructive vs. non-obstructive HCM cats.
Rhythm TotVent (# complexes) VPCs (# complexes) AIVR (# complexes)
Obstructive HCM
Non-obstructive HCM
P-value
364 140 225
425 49 376
0.88 0.13 0.67
TotVent e total ventricular complexes, VPCs e ventricular premature complexes, AIVR e accelerated idioventricular rhythm.
dogs.4,14 The small number of cats enrolled in this study and low overall arrhythmia counts did not allow accurate evaluation of predictive values or specificity and sensitivity of arrhythmia frequency of in-clinic ECG compared to 24-h Holter. Eleven of 17 (64%) cats in our study had mitral valve SAM and secondary LVOT obstruction, which is similar to the prevalence of 67% previously reported.7 We found that having an obstructive form of HCM did not influence the number of ventricular arrhythmias. Additionally, there was no correlation between severity of obstruction and the presence of arrhythmias. Although the overall arrhythmia numbers were low, given the level of obstruction observed, we may have seen a difference if obstruction was a significant factor in frequency or complexity of arrhythmias. Most owners perceived that wearing the Holter monitor negatively affected their cat’s activity level. This may indicate that some degree of stress persisted during home HR/rhythm assessment. Nevertheless, based on lower at home Holter HRs, stress levels at home appear lower than in the clinic. This study has several limitations. First, the inclusion of only asymptomatic, mildly to moderately affected HCM cats makes extrapolation of these findings to cats with more severe disease or those with clinical signs not possible. Second, the time of day for Holter application was not standardized, making assessment of diurnal variation of heart rate and arrhythmias difficult. Finally, the study included a small population of cats, and different results may have been obtained with a larger population.
Conclusion In this study, cats with mild to moderate asymptomatic HCM had more frequent and complex ventricular and supraventricular arrhythmias than normal cats, but no difference in overall HR, compared to normal cats. Further studies are needed to determine if these arrhythmias are associated with an increased risk of sudden cardiac death or influence long-term survival.
Conflict of interest None.
Acknowledgments Funding for this study was provided by the ACVIM Cardiology Resident Research Grant and MedVet Medical & Cancer Centers for Pets. The authors would like to thank QTest Labs in Columbus, OH for use and training with the emka ECG auto software system. We would like to thank Dr. Thaibinh Nguyenba for submission of cases. We would like to thank Dr. Steven Radecki for assistance with statistical analysis.
References 1. Ware W. Twenty-four-hour ambulatory ECG in normal cats. J Vet Intern Med 1999;13:175e180. 2. Hanas S, Tidholm A, Egenvall A, Holst BS. Twenty-four hour Holter monitoring of unsedated healthy cats in the home environment. J Vet Cardiol 2009;11:17e22. 3. Ferasin L, Sturgess CP, Cannon MJ, Caney SM, GruffyddJones TJ, Wotton PR. Feline idiopathic cardiomyopathy: a retrospective study of 106 cats (1994-2001). J Feline Med Surg 2003;5:151e159. 4. Miller RH, Lehmkuhl LB, Bonagura JD, Beall MJ. Retrospective analysis of the clinical utility of ambulatory electrocardiographic (Holter) recordings in syncopal dogs: 44 cases (1991e1995). J Vet Intern Med 1999;2:111e122. 5. Adabag AS, Casey SA, Kuskowski MA, Zenovich AG, Maron BJ. Spectrum and prognostic significance of arrhythmias on ambulatory Holter electrocardiogram in hypertrophic cardiomyopathy. J Am Coll Cardiol 2005;45: 697e704. 6. Monserrat L, Elliott PM, Gimeno JR, Sharma S, PenasLado M, McKenna WJ. Non-sustained ventricular tachycardia in hypertrophic cardiomyopathy: an independent marker of sudden death risk in young patients. J Am Coll Cardiol 2003;42:873e879. 7. Fox PR, Liu SK, Maron BJ. Echocardiographic assessment of spontaneously occurring feline hypertrophic cardiomyopathy: an animal model of human disease. Circulation 1995; 92:2645e2651. 8. Brown S, Atkins C, Bagley R, Carr A, Cowgill L, Davidson M, Egner B, Elliott J, Henik R, Labato M, Littman M, Polzin D, Ross L, Snyder P, Stepien R. Guidelines for the identification, evaluation, and management
Arrhythmias in HCM cats
225
of systemic hypertension in dogs and cats. J Vet Intern Med 2007;21:542e558. 9. Sahn DJ, DeMaria A, Kisslo J, Weyman A. Recommendations regarding quantitation in M-mode echocardiography: results of a survey of echocardiographic measurements. Circulation 1978;58:1072e1083. 10. Abbott JA, MacLean HN. Two dimensional echocardiographic assessment of the feline left atrium. J Vet Intern Med 2006;20:111e119. 11. Schober K, Todd A. Echocardiographic assessment of left ventricular geometry and the mitral valve apparatus in cats with hypertrophic cardiomyopathy. J Vet Cardiol 2010;12: 1e16.
12. Bland JM, Altman DG. Statistical methods for assessing agreement between two methods of clinical measurement. Lancet 1985;1:307e310. 13. Abbott JA. Heart rate and heart rate variability of healthy cats in home and hospital environments. J Feline Med Surg 2005;7:195e202. 14. Wess G, Schulze A, Geraghty N, Hartman K. Ability of 5minute electrocardiography (ECG) for predicting arrhythmias in Doberman Pinschers with cardiomyopathy in comparison with 24-hour ambulatory ECG. J Vet Intern Med 2010;24:367e371.
Available online at www.sciencedirect.com
ScienceDirect