Information to Users

INFORMATION TO USERS

This manuscript has been reproduced from the microfilm master. UMI films the text directly from the original or copy submitted. Thus, some thesis and dissertation copies are in typewriter face, while others may be from any type of computer printer.

The quality of this reproduction is dependent upon the quality of the copy submitted. Broken or indistinct print, colored or poor quality illustrations and photographs, print bleedthrough, substandard margins, and improper alignment can adversely affect reproduction.

In the unlikely event that the author did not send UMI a complete manuscript and there are missing pages, these will be noted. Also, if unauthorized copyright material had to be removed, a note will indicate the deletion.

Oversize materials (e.g., maps, drawings, charts) are reproduced by sectioning the original, beginning at the upper left-hand comer and continuing from left to right in equal sections with small overlaps.

ProQuest Information and Learning 300 North Zeeb Road, Ann Arbor, Ml 48106-1346 USA 800-521-0600

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. THE ELECTROCARDIOGRAPHIC EVALUATION OF ARRHYTHMOGENIC CARDIOMYOPATHY IN BOXERS

DISSERTATION

Presented in Partial Fulfillment of the Requirements for the Degree Doctor of

Philosophy in the Graduate School of The Ohio State University

Alan W. Spier, D.V.M.

*****

The Ohio State University 2002

Dissertation Committee Approved by Associate Professor Kathryn M. Meurs, Adviser

Professor Robert L. Hamlin Adviser Veterinary Clinical Sciences Professor William W. Muir

Professor John D. Bonagura

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. UMI Number: 3059330

______® UMI

UMI Microform 3059330 Copyright 2002 by ProQuest Information and Learning Company. Ail rights reserved. This microform edition is protected against unauthorized copying under Title 17, United States Code.

ProQuest Information and Learning Company 300 North Zeeb Road P.O. Box 1346 Ann Arbor, Ml 48106-1346

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Copyright by

Alan W. Spier

2002

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ABSTRACT

Arrhythmogenic cardiomyopathy in Boxers is characterized by the development

of ventricular arrhythmias and sudden death. Standard diagnostics, including thoracic

radiographs and echocardiography are of limited use in the evaluation of Boxers with

arrhythmias. Alternative diagnostics are needed to better evaluate these patients, help

determine prognosis and guide therapy.

Electrophysiology testing is used in the evaluation of arrhythmias in people.

The need for anesthesia, the relative expense and risks inherent to the procedure, and its

limited availability preclude its use in veterinary medicine. Inexpensive, noninvasive

diagnostic tests that are associated with fewer risks would significantly improve the

management of patients with ventricular arrhythmias. The goal of this project was to

evaluate Boxers with arrhythmias by the use of noninvasive diagnostic testing,

including ambulatory electrocardiography, signal-averaged electrocardiography, QT

dispersion, and heart rate variability.

The degree of arrhythmia variability evaluated by ambulatory

electrocardiography can affect disease diagnosis and influence assessment of

progression and efficacy of therapy. Substantial variability was identified, and suggests

that an 80% reduction in arrhythmia frequency is necessary to document a drug effect,

while changes of less than 80% may be within the limits of spontaneous variability.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Signal-averaged electrocardiography documented the presence of late potentials

in Boxers that were most severely affected. The presence of late potentials suggests

that abnormal conduction exists in these dogs, which may represent a substrate for

arrhythmia formation. Identification of late potentials may therefore be a predictor for

mortality.

QT dispersion from standard electrocardiographic tracings was used to identify

abnormalities in repolarization. The degree of QT dispersion did not correlate with any

measure of disease severity. Many severely affected dogs had no QT dispersion, and

QT dispersion was identified in some less affected dogs. QT dispersion was therefore

not considered to be a useful diagnostic test in these dogs.

Analysis of heart rate variability was reduced in dogs with congestive heart

failure, but no reduction was identified between affected and unaffected dogs.

Increased sympathetic tone is unlikely to be a discriminating feature between unaffected

Boxers and those with only arrhythmias. Therefore, analysis of heart rate variability

was of little value in assessing arrhythmic disease.

iii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Dedicated to all the Boxers that “volunteered” to be in the study,

and to all those that never had the chance.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ACKNOWLEDGMENTS

I wish to thank my advisor, Dr. Kate Meurs for her seemingly endless support

and encouragement; for the many times she used carrots instead of sticks. Without her

effort and enthusiasm, this project would have never been possible.

I extend my deepest appreciation, respect, loyalty and admiration to Dr.

Robert Hamlin for his guidance and support. He is a role model, whose influence on

me personally and professionally, are beyond words. I am truly priveleged to have had

the opportunity to watch him work.

I thank Drs. John Bonagura and William Muir for their service as committee

members and their intellectual stimulation and constructive criticism.

I am indebted to Nicola Wright, who through the evolution of this project,

became the chief organizer, secretary, dog-holder, and, at times, cheerleader.

A special note of gratitude to Dr. Craig Hassler, Jeff Wallery, and Tom Vinci at

the Battelle institute for their generosity for allowing us both time and space to analyze

Holters.

I would like to acknowledge The Ohio State University canine research funds

and the AKC/ Boxer charitable trust for their financial support.

Finally, to all the veterinary students and volunteers who gave up their Saturday

mornings to help us with holding (and sometimes wrestling) dogs.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VITA

June 25, 1970...... Bom— El Paso, TX

1992 ...... BS Zoology, University of Texas

1993 ...... BS Veterinary Science, Texas A&M University

1996 ...... DVM, Texas A&M University

1996-199 7 ...... Internship, Small Animal Medicine and Surgery Kansas State University

1997-200 2 ...... Resident, Small Animal Medicine (Cardiology) The Ohio State University

!997-present ...... Doctoral Candidate, The Ohio State University

PUBLICATIONS

Research Publication:

1. Spier AW, Meurs KM, Muir WW, Lehmkuhl LB, Hamlin RL. Correlation of QT dispersion with indices used to evaluate the severity of familial ventricular arrhythmias in Boxers. Am J Vet Res 62,2001, 1481-5

2. Spier AW, Meurs KM, Coovert DD, Lehmkuhl LB, O’Grady MR, Freeman LM, Burghes AH, Towbin JA. Western blot analysis of myocardial dystrophin, a- sarcoglycan and P-dystroglycan in dogs with idiopathic dilated cardiomyopathy. Am J Vet Res 62,2001,67-71.

3. Meurs KM, Spier AW, Wright NA, Hamlin RL. Comparison of in-hospital vs 24 hour ambulatory electrocardiography for detection of ventricular premature complexes in mature Boxers. J Am Vet Med Assos 218,2001, 222- 224.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 4. Meurs KM, Spier AW, Wright NA, Hamlin RA. Use of ambulatory electrocardiography for detection of ventricular premature complexes in healthy dogs.J Am Vet Med Assoc 218,2001, 1291-2

5. Meurs KM, Spier AW, Miller MW, Lehmkuhl LB, Towbin JA. Familial ventricular Dysrhythmias in Boxer Dogs. JVIM 13, 1999 437-439.

6. Meurs KM, Magnon AL, Spier AW, Miller MW, Lehmkuhl LB, Towbin JA. Evaluation of the cardiac actin gene in Doberman pinschers with dilated cardiomyopathy. Am J Vet Res 62, 2001, 33-36.

FIELDS OF STUDY

Major Field: Veterinary Cardiology

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. TABLE OF CONTENTS

Page

Abstract ...... ii

Dedication...... iv

Acknowlegments ...... v

Vita...... vi

List of Tables ...... x

List of Figures...... xii

Background and Introduction...... 1

Chapter I—Spontaneous Variability in VPC frequency...... 5 Introduction...... 5 Materials and Methods...... 6 Inclusion criteria...... 6 24-hour ambulatory ECG...... 7 Data analysis and statistical evaluation...... 7 Results...... 9 Discussion...... 35 Footnotes...... 40 Reference List...... 41

Chapter 2—Signal-averaged electrocardiography...... 44 Introduction...... 44 Materials and Methods...... 46 Inclusion Criteria...... 46 Ambulatory Electrocardiography...... 46 Cardiac Ultrasonography...... 47 Signal-averaged Electrocardiography...... 48 Subject Categorization...... 50 Statistical Analysis...... 50

viii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Results...... 51 Discussion...... 62 Footnotes...... 70 Reference List...... 71

Chapter 3—QT dispersion...... 73 Intro...... 73 Materials and Methods...... 75 Animal selection...... 75 12-lead electrocardiography...... 75 24-hour ambulatory electrocardiography...... 76 Echocardiography...... 76 Statistical evaluation...... 77 Results...... 77 Discussion...... 80 Footnotes...... 85 Reference List...... 86

Chapter 4— Heart rate variability...... 89 Introduction...... 89 Materials and Methods...... 90 Patient Selection...... 90 Subject Categorization...... 90 Ambulatory electrocardiography...... 91 Heart rate variability analysis ...... 91 Statistical evaluation...... 92 Results...... 92 Discussion...... 100 Footnotes...... 104 Reference List...... 105

Summary and Conclusions...... 108

Bibliography...... 113

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF TABLES

Table Page

1.1 Holter data for dog 1...... 12

1.2 Holter data for dog 2...... 14

1.3 Holter data for dog 3...... 16

1.4 Holter data for dog 4...... 18

1.5 Holter data for dog 6...... 20

1.6 Holter data for dog 7...... 22

1.7 Holter data for dog 8 ...... 24

1.8 Holter data for dog 9 ...... 26

1.9 Holter data for dog 10 ...... 28

1.10 Holter data for dog 11...... 30

1.11 Descriptive data for all dogs...... 32

1.12 Data for day of recording...... 33

1.13 Data for day of week ...... 34

2.1 Descriptive SAECG data for Boxers...... 56

2.2 Descriptive data for normal (non-Boxers)...... 56

2.3 90/10 percentile for sensitivity, specificity, and predictive value...... 57

2.4 95/5 percentile for sensitivity, specificity, and predictive value...... 57

2.5 SAECG values to maximize specificity...... 58

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 2.6 SAECG values to maximize sensitivity...... 59

2.7 Summary of statistical tests for SAECG...... 60

3.1 Holter summary data for QT dispersion analysis...... 79

3.2 Individual data for dogs with non-zero QT dispersion...... 79

4.1 Heart rate variability summary data...... 95

4.2 Mean RR significance...... 96

4.3 SDNN significance...... 97

4.4 ASDNN significance...... 98

4.5 SDANN significance...... 99

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. LIST OF FIGURES

Figure Page

1.1 Holter placement...... 11

1.2 Dog 1 VPCs by day of recording ...... 13

1.3 Dog 2 VPCs by day of recording ...... 15

1.4 Dog 3 VPCs by day of recording ...... 17

1.5 Dog 4 VPCs by day of recording ...... 19

1.6 Dog 6 VPCs by day of recording ...... 21

1.7 Dog 7 VPCs by day of recording ...... 23

1.8 Dog 8 VPCs by day of recording ...... 25

1.9 Dog 9 VPCs by day of recording ...... 27

1.10 Dog 10 VPCs by day of recording ...... 29

1.11 Dog 11 VPCs by day of recording ...... 31

1.12 Holter hook-up...... 54

2.2 SAECG sketch ...... 55

2.3 Normal vs abnormal SAECG ...... 61

4.1 Holter figure...... 94

4.2 Mean RR interval significance...... 96

4.3 SDNN data and significance...... 97

4.4 ASDNN data and significance...... 98

4.5 SDANN data and significance...... 99 xii

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. BACKGROUND AND INTRODUCTION

Ventricular arrhythmias are commonly encountered in veterinary medicine, and

can be associated with a variety of systemic and metabolic conditions. However, the

majority of significant, often life-threatening, arrhythmias are identified in association

with underlying cardiac disease, either as the chief manifestation of the disease, or as a

complicating factor. Cardiomyopathy in dogs manifests primarily as either a structural

abnormality associated with poor contractile function and congestive heart failure, or by

an electrical instability that predisposes to arrhythmias and sudden death.

Arrhythmogenic cardiomyopathy, or familial ventricular arrhythmias, in Boxers

is a primary heart disease characterized by the development of ventricular arrhythmias

and increased risk of sudden death. The disease, as it was originally described, is

subdivided into three “types”. Type I disease is defined by the presence of ventricular

(and occasionally atrial) arrhythmias in an asymptomatic dog, whereas Type II disease

is characterized by either syncope or sudden death. The majority of affected individuals

have preserved systolic function, making congestive heart failure an uncommon

sequelae. However, a subset of dogs does present with evidence of myocardial failure

and congestive heart failure, similar to overt dilated cardiomyopathy. This so-called

type III disease occurs less commonly, and is not a typical presentation for the disease.

For this reason, standard diagnostics performed in most cardiac patients (such as

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. thoracic radiographs and echocardiography) are of limited use in most dogs.

Alternative diagnostics are needed to better evaluate these patients, and help determine

prognosis and guide therapy.

In human medicine, the definitive test used in the evaluation of arrhythmias is

the electrophysiology study. These types of studies are not commonly performed in

veterinary medicine. The need for anesthesia, the relative expense and risks inherent to

the procedure, and its limited availability preclude the use of routine electrophysiologic

testing. The availability of noninvasive diagnostic tests that are inexpensive and

associated with fewer risks would significantly improve the ability of clinicians to

manage patients with ventricular arrhythmias.

The goal of this project was to evaluate the use of noninvasive diagnostic tests in

the assessment of ventricular arrhythmias associated with arrhythmogenic

cardiomyopathy in Boxers. These tests include those that are currently being used such

as ambulatory electrocardiographic (or Holter) monitoring, and those that are routinely

used in human medicine, but have had only limited application in veterinary medicine.

These latter diagnostics include signal-averaged electrocardiography, evaluation of QT

dispersion from standard electrocardiograms, and analysis of heart rate variability.

Holter monitoring is extensively used in veterinary medicine for a wide variety

of conditions, including determining the cause of syncope or syncopal-like episodes,

documenting the presence of arrhythmias, assessing severity of disease, and monitoring

the effect of therapy. At this time, the diagnosis of arrhythmogenic cardiomyopathy in

Boxers is largely performed by electrocardiography. Because of the relative

insensitivity of standard in-hospital electrocardiograms, most veterinary clinicians

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. advocate the use of ambulatory monitoring. Unfortunately, there is no clear consensus

as to what constitutes an affected animal based on arrhythmia frequency, and even less

agreement regarding the initiation of therapy in an asymptomatic animal. The

identification of ventricular ectopia by Holter evaluation in symptomatic animals

usually warrants therapy, and many asymptomatic dogs that manifest “frequent”

arrhythmias are also placed on antiarrhythmic therapy. Therapy is then assessed with

follow-up ambulatory electrocardiography to document efficacy and to monitor for

proarrhythmia. The assessment of both efficacy and proarrhythmia must take into

account the degree of inherent variability of arrhythmia frequency. There are no studies

in dogs that describe the variability of arrhythmia frequency; consequently evaluating

the effect of therapy is a challenge. The initial study was performed to identify to what

degree the frequency of arrhythmia varies for a particular animal evaluated by

ambulatory electrocardiography.

Signal-averaged electrocardiography is a high-resolution electrocardiogram that

identifies the presence of late potentials in the terminal portions of the QRS complex.

The presence of late potentials suggests that a conduction abnormality exists, which

may serve as a substrate for reentry, and predispose to arrhythmia formation. Signal-

averaged electrocardiography has been used in veterinary medicine in the evaluation of

dilated cardiomyopathy in Doberman pinschers, but its use has not been reported in

Boxers.

In addition to altered conduction, abnormal repolarization can also predispose to

reentry, and facilitate arrhythmia formation. Repolarization can be directly evaluated

with invasive electrophysiologic testing, but can also be evaluated from the surface

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. electrocardiogram. The assessment of changes in the duration of the QT interval

between different leads is termed QT dispersion. QT dispersion has been used in

several disease states in people as a marker for adverse arrhythmic events. Its clinical

utility, however, has not been investigated in veterinary medicine. The identification of

altered repolarization in Boxers may be useful in identifying at-risk patients.

The association between cardiac disease and the autonomic nervous system is

well known. Patients that suffer from significant heart disease, particularly congestive

heart failure, have altered autonomic balance, with an increase in sympathetic tone and

a withdrawal of parasympathetic tone. The manifestation of ventricular arrhythmias is

heavily influenced by the sympathetic nervous system, and excessive sympathetic tone

can create or exacerbate malignant arrhythmias. An indirect measure of autonomic

balance is the assessment of the inherent irregularity of heart rate. Reduction in this

irregularity suggests a loss of normal autonomic modulation of the heart, which may be

associated with increased sympathetic tone and/or a decrease in parasympathetic tone.

The effect of this unbalance, in some diseases, correlates with mortality and cardiac

death. Analysis of heart rate variability has been evaluated in Doberman pinschers with

dilated cardiomyopathy, but has not been evaluated in Boxers with arrhythmogenic

cardiomyopathy. Abnormally reduced heart rate variability may suggest an increase in

underlying sympathetic tone, which may be important in arrhythmia formation,

treatment, and assessment of risk in these dogs.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 1

SPONTANEOUS VARIABILITY IN VPC FREQUENCY

Introduction

Ambulatory electrocardiographic (AECG, Holter) monitoring plays an important

role in veterinary medicine. Such evaluations are used in the diagnosis of disease,

assessment of disease severity, evaluation of the response to antiarrhythmic therapy,

and in many cases monitoring the progression of disease.1'6 Although many parameters

are measured on an AECG, the number and complexity of ventricular arrhythmias are

most widely used. The 24-hour ambulatory evaluation is believed, and has been

recently demonstrated, to be a more sensitive indicator of an individual’s overall

arrhythmic state.7 Furthermore, most reports in veterinary medicine use only a single

recording on which to base clinical decisions. It is unknown, however, to what degree

the frequency of ventricular arrhythmias varies on a day-to-day basis. Unlike human

medicine, very little, if any, information concerning the biologic variability of electrical

activity in dogs has been reported. Biologic variability may affect decisions in disease

diagnosis, disease severity assessment, and therapeutic intervention.

Boxer dogs with arrhythmogenic cardiomyopathy manifest ventricular

arrhythmias and are at risk for sudden death.8 Unlike other forms of cardiomyopathy,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. echocardiography is of limited use because myocardial function is typically preserved;

consequently, the disease in this breed is frequently diagnosed by Holter monitoring.

Based on these recordings, many of these dogs are treated with long-term

antiarrhythmic therapy, and efficacy of that therapy is assessed by performing follow-up

Holter analysis. The objective of this study, therefore, is to document and describe the

degree of spontaneous variability in the frequency of ventricular arrhythmias in Boxer

dogs affected with arrhythmogenic cardiomyopathy, and assess what implications it

may have on the diagnosis and treatment of the disease. In addition, differences in both

the day of recording and day of the week were evaluated to identify what effect, if any,

these factors may have in the frequency of ventricular arrhythmias in these dogs.

Materials and Methods

Inclusion criteria

Boxer dogs were prospectively recruited for evaluation of 24-hour ambulatory

electrocardiograms (AECG, Holter monitoring) over seven consecutive days. Dogs

were included if they were free of structural heart disease based on previous

echocardiographic and Doppler evaluation (including acquired myocardial and valvular

disease and congenital valvular disease) and had a prior Holter recording with at least

500 VPCs per 24 hour period. The Holter recording used for study inclusion was not

included in data analysis. Only those recordings at least 20 hours in duration were

included, and dogs were excluded if less than 6 days of recordings were available. Only

dogs over 2 years of age were considered for inclusion.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24-hour ambulatory ECG

Twenty-four hour ambulatory electrocardiograms11 were obtained using a seven

lead, three channel electrode system, and acquired using an analog tape recorder.

Electrodes were so as to approximate the frontal leads I, II, and III (Figure 1.1).

Recordings were analyzed using a prospective software analysis algorithm with

continuous user interaction15 by a trained cardiology research assistant under the

guidance of a veterinary cardiologist. The following data were obtained for each 24-

hour Holter recording: duration of recording, mean heart rate(HRmean), minimum heart

rate (H R mjn), maximum heart rate (HRmax),number of VPCs per recording (VPC#), day

of recording (DOR, numbered one to seven), day of week (DOW, Monday through

Sunday), total number of beats, and grade of arrhythmia. If VPCs were identified, the

arrhythmia was graded on the basis of a modification of the Lown grading scheme as

follows: grade I = single, uniform VPC only; grade 2 = bigeminy, trigeminy, or

multiform VPCs; grade 3 = presence of couplets or triplets, and grade 4 = RonT

phenomenon or ventricular tachycardia (four or more consecutive VPCs). Animals in

which no VPCs were identified were classified as grade 0.9

Data analysis and statistical evaluation

Variability was evaluated for on a day-to-day basis for each dog. Additionally,

differences in the day of recording (e.g. day 1-7) and day of the week (Monday,

Tuesday, Wednesday, etc) were also analyzed. For dogs in which only six recordings

were available over a seven day period, data points for both day of recording and day of

week analysis were missing . For dogs in which recordings were taken over an eight-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. day period, two recordings were obtained on the same day of the week (due to a missing

day). When evaluating for differences in VPC# relative to day of the week, a mean

value from each of the two recordings was calculated, and that combined data point for

that particular weekday was used in the statistical analysis. This created a missing data

point for day of week analysis (due to the missing day), but did not affect the day of

recording analysis because a total of seven days were obtained.

Statistical evaluation included Repeated Measure One Way Analysis of

Variance (RM-ANOVA) to identify significant differences in the variability of VPC #

among dogs relative to DOR and DOW. The non-parametric RM-ANOVA on ranks

was used because the data were not normally distributed, despite multiple transforms to

normalize the data. The presence of the missing data points (for both day of recording

and day of the week) did not allow simultaneous evaluation and assessment of

interaction between DOW and DOR; consequently two independent RM-ANOVA on

ranks were performed. Significance was defined as a < 0.05.

Evaluation of day-to-day variability for each dog was evaluated by calculating

the percent difference between the maximum value and minimum value obtained during

the week. A percent difference was also obtained between the median value of the

week, and the minimum or maximum value of the week (whichever yielded the greater

value).

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Results

Eleven dogs were initially evaluated by seven-day ambulatory monitoring.

However, one dog was excluded because more than one 24-hour Holter recording was

less than 20 hours in duration. Consequently, a total of ten dogs met the inclusion

criteria. Due to either technical difficulties or client compliance issues, seven

consecutive days were obtained in only six of ten dogs; six recordings were obtained

over a seven day period in two dogs, and seven recordings were obtained over an eight

day period in two dogs. The maximum, minimum and mean heart rate, number of

VPCs per 24-hour recording, grade of arrhythmia, day of recording, and day of the

week for each dog are presented in Tables 1.1-1.10. Those recordings that were

missing for dogs in which only six recordings were included (both DOW and DOR), or

missing days for dogs that completed all seven recordings over an eight day period

(DOW only), are represented by “—” in the table. For dogs in which seven days were

recorded, but were obtained over an eight day period, the weekday that was duplicated

and subsequently averaged is qualified by an For each dog, the number of VPCs

per recording, the mean, median, 25th and 75th percentiles, and calculations of the

percent change in the number of VPCs between the minimum and maximum values

during the week, as well as a percent change between the median and either the

minimum or maximum values during the week (whichever yielded the greater percent

change) are shown in Table 1.11.

Composite data for all dogs for day of recording (DOR) and day of the week

(DOW) are shown in Tables 1.12 and 1.13, respectively. With regard to day of

recording analysis (e.g. day 1-7), no difference between each day of recording was

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. identified. In other words, day 1 was not different relative to day 2, etc. For day of

week analysis (e.g. Monday, Tuesday, etc), no difference was identified relative to the

days of the week. Graphs representing the variation of VPC# for each dog are presented

in Figures 1.2-1.11.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Black Orange White Red Green FROM FROM RIGHT FROM LEFT Whi Figure 1.1. Figure 1.1. Positionof electrodes placement Hotter for as both seen right and thesides. from left There pair 7 electrodes; 3 bipolarare create 3 to (orchannels), to serve leads and as ground. 1 Channel 1 II. consists Channel 2 brown electrodeofthe as positive black and the (+) (-), negativeelectrode as consiststheof electrode red positive (+)as white theelectrodeand negative as (-), approximating lead approximating consistsChannel III. of3 lead orangethepositive electrode as blue the electrode (+) and approximating(-), negative as I. lead The green electrode as serves ground. Blue Brown

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 83 62 97 — 125 VPC# 73 66 69 7173 79 90 66 144 —— 146 207 68 237 222 Max Max HR Mean HR — 45 45 191 47 46 Min Min HR 2 32 52 2 2 — Monday Saturday Thursday 3 45 192 Day of Day week Grade 1 Wednesday 3 Friday 2 5 Sunday 3 45 196 7 Wednesday 4 6 Tuesday — 1 1 1 1 1 1 1 1 Dog#of recording Day Wednesday. Wednesday. dayof weeksubsequent analysis (DOW), data In two pointsthevalues these are for Table 1.1. Table 1.1. data#1. HolterDog for of Daytheofdayrecording, of thearrhythmia, week, grade numberofpresented. (VPC#) are VPCs this particular obtained daysofdog, were In over recording 7 eight an dayperiod. data There no is (denoted Monday“—”), data and are two forbypoints there for averaged, and the mean value is used. minimum(Min maximumrate heart (Max HR), rate heart heartrate (Mean mean HR), HR), and total

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

Dog 1 #DdA

Figure 1.2. The number of VPCs per 24 hour recording (VPC#) are presented for dog # 1. 33 121 VPC# 73 228 71 35 71 20 Mean Mean HR 186 175 69 44 159 159 66 48 Max Max HR 43 162 70 43 155 65 Min HR 2 42 2 2 44 2 Grade Friday 2 41 Sunday Monday 3 44 178 Tuesday 2 39 Saturday Thursday Wednesday Day ofweek 1 3 5 7 4 2 22 2 2 2 2 6 2 Dog#of recording Day Table 1.2. Table 1.2. data Dog#2. Holler for dayofthe ofof Dayarrhythmia, recording, thegrade week, (Min minimum maximumHR), rate (Max rate (Mean heart HR), rate heart heart HR), mean total and numberofpresented(VPC#) are VPCs

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

Dog 2 #OdA 1®

Figure 1.3. The number of VPCs per 24 hour recording (VPC#) are presented for dog #2. 1 2 27 VPC# 72 4 77 9 65 0 67 144 154 154 134 159 64 Max HR Mean HR 39 38 41 170 70 42 45 141 68 1 Min HR Min 1 46 Grade Friday Monday 2 40 Tuesday 1 Saturday 2 Thursday 1 Wednesday 0 3 6 3 2 3 1 3 4 3 3 5 Sunday 1 3 7 3 Dog#recording of Day ofweek Day Table 1.3. Table 1.3. data Holter #3. forDog ofof Day day recording, the week, of arrhythmia, the grade (Min heart rate minimum maximum HR), and heart mean (Mean heart HR), (Max HR), rate total rate number ofVPCsnumber (VPC#) presented. are

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Day of RecordingDay Figure Figure 1.4. of The number per 24 hour presented recording (VPC#) VPCs are dog #3. for #OdA

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1951 1257 1542 1608 1988 2325 VPC# 129 115 1599 114 193 187185 128 198186 128 120 Max Max HR Mean HR 86 72 82 201 86 196 116 79 3 3 3 81 3 Friday Sunday 3 Monday 3 86 Tuesday 3 Saturday Thursday Wednesday 2 3 5 7 4 6 Day of recordingof Day of Day week Grade Min HR 4 1 4 4 4 4 4 4 Dog# Table 1.4. Table 1.4. data Holter #4. Dog for of ofdayrecording, Daytheweek, of the grade arrhythmia, minimum rate(Min heart (Max HR), maximumrate heartheart (Mean rate mean HR), and total HR), of(VPC#)number areVPCs presented.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure Figure 1.5. of number per The (VPC#) VPCs recording dog 24 hour presented #4. are for #OdA

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 728 623 1336 VPC# 64 6567 1609 817 6573 905 702 162 140 149165 68 70 Max Max HR HR Mean 39 42 144 40 164 44 42 158 42 2 2 2 2 2 Friday 3 Monday Saturday 3 40 Day ofDay week Grade Min HR 3 Sunday 2 57 Tuesday Thursday 4 6 Wednesday Day ofDay recording 66 6 1 6 6 6 6 Dog# Table 1 .5. .5. dataTable #6. Dog Hotter for 1 ofday Dayrecording, of the the week, ofgradearrhythmia, ofVPCspresented. number(VPC#) are minimum maximum(Min heart (Max rate HR), heart heartmean HR), and total rate HR), rate (Mean

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. without prohibited reproduction Further owner. copyright the of permission with Reproduced

Dog 6 #OdA

Day of Recording Figure 1.6. The number of VPCs per 24 hour recording (VPC#) are presented for dog #6. — 5552 5152 4826 4500 4447 VPC# 20679 89 86 93 Ill 174 183169 92 75 161 ——— 62 3 61 — Grade Min HR Max HR Mean HR Friday Monday Tuesday 3 59 Saturday 3 Thursday 3 56 183 94 Wednesday 3 60 166 Day ofweek Day 1 3 5 Sunday 2 7 Day of recordingofDay 7 7 2 7 7 4 7 7 6 7 Dog# Table 1.6. Table 1.6. data Hotter #7. for Dog of day Dayrecording, of the the week, arrhythmia, grade of (Min minimum maximum HR), rate heart rate (Max heart mean HR), heart total rate (Mean and HR), over a 7 day period. period. day over a7 Therenodatais which by“—”.for Friday, denoted is number of VPCs (VPC#) are presented. (VPC#)ofnumberVPCs arepresented. thisparticular daysdog, of6 In obtained recording were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. m m wmjgmgjjjjg^ m mmmm Dog 7 Dayof Recording 1 1 2 3 4 5 6 Figure 1.7. Figure 1.7. The numberof24 VPCsperhour recording (VPC#) presenteddog are #7. for a There is missing data point missing dayrecording for#3. 5000 15000 10000 25000 20000 Q. > s u» N>

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9 6 17 32 VPC# 65 25 63 3 64 8 61 Mean HR Mean 154 61 150169 62 192149 66 154 151 39 40 2 2 2 43 Grade Min HR Max HR Friday Sunday Wednesday Day ofDay week 2 Monday 3 40 3 Tuesday7 2 Saturday 38 1 40 4 6 Day of recordingofDay 8 1 8 8 8 5 Thursday 1 40 8 8 8 Dog# Table 1.7. Table 1.7. data #8. Dog for Hotter dayofof the Daytherecording, of week, grade arrhythmia, minimum heart rate (Min HR), minimum(Min maximumHR), (Max rate heart HR), rate rate heart mean heart (Mean and total HR), ofnumberare(VPC#)VPCs presented.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. — Dog 8 Day ofRecording 1 1 2 3 4 5 6 7 Figure 1.8. Figure 1.8. of Thenumberper dog(VPC#)24 hour VPCs presented recordingare #8. for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ------9095 2644 4390 11215 10623 VPC# 58 63 ------64 107111 55 59 115 Max Max HR Mean HR 3839 101 38 37 108 60 6522 —— 43 — Tuesday 3 Wednesday 3 Day ofweekDay Grade Min HR 1 2 3 Thursday 3 5 Saturday 3 41 114 7 Monday 3 6 Sunday 4 Friday 3 9 9 9 9 9 9 9 Dog#recordingof Day Table 1.8. Table 1.8. data Holter #9. Dog for of recording, Dayof week, theday the ofgrade arrhythmia, (Min rate minimum heart maximum rate (Max HR), heart heart (Mean mean rate total HR), and HR), dayperiod. 7 There data is no denoted Sunday, by“—”. for number of(VPC#) number VPCs presented. are this particular ofdays dog, recording6 over In obtaineda were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure Figure 1.9. of The number VPCs per recording hour 24 (VPC#) presented dog are #9. for There is a missing data missing point There day a recording is #6 for

#OdA

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1384 2125 2751 5103 95 94 96 3831 Mean Mean HR VPC# ——— 173176 96 187 105 3795 174 190 99 67 186 106 6194 3 50 184 3 —— Friday 3 60 Sunday 3 55 Tuesday 3 62 Saturday 3 64 Saturday Thursday 3 59 Wednesday 2 3 Monday 7 — Day of Day recording of Day week Grade HR Min Max HR 10 10 4 10 1 10 10 6 10 10 5 10 Dog# Table 1.9. Table 1.9. data Holter#10. Dog for of ofrecording, Day day the the week, of arrhythmia, grade minimum maximumheart rate(Min HR), heart rate (Max heartmean total (Mean HR), rate and HR), ofnumberVPCs (VPC#)presented. are thisparticulardays of dog, In obtained were7 recording eight dayover an period. There data no is by Wednesday(denoted for “—"), data two and there are points Saturday. for day of subsequent week analysis In the(DOW), datavalues points two these for theareand averaged, used. value mean is

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Figure Figure 1.10. 10. of The per 24 number hour recording dog (VPC#) VPCs are presented # for

#OdA

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 5507 2037 2243 VPC# 77 76 4730 6668 1946 77 1316 Mean Mean HR 139 138 82 154 142 66 2670 Max Max HR 53 132 54 41 53 140 3 3 3 Friday 3 41 Sunday 3 46 146 Tuesday Monday 2 Saturday 3 40 Wednesday Day ofweek Day Grade Min HR 1 2 5 3 Thursday 7 6 Day of recording ofDay 11 11 11 11 4 11 11 11 Dog# minimum heart (Minminimum rate heart maximum rate (Max HR), heart (Mean heart mean rate total and HR), HR), Table 1.10. Table 1.10. data Hotter #11. Dog for of ofrecording, day week, of the Day the grade arrhythmia, numberofare (VPC#) presented. VPCs

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ■‘SfcCtfr'C Day Day of Recording Figure Figure 1.11. 11. The of 24 number per hour presented recording dog (VPC#) VPCs are # for

#OdA

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 31 76 64 59 49 66 100 % Diff (median) 57 38 78 76 91 81 76 61 100 % % Diff (min/max) 2 9 91 72 817 3795 2243 6 97 90 2921 27 228 76 44 1609 960 2325 1753 1608 46 5507 6194 3598 11215 7415 7808.5 3 32 14 0 20 Min Max Mean Median 1384 1257 Dog Dog Dog 1 62 144 Dog 6Dog 623 Dog 4 Dog Dog 3 Dog Dog 2 Dog Dog 9 Dog 2644 Dog Dog 7 8 Dog 4447 20679 7526 4989 78 Dog Dog 10 Dog Dog 11 1316 Table 1.11. Table 1.11. data Descriptive including minimum rate, (Min) heart maximum heart heart (Max) rate, median and rate dogs. Boxer for 10 percent The of(% VPCsnumber perrecording calculatedmin/max)Diff was each dog. for difference betweendifference the maximum of numberVPCs per and recording minimum differencepercent The the between median valuenumberof the for VPCs the during theweek minimumand either or of maximum also numberVPCs was whichevercalculated; the changegreateryielded percent reflected is the table. in K) w

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 7.0 1570.5 4581.5 6.5 1.0 84.5 34.0 111.0 81.0 1120.5 715.0 1969.5 3700.0 1991.5 10241.0 4923.0 817.0 1608.0 7808.5 3795.0 4467.0 2438.0 4989.0 5452.0 Median % 75th % 25th 6.3 2.0 14.3 9.0 21.0 97.1 90.0 960.0 3597.6 2921.3 2243.0 33 75.6 44.0 623 1316 47.3 4500 7526.0 6194 4390 7414.8 1947.8 2 1 79 90 48 — 702 2243 4447 5103 702.0 969.5 1624.4 9 35 144 728 5152 Day 5Day 6 Day Day 7 Mean 2480.3 17 3 8 6 27 125 1599 1608 1988 2325 1752.9 51.5 62.3 48.0 2751 3831 Day 4 Day 20679 — 32 817 905 1542 1384 83.0 2037 1946 2670 Day 3 Day 11215 6522 10623 1542.0 2549.8 3540.8 2243.0 3873.8 97 83 44 121 20 25 1257 57.3 2644 2125 4730 2514.3 9 62 4826 5552 9095 103.5 Day Day 1 2 Day 1643.5 1433.0 817.0 1252.0 1168.0 4568.3 Dog# Mean 2680.9 1808.4 1915.0 3459.1 Dog Dog 1 2Dog 228 Dog 3Dog 0 1 4 Dog 4 Dog 1951 Dog 6 Dog 1336 1609 Dog Dog 7 8 Dog 9 Dog 75th % 75th Dog Dog 11 5507 Dog Dog 10 3795 25th % 25th Median were obtained (dogsdays and missingwere 9), by7 are represented dog dayof and each recording. each for shown are percentiles for and 75th median, 25lh Mean, Table 1.12. Table 1.12. of Thenumbereach day areofVPCs recording for presented all dogs. for 6 days only thosedogs which For in W w

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 9 9 ------817 5152 2125 2243 1385.6 Sunday 2125.0 6 27 121 20 2325 1951 10623 20679 1609.0 817.0 2670.0 4242.8 Saturday 83 125 144 — 1988 1946 2670 44.0 121.0 20.0 1892.7 1 4 623 1336 1609 1608 5552 3831 5103 4995* 2037 129.8 11215 6522 0 17 3 8 33 228 44 — 76* 97 702 1599 4826 4730 1150.5 1115.5 1336.0 3165.5 3382.5 1988.0 Wednesday Thursday Friday 1 79 4500 9095 2644 1135.0 2428.3 1818.9 2519.5 4062.8 2 — 35 48 25 32 905 728 1257 1542 1384 2751 1316 5507 35.0 55.8 29.0 4447 Monday Tuesday 1 2 3 7 8 4 6 9 4390 11 10 Mean 1529.0 Dog# 75th %75th 1384.0 25th % Median 1257.0 Table 1.13. Table 1.13. numberof The dayofVPCs week each the for presented. is dogs which only 6 in For were obtained over an 8 day period (dogs 1 obtainedwere dayperiod(dogs overan 8 and missingdaysrepresented by“—” also are 1 10), In these obtained dogs, 2 recordingwere on the weekday same and dog Saturday(Wednesday for for 1 value by the represented is days obtained (dogs were and 9), missingdenoted days are 7 by “—”. dogs days which For in 7 dog 10). cases, the numberofthese those particular In VPCs daysand thewere averaged, for mean

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Discussion

The results of this study indicate that the degree of variability is inversely

related to the frequency of VPCs, with higher VPC frequency (>500 VPCs/24 hrs)

associated with variability as high as 80%. This translates into a 5-fold difference

between the minimum and maximum values obtained over the period of a week.

Therefore, changes in VPC frequency of less than 80%, or less than a 5-fold difference,

may be within the limits of physiologic variability.

By contrast, the grade of arrhythmia did not vary to the same degree. In 8 of the

10 dogs, the grade of arrhythmia only differed by one category, and in 3 did not differ at

all during the week. All the dogs that did not exhibit any variability in grade had very

frequent VPCs and a grade of 3 each day (the highest grade observed in the study). In

only 2 dogs was there a difference of greater than one grade category within the week;

this finding was observed in dogs with the least frequent arrhythmias—one with no

more than 32 VPCs in any given day, and the other in the dog with the fewest number

of VPCs, including zero. Therefore, the grade of arrhythmia, as defined in this study,

appears less variable than the frequency of premature complexes, but appears to also

vary inversely with VPC frequency.

Statistical evaluation of potential differences relative the day of recording and

day of the week suggest that no predictable trends exist for either scenario. It was

hypothesized that during the initial days of recording, dogs would be more agitated by

the monitors, which might be reflected by an increase in the frequency of arrhythmia; as

the week progressed, the dogs would become more adapted to wearing the monitors,

and the VPC frequency would decrease. However, no such consistent relationship

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. between VPC frequency and recording day was observed. It was also postulated that

the day of the week may have influence on arrhythmia frequency; the owner’s activity

might vary between days (for example weekend versus weekday), which could

indirectly affect the dog’s activity and subsequently affect the number of VPCs. Again,

however, no relationship between day of the week and number of VPCs was observed.

These findings are consistent with results of similar human studies evaluating

the variability of ventricular arrhythmias. In the late 1970’s, it was observed that people

with VPCs exhibited significant variability in the frequency of these abnormal beats.

Analysis using three consecutive 24-hour AECG recordings suggested that changes in

VPC frequency independent of therapeutic intervention could account for as much as

an 83% reduction in VPC frequency.10 Antiarrhythmic trials evaluating variability of

VPC frequency suggested that a reduction in VPC frequency between 80-90% may be

necessary to document drug response.11'12 Concern that variability in VPC frequency

could mimic the effects of antiarrhythmic therapy led to a trial conducted in a manner

similar to antiarrhythmic drug trials, except that no medication was administered.

Using the accepted standard criterion for that time of 50% reduction in the number of

VPCs to document antiarrhythmic response, 65% of individuals would have been

determined “drug responders” had they received medication.13

Few reports describing the frequency of ventricular arrhythmias in dogs are

available. Early reports suggest that the occurrence of arrhythmias in clinically normal

dogs is rare.14 Recent investigation into the electrical activity in normal dogs suggests

that clinically normal dogs do not possess the same degree of ectopic ventricular

activity that is seen in clinically normal people. In a study of 228 clinically normal

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Beagles, only 49 (21.5%) of the dogs had VPCs, and only four of these (3.125%) had

more than nine VPCs in 24 hours.15 In a group of 49 dogs of varying breeds, no dog

had greater than 24 VPCs per 24 hours, and the median value was zero.16 In a group of

16 healthy dogs, only two dogs had greater than 24 VPCs per recording, and they were

both Boxers.17 These reports describe a strikingly different scenario than exists in

people, where studies have shown as many as 50% of clinically normal people exhibit

ectopic ventricular activity as frequent as several thousand VPCs per 24 hours.18'20

Recently, there have been more studies evaluating Holter-derived assessment of

ventricular arrhythmias in dogs with cardiovascular disease. Many of these reports

identify the presence of ventricular arrhythmias as having diagnostic and prognostic

value in Doberman pinschers with dilated cardiomyopathy.5,21'23 Others have described

the Holter findings of inherited ventricular arrhythmias of German shepherd dogs and

Boxers. 1 ? 4 Holter monitoring has also been performed in the assessment of ventricular

arrhythmias associated with non-cardiac disease (traumatic myocarditis and

splenectomy).25,26

The importance of spontaneous variability in ventricular arrhythmias for the

purposes ofdisease diagnosis in the previous examples is not clear. However,

spontaneous variation in the frequency of ventricular arrhythmias is particularly

important as it applies to response to therapy. Studies that attempt to evaluate the

antiarrhythmic efficacy of a drug must take these variations into account. Some

antiarrhythmic trials in dogs have been performed, but in earlier studies the response to

therapy was not based on results obtained by the use of Holter monitors. 27,28 More

recent studies evaluating the effect of antiarrhythmic therapy have been reported using

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Holter monitor evalaution.6,29 In both of these later studies, criteria used for an

antiarrhythmic response were consistent with the guidelines presented here, as well as

those described in the human literature.

Twenty-four hour AECG remains the most reliable method available for

obtaining electrocardiographic data in people and animals. However, significant

spontaneous variability makes repeatable sequential analyses of cardiac electrical

activity difficult. The implication of this variability appears to be most important when

assessing efficacy of antiarrhythmic therapy. The effect of this variability, however, is

less clear in the application of Holter monitoring for the purpose of disease diagnosis.

Future studies will hopefully aid in identifying the utility of ambulatory monitoring for

this purpose.

As with any study, this report has limitations. The patients included in this

analysis were all Boxers, and all were believed to have the same cardiovascular disease

process (arrhythmogenic cardiomyopathy). Therefore, the results of this study may

only apply to this patient population. However, as previously described, the degree of

variation in this study very closely resembled the values obtained in the human

literature. It may therefore be reasonable to apply these guidelines to disease processes

associated with ventricular arrhythmias in other breeds with other disease processes.

Another problematic finding in this study was the disparity between the Holter

evaluations in some of the dogs relative to their inclusion criteria. The inclusion criteria

for this particular study was that a previous Holter (not analyzed as part of the 7 day

evaluation) identified at least 500 VPCs. In four of the dogs, the number of VPCs was

far less than 500, and in one case reached zero VPCs, despite the inclusion Holter.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Additionally, these dogs manifest a variability as high as 90% (and 100% in the case of

the dog with zero VPCs). These findings were a surprise, and are difficult to explain.

However, studies performed in people have shown that in addition to acute variations

in VPC frequency, spontaneous variability also occurs over time. To determine short

term and long-term variability, two AECG recordings were taken 2-14 days apart and

were then repeated 6-12 months later.30 Statistical analysis of recordings separated by

the short interval (2-14 days) revealed variations in VPC frequency of 69% for the early

pair, and 73% for the pair taken later. However, when comparing two recordings

separated by the long interval (6-12 months), variability in VPC frequency was as high

as 98-100%. This longer term variability may provide some explanation for the

difference seen between the inclusion Holter, and those obtained during the 7-day

recording period. Great care must be taken, therefore, in assessing Holter recordings

taken many months apart.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Footnotes

a Cardiocorder cassette recorder, Del Mar, Irvine C A

b Accuplus Holter analysis system, Del Mar, Irvine CA

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reference List

1. Meurs KM, Spier AW, Miller MW, Lehmkuhl L, Towbin JA. Familial Ventricular Arrhythmias in Boxers. J Vet Intern Med 1999;13:437-439.

2. Calvert CA, Jacobs GJ, Smith DD, Tathbun SL, Pickus CW. Association between results of ambulatory electrocardiography and development of cardiomyopathy during long-term follow-up of doberman pinschers.J Am Vet Med Assoc. 2000;216:34-39.

3. Calvert CA, Jacobs G, Pickus CW, Smith DD. Results of Ambulatory Electrocardiography in Overtly healthy Doberman Pinschers with Echocardiographic Abnormalities. JAVMA 2000;217:1328-1332.

4. Ware WA. Practical Use of Holter Monitoring. Compendium on Continuing Education fo r the Practicing Veterinarian 1998;20:167-177.

5. O'Grady, M. R. DCM in Doberman pinschers: Lessons learned in the first decade of study. 2002. Dallas, Tx. Proceedings of the 20th annual ACVIM forum.

6. Meurs KM, Spier AW, Wright NA, Atkins CE, DeFrancesco TC, Gordon SG, Hamlin RL, Keene BW, Miller MW, Moise NS. Comparison of the effects of four antiarrhythmic treatments for familial ventricular arrhythmias in Boxers. Journal o f the American Veterinary Medical Association 2002;221.

7. Meurs KM, Spier AW, Wright NA, Hamlin RL. Comparison of in-hospital versus 24-hour ambulatory electrocardiography for detection of ventricular premature complexes in manure Boxers.JAVMA 2001;218:222-224.

8. Harpster NK. Boxer cardiomyopathy. In: Kirk RW, ed. Current Veterinary Therapy: Small Animal Practice. Philadelphia: WB Saunders, 1983:329-337.

9. Lown B, Calvert C, Armington R. Monitoring for serious arrhythmias and high risk of sudden death. Circulation 1975;51-52 (Suppl 3): HI-189-HI-198.

10. Morganroth J, Michelson E, Horowitz L. Limitations of routine long-term electrocardiographic moitoring to assess ventricular ectopic frequency.Circulation 1978;58:408-414.

11. Engler R, Ryan W, LeWinter M, Bluestein H, Karliner JS. Assessment of long term antiarrhythmic therapy: studies on the long-term efficacy and toxicity of tocainide. American Journal of Cardiology 1979;43:612-618.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12. Winkle R, Gradman A, Fitzgerald J. Antiarrhymic drug effect assessed from ventricular arrhymia reduction in the ambulatory electrocardiogram and treadmill test: comparison of propranolol, procainamide and quinidine.Am J Cardiol 1978;42:473- 480.

13. Winkle R. Antiarrhythmic drug effect mimicked by spontaneous variability of ventricular ectopy.Circulation 1978;57:1116-1121.

14. Patterson D, Detweiler D, Hubben K. Spontaneous abnormal cardiac arrhythmias and conduction disturbances in the dog. Am J Vet Res 1961 ;355-369.

15. Ulloa H, Adams W, Houston B, Altrogge D. Arrhythmias prevalence during ambulatory eclectrodardiographic monitoring of beagles. Am J Vet Res 1995;56:275- 281.

16. Meurs KM, Spier AW, Wright NA, Hamlin RL. Use of ambulatory electrocardiography for detection of ventricular premature complexes in healthy dogs. JAVMA 1 A.D.;218:1291-1292.

17. Hall L, Dunn J, Delaney M. Ambulatory electrocardiography in dogs. Vet Rec 1991;129:213-216.

18. Takada H, Mikawa T, Murayama M. Range of ventricular ectopic complexes in healthy subjects studied with repeated ambulatory electrocardiographic recordings.Am J Cardiol 1989;63:184-186.

19. de Backer G, Jacobs D, Prineas R. Ventricular premature contractions: A randomized non-drug intervention trial in normal men. Circulation 1979;59:762-768.

20. Brodsky M, Wu D, Denes P. Arrhythmias documented by 24 hour continous electrocardiographic minitoring in 50 male medical students without apparent heart disease. Am J Cardiol 1977;39:390-395.

21. Calvert CA, Jacobs GJ, Smith DD, Rathbun SL, Pickus CW. Association between results of ambulatory electrocardiography and development of cardiomyopathy during long-term follow-up of Doberman Pinschers.Journal o f the American Veterinary Medical Association 2000;216:34-39.

22. Calvert CA, Jacobs GJ, Pickus CW, Smith DD. Results of ambulatory electrocardiography in overtly healthy Doberman Pinschers with echocardiographic abnormalities. Journal o f the American Veterinary Medical Association 2000;217:1328-1332.

23. Calvert CA, Wall M. Results of ambulatory electrocardiography in overtly healthy Doberman Pinschers with equivocal echocardiographic evidence of dilated cardiomyopathy. Journal o f the American Veterinary Medical Association 2001;219:782-784. 42

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24. Moise NS, Gilmour RF, Riccio M. Diagnosis of inherited ventricular tachycardia in German Shepherd dogs. J Am Vet Med Assoc 1997;210:403-410.

25. Snyder PS, Cooke KL, Murphy sT, Shaw NG, Lewis DD, Lanz OI. Electrocardiographic findings in dogs with motor vehicle related trauma.J Am Anirn Hosp Assoc 2001;37:55-62.

26. Marino D, Matthiesen D, Fox PR. Ventricular Arrhythmia in dogs undergoing splenectomy: a prospective study.Veterinary Surgery 1994;23:101-106.

27. Lunney J, Ettinger SJ. Mexiletine administration for management of ventricular arrhythmia in 22 dogs. Journal o f the American Animal Hospital Association 1991;27:597-600.

28. Harpster N. Boxer cardiomyopathy. A review of the long-term benefits of antiarrhythmic therapy. Vet Clin North Am Small Anim Pract 1991;21:989-1004.

29. Calvert CA, Jacobs GJ, Pickus CW. Efficacy and toxicity of tocainide for the treatment of ventricular tachyarrhythmias in Doberman Pinschers with occult cardiomyopathy. Journal o f Veterinary Internal Medicine 1996;10:325-240.

30. Toivonen L. Spontaneous variability in the fequency of ventricular premature complexes over prolonged intervals and implications for antiarrhythmic treatment.Am J Cardiol 1987;60:608-612.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 2

SIGNAL-AVERAGED ELECTROCARDIOGRAPHY

Introduction

Arrhythmogenic cardiomyopathy of Boxers, which has also been referred to as

Boxer Cardiomyopathy and familial ventricular arrhythmias of Boxers, is an inherited

disease of Boxers that predisposes to the development of ventricular arrhythmias,

resulting in syncope or sudden death.1'2 A subset of dogs with this disease may also

develop myocardial failure with progression to congestive heart failure; a condition that

more closely resembles dilated cardiomyopathy in dogs.1 Dogs that manifest

ventricular arrhythmias may never develop clinical signs, may remain asymptomatic for

a variable length of time before developing clinical signs, or may experience sudden

death as the first (and only) manifestation of the disease. Diagnosis of the disease is

typically achieved by the use of routine in-hospital electrocardiography, ambulatory

electrocardiography (Holter monitoring), and 2-dimensional echocardiography.

Standard diagnostic evaluation of these dogs enables, in most (but not all) instances, the

identification affected individuals. However, the identification of affected individuals

that are at risk of developing clinical signs (i.e. syncope and sudden death) is more

problematic.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. In humans, adjunctive diagnostics are used to not only aid in risk stratification,

but identification of arrhythmia mechanism. Most clinically important ventricular

arrhythmias that are associated with sudden death are caused by reentry.3 Reentrant

mechanisms result from abnormalities in either ventricular conduction or

repolarization.4 Identifying the presence of abnormal conduction is important in the

evaluation of ventricular arrhythmias, and is achieved by both invasive and noninvasive

means. Invasive programmed stimulation is commonly performed in people, and has

certain advantages over less invasive diagnostics. In veterinary medicine, however,

these invasive procedures are not commonly performed, and more emphasis is placed

on noninvasive testing.

Signal-averaged electrocardiography (SAECG) is a non-invasive

electrocardiographic technique that identifies the presence of late potentials in the QRS

complex that are otherwise hidden in baseline noise during standard

electrocardiography. These late potentials are believed to represent areas of abnormal

conduction, which may contribute to reentry.5'7 Consequently, the presence of late

potentials serves as pathophysiologic markers of a reentrant substrate. The purpose of

this study was to perform signal-averaged electrocardiography in Boxer dogs to test the

hypothesis that the presence of late potentials increases with disease severity. Thus,

SAECG may prove to be a valuable noninvasive marker of severe disease, and serve to

aid in the identification of at-risk individuals.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Materials and Methods

Inclusion Criteria

Boxer dogs were included for SAECG analysis if they met the following

criteria: a 24 hour ambulatory ECG was obtained, a two-dimensional echocardiogram

was performed to assess function and exclude congenital abnormalities, and a

satisfactory signal-averaged electrocardiogram could be obtained. Dogs included in the

study represent individuals that participated in a larger study as part of a screening

program through The Ohio State University. A group of normal, non-Boxer dogs were

also evaluated by SAECG to serve as a control population. These dogs were screened

by auscultation and physical exam, and were excluded if any abnormalities were

identified. Dogs of mixed breeds that were believed to include either Boxer or

Doberman pinscher, or other breed at high risk for cardiomyopathy, were excluded.

Ambulatory electrocardiography and echocardiography were not performed in these

dogs. Normal dogs were recruited on a voluntary basis, chiefly from the staff and

students associated with the veterinary school at The Ohio State University. Informed

consent was obtained for all dogs evaluated.

Ambulatory electrocardiography

Twenty-four hour ambulatory electrocardiograms3 were obtained using a seven

lead, three channel electrode system, and acquired using an analog tape recorder. Leads

were arranged to approximate the frontal leads I, II, and m (Figure 2.1). Recordings

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. were analyzed using a prospective software analysis algorithm with continuous user

interaction b by a trained cardiology research assistant under the guidance of a

veterinary cardiologist. The presence and complexity of ventricular arrhythmias were

identified. The total number of ventricular premature complexes (VPCs) were

tabulated, and the complexity of the ventricular arrhythmia was graded according to the

following scheme: grade 0 = no VPCs were identified; grade 1 = only single,

monomorphic VPCs were present; grade 2 = single VPCs were identified, and were

present either in a bigeminal or trigeminal pattern, or multiform complexes were

identified; grade 3 = presence of couplets or triplets were present; grade 4 = runs of

ventricular tachycardia (four beats or longer) or the presence of RonT phenomenon.

Cardiac ultrasonography

Cardiac ultrasound0 was performed in unsedated dogs positioned in right lateral

recumbency. Aortic velocity was obtained with a subcostal transducer position to

exclude the presence of congenital valvular or sub valvular aortic stenosis.8 Two-

dimensional and M-mode imaging in the parasternal right and long axis views were

obtained to assess cardiac dimensions and function to identify the presence of

associated myocardial failure or other significant cardiac lesions. Color flow Doppler

was also used when indicated to confirm or exclude the cause of any auscultable

murmurs. Dogs were excluded from the study if they had aortic velocities greater than

2.25 m/s (pressure gradient greater than 20mmHg), suggestive of aortic stenosis. Dogs

were also excluded if there was evidence for other congenital heart disease (i.e.

pulmonic stenosis). Other exclusion criteria included additional congenital lesions or

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. significant acquired valvular disease; these conditions, however, were not encountered

in the dogs evaluated for the study, and therefore no dogs were excluded on these bases.

Signal Averaged Electrocardiography

Signal averaged electrocardiography (SAECG) was similarly performed in

unsedated dogs in right lateral recumbency. Electrodes were placed on the body surface

after hair was clipped and skin debrided to facilitate good skin-electrode interaction.

Leads were placed in such a way as to approximate Leads I, aVF, and V10, creating an

orthogonal lead system comprising the X, Y, and Z axes, respectively. A commercial

SAECG system was usedd to achieve a low noise baseline to facilitate the identification

of late potentials. Data points were sampled at 2000 samples per second and were

filtered with a high pass frequency cutoff of both 25 and 40Hz. At the beginning of the

recording, a template QRS beat was identified to define the “typical” morphology

against which other QRS complexes would be compared to exclude both artifact and

ectopic/abnormal QRS morphologies. This “template beat” was also used to identify

the limits for the region of interest for baseline noise calculation; the isoelectric portion

of the S-T segment was used for this purpose. QRS morphologies that exhibited a 95%

homology with the predetermined template beat were accepted and included in the

analysis. Successive QRS complexes were included in the averaging process until the

averaged value of the baseline noise was at or below 0.75pV. QRS complexes of each

lead were signal-averaged, and then filtered using a 25 Hz and 40 Hz high pass filter.

These averaged and filtered QRS complexes were summated to yield a composite

filtered QRS complex. A satisfactory SAECG allowed evaluation of a sufficient number

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of normal ventricular complexes so as to achieve a final noise of 0.75pV. Dogs with

bundle branch block causing QRS complex widths of over 80 msec were excluded from

evaluation, as identification of late potentials is complicated by pre-existing conduction

defects and wide QRS complexes.9,10

Several parameters were generated during SAECG analysis that were used to

document the presence of late potentials (Figure 2.2). The width of the filtered QRS

complexes (QRSd), as defined by computer determined QRS onset and offset points,

increases in proportion to the severity of late potentials. The duration of time that the

voltage of the terminal QRS complex remains below 40pV before its termination is

termed the low amplitude signal (LAS). As with the QRSd, the duration of LAS is

proportional to degree of late potential. The average value o f the voltage of the terminal

QRS, also known as root mean square (RMS) is another estimate of late potential

activity. Unlike QRSd and LAS, RMS values are inversely proportional to the degree

of late potentials. QRS complexes that do not exhibit late potential activity will

abruptly terminate, yielding a relatively large RMS value. Conversely, the presence of

late potentials allows for a more gradual QRS offset, and results in a smaller RMS

value. In this study, root mean square values were obtained over the terminal 30msec

(RMS30) and 40 msec (RMS40). All parameters were calculated at high pass

frequency cut-off values of both 40Hz and 25Hz. QRSd values remained constant

independent of the high pass frequency filter used. Therefore, seven parameters were

used: QRSd, LAS-40, RMS40-40, RMS30-40, LAS-25, RMS40-25, and RMS30-25.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Subject Categorization

Dogs were grouped on the basis of disease severity, as measured both by VPC

frequency and by clinical assessment. Based on VPC frequency, dogs were grouped

into four categories as follows: group 1 = <10 VPCs/24 hours; group 2 = 10-100

VPCs/24 hours; group 3 = 100-1000VPCs/24 hours; group 4 = >1000VPCs/24 hours.

Based on clinical assessment, dogs were classified as either asymptomatic (A),

symptomatic but not suffering a cardiac death (NCD), or those dogs dying from a

cardiac cause (CD). Dogs were considered symptomatic if a history consistent with

syncope was identified or if a syncopal episode was witnessed in the clinic. Dogs in the

CD category were considered to have suffered a cardiac death if their death was a

complication of congestive heart failure, or a sudden, unexplained death with no other

known concomitant disease capable of causing such a death.

Statistical Analysis

Data were analyzed on the basis of both VPC frequency and clinical assessment.

The number of VPCs were grouped as previously described (groups 1-4), and One-way

Analysis of Variance was used to identify differences in each SAECG parameter based

on VPC frequency group. Similarly, One-way Analysis of Variance was used to

identify differences in each SAECG parameter based on the clinical status of each dog

as categorized previously (A, NCD, CD). One-way Analysis of Variance was also used

to identify differences between the groups of dogs based on clinical status and VPC

frequency (absolute numbers, not grouped). Student’s t-test was used to identify

differences in SAECG parameters between dogs that were symptomatic (CD and NCD)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and those that were asymptomatic. A Student’s t-test was also used to identify

differences in SAECG parameters between symptomatic dogs that died from their

disease (CD) and symptomatic dogs that did not (NCD). Non-parametric tests were

used when indicated (if the data did not exhibit normal distributions). Significance

level was tested at a = 0.05. Tukey’s post hoc tests were performed when a significant

difference was identified.

Results

A total of 89 Boxer dogs satisfied the inclusion criteria, and one additional dog

was included for a total of 90 dogs. The additional dog that was included died before a

24 hour ambulatory ECG could be recorded, but a signal-averaged ECG and

echocardiogram were obtained prior to death. He was included because the information

obtained from echocardiography and in hospital ECG were sufficient to accurately

diagnose and classify his disease, and a Holter would not have provided additional

information in this particular dog. In addition, a total of 49 normal, non-Boxer dogs

were evaluated as a control group.

Of the 90 Boxer dogs, 33 were male (37%) and 57 were female (63%). The age

of the dogs ranged from 1 to 12 years, with a median of 5 years. The number of VPCs

ranged from 0 to 23,699 VPCs/24 hours, with a median of 7 per 24 hour period. Based

on VPC frequency, 48 dogs (53%) were classified into group 1 (<10VPCs/24 hours),

17 dogs (19%) were placed into group 2 (10-100 VPCs/24 hours), 8 dogs (9%)were

placed into group 3 (100-1000 VPCs/24 hours), and 17 dogs (19%) were placed into

group 4 (>1000 VPCs/24 hours). When grouped by clinical status, 69 dogs (77%) were

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. asymptomatic (A), 12 dogs (13%) were symptomatic but did not die of cardiac causes

(NCD), and 9 dogs (10%) suffered a cardiac death (CD), either syncope or CHF.

Descriptive data for both Boxer and normal, non-Boxer dogs for each SAECG

parameter is presented in Tables 2.1 and 2.2, respectively. In addition to minimum,

maximum, mean and median values, the 90th and 95th percentile values are presented

for parameters that increase with degree of late potential severity (QRSd, LAS). For

parameters that vary inversely with the degree of late potentials (RMS40 and RMS30),

the 5,h and 10th percentile are presented.

The 90th/10th and 95th/5th percentiles for each SAECG parameter from the

normal dogs were used to calculate sensitivity, specificity, and positive and negative

predictive values for the Boxer dogs. These results are presented in Tables 2.3 and 2.4.

Finally, the values that optimize sensitivity/specificity and predictive values are

presented in Tables 2.5 and 2.6.

Results of the One-way Analysis of Variance for SAECG parameters, based on

both VPC frequency and clinical status groups, are summarized in Table 2.7. Relative

to VPC frequency, significant differences were identified for RMS40 at both 40 and

25Hz and RMS30 at 40Hz, primarily due to differences between groups 1 and 4. For

clinical status, significant differences were identified for all seven SAECG parameters.

Post hoc evaluation revealed that the CD group was different than the A group in all

parameters, and that the A and NCD group were not different in any parameter.

Differences were identified between the CD and NCD group in all parameters except

LAS at 40 and 25Hz. When clinical status was considered as either symptomatic (CD

or NCD) or asymptomatic, differences were identified in all SAECG parameters except

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. QRSd and LAS at 40Hz. When comparing differences in VPC frequency between dogs

grouped by clinical status, the nonparametric ANOVA on ranks was required due to the

non-normal distribution of VPC frequency. A significant difference was identified

between the groups, and post hoc evaluation (Dunn’s method) identified the major

difference was between group A and CD. Examples of SAECGs taken from a normal

dog and one taken from a dog that subsequently died are presented in Figure 2.3.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Black Orange White Red Green FROM FROM RIGHT FROM LEFT Whi 11. 11. Channeltheconsists of brown electrode2 (+)positive and as asnegativethe black electrode (-), Figure 2.1. 2.1. Figure electrodes Position of placementHolter seenasfor rightboth the and sides. from left There electrodes; create to bipolar pairare 7 3 channels), 3 (orto and serve as ground. leads 1 Channel 1 Channel consists oftheorange3 bluethe electrodepositive(+) electrode and as approximating lead 111. (-), negative as approximating lead I. The electrode serves ground. green as consistspositiveof as electrode the red electrode(+) and the white negativeasapproximating (-), lead Blue Brown

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 1989;63:556-60. Am JAm Cardiol Figure 2.2. 2.2. Figure of descriptiona with a QRS (fQRS) Schematic basic of duration the filtered SAECGparemeters: the of filtered QRS (QRSd), the duration of signal below 40uV durationof the QRS (LAS40),40uV below of (QRSd), signal terminal (RMS4040and voltage msec RMS30, 30 and and respectively). taken JS Figure Steinberg and JT, of Jr. from ofthe Biggerreduction noise analysis endpoint Importance in of the signal-averaged electrocardiogram. electrocardiogram. ofthe signal-averaged in in

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 15 24.5 280.2 95/5% 75.5 23.5 44.5 54.7 143.6 95/5% 75 62.1 76.5 78.5 20.1 16 19 71.3 22.8 120.7 78.5 63 337.5 207 175.1 159 15.2 Median 90/10% 354.8 187.6 246.7 98.6 Median 90/10% 14 14 7.2 7 14 63.9 171 66.7 67.2 74 534.4 530.4 321.2 Mean 375.1 27 15 83 25 625 342.6 710.9 404.8 435.5 238.7 238 Maximum Mean 0 0 22 9.9 10.2 53 29 22 22.1 63.9 41.1 271.2 919.1 Parameter Minimum Parameter Minimum Maximum QRSd (msec) LAS40 (msec) 4.5 LAS25 (msec) LAS25 QRSd (msec) 49.5 81.5 LAS40 (msec) LAS40 LAS25 (msec) LAS25 RMS40-40 (uV)RMS40-40 (uV)RMS30-40 169.1 35.5 459.4 229.8 220.1 RMS40-25 (uV) RMS40-25 RMS40-40 (uV) RMS40-40 RMS30-40 RMS30-40 (uV) RMS40-25 (uV)RMS40-25 RMS30-25 (uV)RMS30-25 30.5 601.5 259.2 Table 2.1. Table 2.1. data Descriptive SAECG parameters 90 in for Boxers. parameters that increase value For in arepercentiles reported. oflatedegreewith 95lh and decrease parameters For that in potentials, the 90th percentiles are reported. th oflatewith degreevalue and 5 potentials, the 10lh increase in value with degree oflate potentials, the 90lh and 95lh percentiles are reported. arepercentiles reported. with degreeoflate parameters value in 95lh increase and For potentials, the 90lh Table 2.2. data 2.2. parameters49Descriptive Table SAECG dogs. for normal, non-Boxerin parameters that For percentilesreported. are These values with degreedecrease in value oflatethat and Sth potentials, the 10th calculationsof subsequently are used dogs. sensitivity, predictivespecificity, thefor and Boxer value for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 94% NPV 100% 100% NPV 100% 100% 94.30% 96.10% 94.70% PPV 25% PPV 35% 36% 100% 91% 100% 93% 24.30% 24.30% 46.20% 65% 80% 100% 100% 87.60% 37.50% 96% 96.30%81.50% 66.60% 96.30% 91.40% 65.40% 87.60% 33.30% 97.50% 71.40% 95.20% 93.80% 54.50% 96.20% 33% 100% 55.60% 55.60% 66.70% Sensitivity Specificity Sensitivity Specificity <75 66.60% >14 <207 100% <62.1 95/5% <175.1 100% 79% <280.2 100% 90/10% Parameter Parameter QRSd(msec) >76.5 44% LAS40 (msec) LAS40 >20.1 66.70% LAS25 (msec) LAS25 QRSd(msec) >78.5 11% 100% LAS40 (msec) LAS40 >24.5 LAS25 (msec)LAS25 >15 55.60% RMS30-25 RMS30-25 (uV) <107.5 RMS40-40 RMS40-40 (uV) RMS30-40(uV) RMS40-25 RMS40-25 (uV) <321.2 100% RMS30-25 RMS30-25 (uV) <119.3 66.70% RMS40-40 RMS40-40 (uV) (uV) RMS30-40 RMS40-25 RMS40-25 (uV) Table 2.3. Table 2.3. positive Sensitivity, specificity, and negative predictive presented values are at cut-offvalues for percentiles of parametersSAECG (fornormal dogs). parameters that For increase value in or 10th the 90th with degree oflatepercentile is used; potentials, the 90th parametersdecrease that value degree in with for oflatepercentile reported. is potentials, the 10th in oflatedegree with value in percentile is reported. potentials, the 5lh Table 2.4. Table 2.4. positive Sensitivity, specificity, predictiveand negative presented values are cut-off for percentilesof SAECG parameters(for dogs). parametersnormal For that or5th valuesat the 9S'h degreewith oflate increase value in percentile used; is potentials, 95th the parameters thatdecrease for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 96% 95% 96% 95% 95% 95% 93% Specificity NPV 83% 99% 71% 98% 100% 100% 94% 67% 67% 96% 56% 44% 56% 83% 99% 56% Sensitivity PPV 1 3 2 67% 75% 98% Number of false positives 4 0 6 5 1 5 5 2 3 4 33% 43% 6 captured (n=9) Numberof CD dogs >77 >23 <84 >18 <67 <76 <147 false positives level to minimize (maximize specificity) Table 2.5. Table 2.5. each optimizeSAECG Values parameterfor that minimize and positivesspecificity ofpresented. are number false (PPV), specificity, valueand negativepredictive value(NPV) each arevalue shown. for The number of CD dogsCDcorrectlyof (outThe numberof the 9 dogs), of identified numbertotal positivepredictive positives, sensitivity, false Parameter QRSd(msec) LAS40 (msec) LAS25 (msec) RMS40-40 (uV) RMS30-40(uV) RMS40-25 (uV) RMS30-25 (uV)

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13% 11% 19% 32% 26% 25% 76% 52% 31% 14% 48% 18% Specificity PPV 19 75 7% 61 39 56 42 positives Numberof false >7 >6 >63 <154 <142 <232 <279 25 69% (sens &NPV=100) level to dogs capturelevel CD all Parameter QRSd (msec) LAS40 (msec) LAS25 LAS25 (msec) RMS40-40 (uV) RMS30-40 (uV) RMS30-25 RMS30-25 (uV) RMS40-25 RMS40-25 (uV) Table 2.6. 2.6. Table each value SAECG The correctly parameterthat for CD dogs, all 9 maximizingand sensitivity identifies negative predictive value (NPV) negativevaluepredictivepresented. is numberof The positives, predictivepositive value specificity, and false (PPV) are value shown. each for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. N/A N/A 1 4 and 1 CD/Aonly CD/A only Groups different CD/A, CD/NCD CD/A, CD/NCD CD/A, CD/NCD CD/A, CD/NCD CD/A, CD/NCD N/A N/A N/A N/A Tukey 4 and 1 Tukey TukeyTukey and 4 1 Tukey Tukey Tukey Tukey Tukey Tukey NS N/A 0.003 0.003 0.015 <.001 <0.001 ORSd NS N/A ORSd LAS40 LAS25 NS LAS25 RMS40-40 RMS30-40 0.018 RMS30-25 NS RMS40-25 <0.001 RMS30-40 <.001 RMS40-40 <.001 RMS40-25 RMS30-25 0.001 SAECG ParameterSAECG p-value posthoc test Status VPC VPC freq. (Groups 1-4) (A, (A, NCD, CD) LAS40 frequency or clinical orstatus. frequency text definitions. See category for Table 2.7. Table 2.7. Summaryofstatistical tests differencesto identifydogs SAECG by in parameters grouped VPC for

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27 . 00 27 . 69 . 00 69. Z V e c to r 21 21 .00 62.00 93 57 . .56 SO 42 107. 70.52 Y Scale:aV/aa S Vector ea|tilade X 19.00 50 . 32 60 . 50 60 . 00 69. 70 147 300 ■! 143.90 117.93 B uV <■•> (•V) (-•) 0 600 300 400 300 100 200 LAS RMS RMS40 300 LAS RMS RMS40 ORSD Figure 2.3. 2.3. Figure ofa An example SAECG doganormal (A) compared abnormalfrom to an an SAECG affected Boxer from QRSdwith and increased adecreased LAS, RMS40. (B) with late potentials that subsequently died. (B) that subsequently potentialswith died. late The SAECG Boxer with abnormal an the and potentials had late Os

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Discussion

To our knowledge, this is the first study to evaluate the utility of SAECG in the

assessment of arrhythmogenic cardiomyopathy in Boxer dogs. The results of this study

document that signal-averaged electrocardiography is a useful test in the evaluation of

this disease in Boxers. We were able to identify differences between groups of dogs

categorized by disease severity, measured both by quantification of VPC frequency and

clinical status. In most instances, the majority of the difference was identified in the

group that was most severely affected. SAECG proved useful in identifying those dogs

that are most likely to suffer a cardiac death, as well as identifying differences between

dogs that were asymptomatic and those that were symptomatic (manifesting either as

syncope or cardiac death). For each parameter, values could be identified that were

associated with either high sensitivity (with high negative predictive value) or high

specificity (with high positive predictive value) for a cardiac death; however, numerous

false positives and false negatives occurred.

Signal-averaged ECG has been used infrequently in veterinary medicine in the

evaluation of canine heart disease. In the assessment of dilated cardiomyopathy in

Doberman pinschers11, and in 4 dogs with sustained ventricular arrhythmias12, the

presence of late potentials exhibited the ability to potentially identify dogs which were

more likely to suffer a cardiac death. In these dogs, the manifestation of cardiac disease

was similar to the subset of Boxers that either have severe arrhythmias or develop

systolic dysfunction and congestive heart failure.

In human medicine, the evaluation and assessment of patients with ventricular

arrhythmias employs a variety of diagnostic approaches. A major goal in the evaluation

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. of such patients is to identify those individuals who are at the highest risk of suffering

adverse events, particularly sudden death. Diagnostic testing used for these purposes

include Holter evaluation, exercise testing, radionucleotide ventrigulography, and

programmed stimulation.9 With the exception of Holter monitoring, the usefulness of

these other modalities is limited in veterinary medicine due to a combination of

expense, invasiveness, and patient compliance. Consequently, this has resulted in the

search for less invasive testing that may be more readily performed in veterinary

patients.

Signal-averaged electrocardiography is a high resolution ECG that allows the

identification of late potentials occurring in the terminal portion of the QRS complex.13

The presence of late potentials is believed to represent areas of abnormal conduction

that contributes to anatomic reentry.7,9 Reentrant circuits are considered the principal

causes of the arrhythmias that predispose to sudden death.3 Thus, the presence of late

potentials is a marker for an arrhythmic substrate, and may identify individuals at risk

for adverse arrhythmic events. Identification of late potentials has been documented as

a predictor of adverse arrhythmic events, particularly in conjunction with Holter

monitoring and assessment of ventricular function.7 Diseases for which SAECG has

been used in the evaluation of ventricular arrhythmias include dilated cardiomyopathy,

hypertrophic cardiomyopathy, arrhythmogenic right ventricular

dysplasia/cardiomyopathy, ischemic heart disease, and nonspecific causes of

syncope.5,7,9,14

The utility of signal-averaged electrocardiography in human medicine is not

limited as a single diagnostic test, but is frequently used for screening of patients to

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. identify at-risk individuals in whom additional testing is indicated, particularly the

electrophysiologic study.6,710 As stated previously, the expense and invasive nature of

EP testing is, for the most part, prohibitive in veterinary medicine. In human studies,

the presence of late potentials as identified by SAECG has predicted the development of

inducible ventricular arrhythmias on EP testing, and SAECG may be used as an

independent diagnostic test.6,7,9

The identification of late potentials on SAECG is dependent on the technique

used, as normal values are highly dependent on recording and filtering technique.6

Identification of late potentials can be performed several ways, including temporal

(time), spectral (frequency), and spatial averaging techniques.6,7 Spatial averaging

allows identification of late potentials of a single QRS complex by placing multiple

electrodes over the body surface. Signal noise is reduced by the comparing closely

spaced electrodes, with the assumption that cardiac electrical activity between these

electrodes is more consistent than random noise. In this way, random noise can be

identified as differences between closely spaced electrodes, allowing the more

consistent cardiac generated potentials to persist. The advantage of such a system is

that it allows identification of late potentials for a single beat, which may be particularly

important in the presence of significant arrhythmia It is also particularly useful when

late potential activity fluctuates as a result of periodic conduction abnormalities. The

disadvantage is the difficulty of the technique, which requires that the patient be

electrically shielded. Additionally, electromyographical noise, which tends to be more

consistent and coherently related to cardiac electrical activity, may not be attenuated to

the same degree as more random electrode/amplifier noise.7

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Spectral, or frequency domain, analysis is achieved by analyzing the frequency

content of the terminal QRS signal. The generation of late potentials, caused by

abnormal fractionated conduction through diseased myocardium, not only causes

changes in conduction velocity, but also contains higher frequency components.7 The

evaluation of the frequency content of the region of interest (i.e. the terminal QRS) is

achieved by utilizing fast Fourier transform (FFT) to decompose a signal into its

frequency components. The identification of high frequency content within a signal

may correlate with patients that have ventricular arrhythmias; however, this technique

has been shown to lack reproducibility and is very sensitive to recording parameters.7

Furthermore, frequency domain analysis has not been shown to be more informative

compared to temporal averaging.7’1015 A more useful application of spectral analysis is

in conjunction with temporal analysis, or so-called spectrotemporal averaging.6,7 For

these reasons, frequency analysis was not performed in this study.

In this study, temporal, or time domain, analysis was used for identification of

late potentials. Unlike spectral analysis, this technique identifies late potentials by

reducing baseline noise through averaging of successive QRS complexes and

application of digital filters. The averaging process allows repeatable, consistent

signals to persist, while eliminating random signals. Temporal averaging is based on

the principle that the late potentials are fixed in time relative to overall cardiac

depolarization, and that noise is a random event. With these assumptions in place,

baseline noise can be reduced as a function of the number of beats averaged,

proportional to Vn, where n=number of averaged beats. Other factors involved include

the initial noise level, the degree of muscular activity, and adequacy of skin

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. preparation.6,7,16,17 An orthogonal lead system is used to record the ECG, and each lead

is individually averaged until a specified end point. Once the endpoint is reached, the

individual leads are mathematically combined into a vector, according to the formula

V(x2 + y2 + z2), where x=lead I, y=lead aVF, and z=lead Vi0. Digital filters are applied

to this composite QRS to facilitate the identification of late potentials, creating a filtered

QRS complex. High pass filters, which allow only certain frequencies above a

particular range to pass unimpeded, are used to eliminate low frequency components,

such as baseline drift, and better identify QRS onset and offset. Low pass filters, which

allow only certain frequencies below a particular range to pass unimpeded, are used to

eliminate high frequency noise, such as electromyographic artifact.7

An important feature of time domain analysis is maintaining a consistent

temporal relationship between the QRS inscription and high frequency activity such as

noise or late potentials. This is achieved with a marker referred to as a the fiducial

point, which serves as a reference point for subsequent noise/late potentials.7 Technical

limitations of temporal averaging are largely dependent on this relationship. Signals

that are not fixed relative the fiducial point, or movement of the fiducial itself, called

jitter, can distort and alter the presence of late potentials.7 Because multiple QRS

complexes are included in the averaging process, the individuals beats must be almost

identical. Therefore, premature complexes, aberrant conduction (such as bundle branch

block), and noisy beats need to be excluded from the analysis.7,10

Another important factor influencing the time domain analysis is the manner of

end-point determination. The end-point of the procedure can be defined by either a pre

determined number o f averaged beats, or by a pre-determined noise level. 5’7'10'16'18

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Both methods have been used, but end-point determination based on noise level may

allow for better reproducibility.7,16 The ideal level of end-point noise has not been

absolutely defined, but is likely to be between lpV and 0.3pV.5'7’10,16'18 There exists a

“trade-off’ between accuracy associated with lower levels of noise, and reproducibility

associated with higher levels of noise.6'7,18 For these reasons, and to increase the chance

that a particular noise level could consistently be achieved in unsedated dogs, we chose

to determine an end-point using a noise level of 0.75pV, regardless of the number of

beats necessary to achieve that level.

Boxers with arrhythmogenic cardiomyopathy manifest pathologic and clinical

similarities to human patients with arrhythmogenic right ventricular

cardiomyopathy/dysplasia (ARVC). ARVC in people is an inherited disease in which

affected individuals are at risk for the development of ventricular arrhythmias, and are

also at an increased risk of sudden death. Myocardial infiltration of fibrous and adipose

tissue is the predominant pathologic finding in these patients, and the degree of

infiltration correlates to severity of disease.5'10,19 The presence of late potentials

identified in patients with ARVC is believed to arise from conduction abnormalities

secondary to myocardial fibrofatty infiltration, and the presence of late potentials is

dependent on the severity of disease.5 This finding supports the conclusion that

SAECG is most useful in identifying abnormal conduction and increased risk of adverse

arrhythmic events in patients with the more severe forms of the disease, suggestive of

more significant myocardial infiltration.5,10 This pattern was also identified in the

current study of Boxer dogs, and lends credence to the similarities between the two

diseases. In our current study, the dogs with more severe disease, measured by either 67

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. VPC frequency or clinical status, were more likely to have late potentials on their

SAECG. Those dogs with less severe disease were often indistinguishable from

unaffected, asymptomatic dogs.

Several limitations exist in the current study. As with many clinical studies in

veterinary medicine, this study suffers due to the relatively low number of dogs

included. This is particularly the case with symptomatic dogs. Based on VPC

frequency, less than 20% of the dogs had >1000VPCs /24 hrs, and almost 50% had less

than 10VPCs/24 hrs. Less than 25% were symptomatic, and o f those only 10% suffered

a cardiac death. Because of the low numbers, no attempt was made to distinguish

between dogs that had died with systolic dysfunction in addition to an arrhythmia from

those that died of without myocardial failure. As the number of dogs evaluated

increases, a distinction between these groups may be possible.

The identification of a particular Boxer as either affected or unaffected can be

very difficult, as no clear diagnostic criteria currently exists for this disease in Boxers.

Consequently, normal values were generated from clinically normal, non-Boxer dogs.

It may be more desirable to compare normal Boxers to abnormal Boxers, but the

uncertainty of identifying a Boxer as “normal” or “abnormal” precludes this distinction

at this time. Thus the designation of symptomatic or asymptomatic was chosen, and

SAECG parameter values were derived from a clinically normal, non-Boxer control

group that was screened for heart disease.

The distinction between symptomatic and asymptomatic dogs was not absolute,

even in the most severely affected groups. Some dogs that died from cardiac causes had

fewer late potentials than asymptomatic animals that were otherwise clinically normal.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reasons for the presence of late potentials in an otherwise healthy individual may be

due to technical artifacts or noise level. It is possible that the noise level chosen was too

low, or spurious signals persist in the averaging process that do not necessarily

represent a true late potential.610 Alternatively, the presence of late potentials

documents the existence of an anatomic substrate for arrhythmia formation. However,

the presence of a reentrant circuit requires additional modifiers or triggering

mechanisms to produce reentry and ventricular arrhythmias.10 Up to 4-6% of normal

controls in people manifest the presence of late potentials.5,7 The absence of late

potentials in otherwise affected individuals may simply reflect a decreased amplitude in

potential, or abnormal potentials that are “buried’ in a baseline with too high a noise

level.5 It is also possible that the mechanism for arrhythmia formation differs between

subsets of patients, and that in some patients the arrhythmias are automatic in nature,

and not reentrant.5

Despite these limitations, the conclusions of this study support the findings

observed in people with ARVC that late potentials do occur in patients with arrhythmic

disease, and the presence of abnormal electrical activity parallels disease severity. We

also conclude that SAECG has clinical utility in Boxers with arrhythmogenic

cardiomyopathy, and future studies may be necessary to determine if different subsets

of these dogs can be differentiated based on the presence of late potentials.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Footnotes

a Cardiocorder cassette recorder, Del Mar, Irvine CA

b Accuplus Holter analysis system, Del Mar, Irvine CA

c System 5, VingMed, Denmark

d Pagewriter Xli with SAECG module M l754, Hewlett Packard,

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reference List

1. Harpster NK. Boxer cardiomyopathy. In: Kirk RW, ed. Current Veterinary Therapy: Small Animal Practice. Philadelphia: WB Saunders, 1983:329-337.

2. Meurs KM, Spier AW, Miller MW, Lehmkuhl L, Towbin JA. Familial Ventricular Arrhythmias in Boxers. J Vet Intern Med 1999;13:437-439.

3. Zipes DP. Sudden cardiac death: Future approaches.Circulation 1992;85(Suppl I):I-160-I-166.

4. Zipes DP. Genesis of cardiac arrhythmias: Electrophysiological considerations. In: Braunwald E, ed. Heart Disease: A textbook of cardiovascular medicine. Philadelphia: W.B. Saunders Company, 1997:548-592.

5. Nava A, Folino AF, Turrini P, Buja GF, Daliento L, Thiene G. Signal-averaged electrocardiogram in patients with arrhythmogenic right ventricular cardiomyopathy and ventricular arrhythmias. Eitr Heart J 2000;21:58-65.

6. Breithardt G, Wichter T, Fetsch T, Seifert T, Borggrefe M, Shenasa M, Reinhardt L. The signal-averaged ECG: Time-domain analysis. Eur Heart J1993; 14 (Suppl. E):27-32.

7. Kuchar DL, Rosenbaum DS. Noninvasive recording of late potentials: Current state of the art. Pacing Clin Electrophysiol 1989;12:1538-1551.

8. Lehmkuhl LB, Bonagura JD. Comparison of transducer placement sites for Doppler echocardiography in dogs with subaortic stenosis.American Journal o f Veterinary Research 1994;55:192-198.

9. Schactman M, Greene JS. Signal-averaged electrocardiography: A new technique for determining which patients may be at risk for sudden cardiac death. Focus on Critical Care 1991;18:202-210.

10. Turrini P, Angelini A, Thiene G, Buja G, Daliento L, Rizzoli G, Nava A. Late potentials and ventricular arrhythmias in arrhythmogenic right ventricular cardiomyopathy. Am J Cardiol 1999;83:1214-1219.

11. Calvert CA, Jacobs GJ, Kraus MS. Possible ventricular late potentials in Doberman Pinschers with occult cardiomyopathy. Journal o f the American Veterinary Medical Association 1998;213:235-239.

12. Calvert CA, Jacobs GJ, Kraus MS, Brown J. Signal-averaged electrocardiograms in normal Doberman Pinschers. Journal o f Veterinary Internal Medicine 1998;12:355-364.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13. Simson MB. Noninvasive identification of patients at high risk for sudden cardiac death: Signal-averaged electrocardiography. Circulation 1992;85(Suppl I):I- 145-1-151.

14. Fauchier L, Babuty D, Cosnay P, Poret P, Rouesnel P, Fauchier JP. Long-term prognostic value of time domain analysis of signal-averaged electrocardiography in idiopathic dilated cardiomyopathy.Am J Cardiol 2000;85:618-623.

15. Calvert CA, Jacobs GJ, Kraus MS. Failure of spectro-temporal mapping to detect ventricular late potentials in dogs.American Journal o f Veterinary Research 1999;60:396-401.

16. Steinberg JS, Bigger JT, Jr. Importance of the endpoint of noise reduction in analysis of the signal-averaged electrocardiogram. Am J Cardiol 1989;63:556-560.

17. Calvert CA, Jacobs GJ, Kraus M, Brown J. Signal-averaged electrocardiograms in normal Doberman pinschers. J Vet Intern Med 1998;12:355-364.

18. Goldberger JJ, Challapalli S, Waligora M, Kadish AH, Johnson DA, Ahmed MW, Inbar S. Uncertainty principle of signal-averaged electrocardiography. Circulation 2000; 101:2909-2915.

19. Thiene G, Basso C, Danieli G. Arrhythmogenic right ventricular cardiomyopathy. TCM 1997;7n3:84-91.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. CHAPTER 3

QT DISPERSION

Introduction

Arrhythmogenic cardiomyopathy of Boxer dogs is commonly encountered in

veterinary medicine, and is characterized by the development of ventricular

tachyarrhythmias resulting in sudden death.1,2 The mechanism of the arrhythmia in

Boxer dogs is unknown. There are three mechanisms postulated for the genesis of

ventricular tachyarrhythmias, including enhanced automaticity, triggered activity, and

reentry. In people, the predominant mechanism of ventricular tachyarrhythmias

resulting in sudden cardiac death is believed to be reentry.3 Reentrant circuits are

created when adjacent myocardial tissues exhibit different electrophysiologic

properties, allowing for the development of abnormal impulse conduction. Tne

disparity in electrophysiologic properties may be due to structural abnormalities or

functional changes. Anatomic abnormalities, including hypertrophy, fibrosis and

ischemia, create areas of differing excitability and electrophysiologic properties

resulting from the presence of abnormal tissue. Functional reentry occurs in the

absence of any identifiable anatomic substrate, presumably as the result of inherent

electrophysiologic differences in membrane and ionic properties of excitable tissue.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Differences in the electrophysiologic properties of excitable tissue may manifest as

inhomogeneity (or heterogeneity) of ventricular depolarization and repolarization.4

The QT interval, measured from the surface electrocardiogram, is an

electrocardiographic parameter of ventricular repolarization.s The duration of the QT

interval is not identical in all leads; this interlead difference between the maximum and

minimum QT durations is known as QT dispersion.6 It is hypothesized that the degree

of variation in QT interval duration among leads (dispersion) reflects regional

differences of ventricular repolarization.5,6 Increased QT dispersion reflects increased

inhomogeneity of ventricular repolarization.7 Dispersion of ventricular repolarization

reduces the threshold for ventricular fibrillation and facilitates the development of

reentrant ventricular tachyarrhythmias. Consequently, QT dispersion may be used as a

“marker of arrhythmogenicity”, and may support a reentrant mechanism.5,8,9

QT dispersion may be an indicator of a reentry and serve as a non-invasive

means to identify electrical dispersion in Boxer dogs with FVA. This could provide

evidence for the presence of an arrhythmogenic substrate, and be useful in identifying

dogs at risk for arrhythmia development and sudden cardiac death. In humans, QT

dispersion has been used with some success for identifying patients at risk for

developing ventricular tachyarrhythmias associated with a number of cardiovascular

diseases, including chronic heart failure (CHF), hypertrophic cardiomyopathy,

myocardial ischemia, aortic stenosis, mitral valve prolapse, and long QT syndrome.6,10'

14 The development of a non-invasive diagnostic test capable of identifying patients

who are most severely affected with FVA that may be at highest risk for developing

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. ventricular tachyarrhythmias and/or sudden death would represent a significant

advancement in the management of Familial ventricular arrhythmias in Boxer dogs.

The purpose of this study was to (1) measure QT interval durations and QT

dispersion in a group of Boxer dogs, and (2) determine if QT parameters correlate with

indices of disease severity of Familial Ventricular Arrhythmias, including the number of

ventricular premature complexes, grade of arrhythmia, and left ventricular function (as

measured by percent fractional shortening).

Materials and Methods

Animal Selection

Twenty-five mature Boxer dogs were recruited for evaluation and evaluated

based upon physical examination, 12-lead electrocardiograms (ECGs), ambulatory

electrocardiography and 2-dimensional echocardiography. Historical information

regarding exercise tolerance and episodes of syncope was also obtained.

12-lead electrocardiography

A 12-lead ECG was obtained in unsedated dogs positioned in right lateral

recumbency. Tracings were recorded at a paper speed of 25 and 50 mm/sec, with a gain

of 10 mm/mV. In some instances, the recording of precordial leads overlapped with the

limb leads, and in these cases the gain of the precordial leads was reduced to 5 mm/mV.

QT intervals were measured manually as the duration from the onset of the QRS

complex to the termination of the T wave.6 Three consecutive beats were measured,

and the RR interval immediately preceding each complex was recorded. A heart rate-

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. corrected QT interval (QTc) was calculated for each beat as the duration of the QT

interval divided by the cube root of the RR interval (measured in seconds) of the

preceding cardiac cycle.15 A mean QT and QTc was calculated for each lead based on

the three measured beats, from which QT and QTc dispersion were determined. QT and

QTc dispersion were defined as the difference between the maximum and minimum

value of the mean QT and QTc for each lead, respectively.6 In addition to QT and QTc

dispersion, an average value for QT and QTc (QTavg and QTcavg) values were calculated

for each dog using the mean values from each of the 12 leads.

24-hour ambulatory electrocardiography

Twenty-four hour ambulatory electrocardiography was performed using three-

channel cassette recorders and analyzed with a Holter analysis system with prospective,

user interaction.8 Total number of ventricular premature complexes (VPCs) and

duration of recording were obtained. Only those recordings at least 20 hours in duration

were included. If VPCs were identified, the arrhythmia was graded based on a

modification of the Lown grading scheme as follows: grade 1= single, uniform VPCs

only; grade 2= presence of bigeminy, trigeminy, or multiform VPCs; grade 3= presence

of couplets or triplets, and grade 4= presence of RonT phenomenon or ventricular

tachycardia. Animals in which no VPCs were identified were classified as grade zero.16

Echocardiography

Two-dimensional echocardiography was performed in unsedated dogs

positioned in right lateral recumbency, using standard techniques. Percent fractional

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. shortening (FS%) was calculated as the percent difference between diastolic (LVIDd)

and systolic (LVEDs) left ventricular dimensions according to the formula FS% =

(LVIDd-LVIDs)/LVIDd X 100.17’18

Statistical evaluation

QT parameters (QT and QTc dispersion, and average QT and QTc) were

evaluated for correlation to total number of VPCs, arrhythmia grade, and percent

fractional shortening. For the purpose of statistical calculations, arrhythmia grade was

converted to a numeric ordinal scale as described. The non-parametric Spearman

correlation coefficient was used to identify significant correlation between the QT

parameters and indices of disease severity. A Mann-Whitney rank sum test was used to

identify a difference between QT dispersion and QTc dispersion. In all cases,

statistical significance was tested at a=0.05.

Results

Twelve-lead electrocardiography, ambulatory electrocardiography, and

echocardiography were performed in 25 Boxer dogs (20 females, 5 males). Descriptive

statistics for total number of VPCs on 24-hour ambulatory ECG (VPC#), grade of

arrhythmia, and percent fractional shortening (FS%), in addition to the QT parameters

(QT and QTc dispersion, average QT and QTc) are shown in Table 3.1. In 22 of 25

dogs, QT dispersion (and consequently QTc dispersion) was zero. Data for the three

dogs with QT dispersion not equal to zero are presented in Table 3.2. Values for QT avg

and QTc avg are within one standard deviation of the mean, and indicators of disease

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. severity do not suggest presence of disease. Two of the 25 dogs had a history of

syncope. In both dogs, QT and QTc dispersion are zero, and QT parameters are either

less than the mean, or are within one standard deviation of the mean. None of the dogs

with non-zero QT dispersion had a history of syncope.

QT dispersion, QTc dispersion, average QT, and average QTc were evaluated

for any correlation to VPC#, arrhythmia grade, and percent fractional shortening (FS%).

No significant correlations between any of the QT parameters and the indices of disease

severity were found.

Evaluation of heart rate correction for QT interval durations revealed that

individual measurements of QT intervals often did not change on a beat-to-beat basis,

despite beat-to-beat variation in heart rate. A Mann-Whitney rank sum test failed to

identify a difference between QT dispersion and QTc dispersion.

Two of the dogs (a 6 year old male and a 10 year old female) were syncopal, and

their QT, Holter, and echocardiographic parameters were as follows: both had QT

dispersion of zero, and QT average (213, 200) and QTc average (230, 259) were below

and within 2 standard deviations of the mean. The VPC# and grade from Holter

recordings were 68 VPCs, grade 2 and 505 VPCs, grade 3, respectively. Percent

fractional shortening (32,25) was normal for each dog.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Parameter Median Mean SD Min Max Range

QT disp(ms) 0 3.9 12.5 0 50 50

QTc disp(ms) 0 4.8 15.6 0 61.3 61.3

QT avg(ms) 213 218.0 18.6 180 253 73

QTc avg(ms) 261 264.5 21.5 230 317 87

Age(yr) 4 4.5 3.2 1 10 9

FS% 31.0 30.5 4.0 20.0 36.0 16.0

VPC#(per day) 5 2,776 12,489 0 62,622 62,622

Arrhythmia Grade 1 2.0 1.4 0 4 4

Heart Rate(bpm) 111 109 19.9 65 148 83

Table 3.1. 25 Boxers with echocardiogram and 24 hour Holter. Median, mean, standard deviation (SD), minimum (min), maximum (max), and range values for QT parameters, indices of disease severity, age, and heart rate o f 25 Boxer dogs evaluated by 12-lead electrocardiography, ambulatory electrocardiography and echocardiography.

QT QTc QT QTc Arrhythmia Age Gender FS% VPC# Heart rate disp disp avg avg grade 50 61.3 224 280 1.5 M 28 1 1 110 7 7.9 223 262 2 F 36 6 3 97 40 50.6 225 276 1.5 F 35 24 2 111

Table 3.2. 3 dogs with non-zero QT dispersion. QT parameters, signalment, and indices of disease severity for the three Boxer dogs in which QT and QTc dispersion are non-zero. Values for QT avg and QTc avg are within one standard deviation of the mean, and indicators of disease severity do not suggest presence o f disease.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Discussion

This is the first study to report the role of QT dispersion in the assessment of FVA

in Boxer dogs. Results obtained in this study indicate that QT interval duration or

dispersion of QT do not correlate with either total number of VPCs, arrhythmia grade,

or percent fractional shortening. The majority of dogs had QT dispersion equal to zero,

implying that substantial dispersion of repolarization was not present in this group of

dogs, and would not be expected to discriminate dogs based on disease severity. There

was no difference between dispersion of QT and QTc intervals, and individual

measurements of QT intervals did not change on a beat-to-beat basis, despite beat-to-

beat variation in heart rate resulting from respiratory sinus arrhythmia. Thus, heart rate

correction of the QT interval did not appear to be necessary for QT dispersion

calculation in this group of dogs.

Human studies evaluating QT dispersion have revealed potential clinical utility

in predicting arrhythmic episodes or sudden death associated with cardiovascular

diseases, such as coronary artery disease, hypertrophic cardiomyopathy, chronic heart

failure, aortic stenosis, mitral valve prolapse, and long-QT syndrome.5,6,10'12’14 Despite

studies that claim clinical usefulness of QT dispersion in identifying patients at risk for

future arrhythmic episodes or sudden death, much controversy still remains. Not all

studies evaluating QT dispersion have arrived at the same conclusions.19 21 Disparity in

findings could result from evaluation of QT dispersion using retrospective study designs

or having protocols that were subject to selection bias. This lack of standardization of

the use of QT dispersion as a diagnostic test makes comparison between studies

difficult, which may account for some of the differences noted.19

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. One of the difficulties in evaluating QT dispersion is with the method of QT

interval measurement itself. There is little difficulty in identifying the onset of the

QRS, but there can be some ambiguity in determining the termination of the T wave.

Low T wave amplitude, variations in T wave morphology, and the presence of a U

wave are all factors that can contribute to the inability to accurately identify termination

of the T wave, resulting in difficulty of QT interval measurement.6 Some studies, in an

effort to eliminate this uncertainty, have excluded leads in which these problems exist.

Because QT dispersion is the difference between the largest and smallest QT interval

between leads, eliminating one or more leads may have an effect on its value. Other

efforts used to avoid complications associated with accurate T wave termination include

measurement of the QT interval from the onset of the QRS complex to the peak of the T

wave (QTpeak)- The peak of the T wave can be more reliably identified than the end of

the T wave, and is thus associated less ambiguity.22 Studies using different techniques

to improve the accuracy of QT measurement have also created a lack of standardization

that contributes to the inability to compare studies.

Results of this study indicate a lack of correlation between QT parameters and

indices of disease severity. As in other studies, the methods used may contribute to

these findings. In this study, QT interval durations were measured manually. Manual

measurement in previous studies has been aided by the use of a digitizer coupled with

magnification to improve resolution and enhance accurate identification of T wave

termination.13,23 Using these criteria, error of measurement has been determined to be

on the order of 20 ms.24 With the use of caliper measurement in this study, accuracy of

measurement was estimated to be 0.25 mm, which corresponds to 10 ms for recordings

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. made at 25mm/sec, and 5 ms for recordings made at SOmm/sec. However, this is an

unsubstantiated estimate, and only takes into account the perceived error of

measurement with the calipers, and does not reflect accurate identification of T wave

termination. In this study, we did not appreciate difficulty in identification of T wave

termination, and did not observe unusual T wave morphology (biphasic T waves,

notching, changes in polarity) or presence of U waves. As a result, we did not use a

digitizer or magnification, nor did we calculate QTpeak intervals. Validation of

computer-free manual measurements may be necessary to fully appreciate the degree of

measurement error. However, the frequency of QT dispersion values of zero and the

magnitude of non-zero measurements, suggest that measurement error is unlikely to

significantly affect the conclusions of this study.

Duration of the QT interval is affected by heart rate, and QT measurements are

frequently corrected for heart rate (QTc). There are many ways in which QT interval

duration can be corrected for heart rate; two of the most commonly used are Bazett’s

formula and Fridericia’s formula.15 Correcting for heart rate in humans may be

unnecessary as adjustments based on Bazett’s formula (square root) are negligible at

heart rates between 50-80 beats per minute.25 In dogs, normal heart rates are higher,

with rates greater than 120 beats per minute frequently occurring in dogs in this study.

Because Bazett’s formula is believed to be less accurate at higher heart rates, we chose

to correct the QT interval using Fridericia’s (cube root) method.26 However, the

importance of heart rate correction for the purpose of QT dispersion calculation in dogs

is unclear. Furthermore, due to the relatively slow response of QT duration changes as

a result of changing heart rate, the practice of measuring RR intervals immediately

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. preceding each beat may be complicated in the presence of respiratory sinus arrhythmia,

which commonly occurs in dogs.25,27 The necessity of heart rate correction for QT

dispersion is also unclear.23,28 In some investigators’ opinion, the most serious

drawback to the use of heart rate corrected QT durations is that studies evaluating QT

dispersion identify differences between patient populations based on differences in

underlying heart rate, and not true differences in repolarization.25 Most studies

examining QT dispersion have reported values in terms of QTc, which may represent a

confounding factor in the interpretation of their results.

In this study, QT and heart rate corrected QT (QTc) measurements were

determined, and no difference between QT and QTc dispersion was identified.

Furthermore, individual measurements of QT intervals did not change on a beat-to-beat

basis, despite beat-to-beat variation in heart rate resulting from respiratory sinus

arrhythmia. Therefore, the use of heart rate correction of the QT interval did not appear

to be necessary for QT dispersion calculation in this group of dogs.

Limitations in this study may arise from the assumption that percent fractional

shortening obtained from 2-dimensional echocardiograms and the frequency and

severity (grade) of the arrhythmia taken from 24-hour ambulatory electrocardiograms

are useful in quantifying the severity of FVA in Boxer dogs. The association between

echocardiographic and electrocardiographic parameters and FVA has not been

demonstrated. The lack of correlation between QT dispersion and the results of these

diagnostic tests may imply the lack of significant heterogeneity of repolarization in dogs

with FVA, or it may simply reflect the inability of these diagnostic parameters to

quantify disease severity. However, these parameters are commonly used in the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. diagnosis of FVA, and serve to guide therapy and provide prognostic information in the

clinical management of the disease.1,2 The gender distribution in this group of dogs was

skewed (20 female, 5 male), which may have influenced the results. The discrepancy

most likely is a result of the fact that most animals evaluated are owned by breeders,

and females were more commonly presented. However, no evidence supporting a

difference in QT parameters based on gender exists in dogs. Furthermore, previous

evaluation of VPC# and arrhythmia grade did not identify significant differences based

on gender.b The low numbers of dogs available for evaluation may have also

contributed to our inability to demonstrate significant correlation between QT

parameters and these indices of disease severity. More studies including a greater

number of animals may be needed to validate these findings.

Conclusions

The results of this study suggest that QT interval duration and QT dispersion do not

correlate with indices of disease severity for FVA, including percent fractional

shortening, number o f VPCs, and grade of arrhythmia. Furthermore, heart rate

correction of the QT interval for the purposes in calculating QT dispersion may not be

necessary.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Footnotes

a. Del Mar Avionics, Irvine Ca

b. Spier AW, Meurs KM, Lehmkuhl LB, et al. Evaluation of ambulatory ECG

monitoring in asymptomatic Boxer dogs. (Abstract) J Vet Intern Med 1999; 13

248.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reference List

1. Meurs KM, Spier AW, Miller MW, et al. Familial ventricular arrhythmias in Boxers. J Vet Intern Med 1999;13:437-439.

2. Harpster N. Boxer Cardiomyopathy. Vet Clin North Am Small Anim Pract 1991;21:989-1004.

3. Zipes DP. Sudden cardiac death: Future approaches.Circulation 1992;85(Suppl I): I-160-1-166

4. Zipes DP. Genesis of cardiac arrhythmias: Electrophysiological considerations. In: Braunwald E, ed. Heart Disease: A textbook o f cardiovascular medicine. 5th Ed. Philadelphia: W.B. Saunders Company, 1997;548-592.

5. Kautzner J, Malik M. QT interval dispersion and its clinical utility. Pacing Clin Electrophysiol 1997;20(Part II):2625-2640.

6. Day CP, McComb JM, Campbell RWF. QT dispersion: An indication of arrhythmia risk in patients with long QT intervals. Br Heart J 1990;63:342-344.

7. Zabel M, Portnoy S, Franz MR. Electrocardiographic indexes of dispersion of ventricular repolarization: An isolated heart validation study.J Am Coll Cardiol 1995;25:746-752.

8. Han J, Moe GK. Nonuniform recovery of excitability in ventricular muscle. Circ Res 1964; 14:44-60.

9. Peeters HA, Sippensgroenewegen A, Wever EFD, et al. Electrocardiographic identification of abnormal ventricular depolarization and repolarization in patients with idiopathic ventricular fibrillation. J Am Coll Cardiol 1998;31:1406-1413.

10. Tielemen RG, Crijns HJGM, Wiesfeld ACP, et al. Increased dispersion of refractoriness in the absence of QT prolongation in patients with mitral valve prolapse and ventricular arrhythmias. Br Heart J 1995;73:37-40.

11. Barr CS, Naas A, Freeman M, et al. QT dispersion and sudden unexpected death in chronic heart failure. Lancet 1994;343:327-329.

12. Ducceschi V, Sarubbi B, D’Andrea A, et al. Increased QT dispersion and other repolarization abnormalities as a possible cause of electrical instability in isolated aortic stenosis. IntJ Cardiol 1998;64:57-62.

13. Savelieva I, Yi G, Guo X, et al. Agreement and reproducibility of automatic versus manual measurement of QT interval and QT dispersion.Am J Cardiol 1998;81:471-477. 86

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 14. Higham PD, Fumiss SS, Campbell RWF. QT dispersion and components of the QT interval in ischaemia and infarction. Br Heart J 1995;73:32-36.

15. Puddu PE, Jouve R, Mariotti S, et al. Evaluation of 10 QT prediction formulas in 881 middle-aged men from the seven countries study: Emphasis on the cubic root Fridericia's equation. J Electrocardiol 1988;21:219-229.

16. Lown B, Calvert C, Armington R. Monitoring for serious arrhythmias and high risk of sudden death. Circulation 1975;51-52(Suppl 3):HI-189-111-198

17. Bonagura JD, O'Grady MR, Herring DS. Echocardiography: Principles of interpretation. Vet Clin North Am Small Anim Pract 1985;15:1177-1194.

18. Vuille C, Weyman AE. Left Ventricle I: General considerations, assessment of chamber size and function. In: Weyman AE, ed. Principles and Practice of Echocardiography. 2nd Ed. Philadelphia: Lea& Febiger, 1994;575-624.

19. Zabel M, Klingenheben T, Franz MR, et al. Assessment of QT dispersion for prediction of mortality or arrhythmic events after myocardial infarction. Circulation 1998;97:2543-2550.

20. Fei L, Goldman JH, Prasad K, et al. QT dispersion and RR variations on 12-lead ECGs in patients with congestive heart failure secondary to idiopathic dilated cardiomyopathy. Eur Heart J1996; 17:258-263.

21. Fu GS, Meissner A, Simon R. Repolarization dispersion and sudden cardiac death in patients with impaired left ventricular function.Eur Heart J 1997;18:281- 289.

22. Manttari M, Oikarinen L, Manninen V, et al. QT dispersion as a risk factor for sudden cardiac death and fatal myocardial infarction in a coronary risk population. Heart 1997;78:268-272.

23. Glancy JM, Weston PJ, Bhullar HK, et al. Reproducibility and automatic measurement of QT dispersion. Eur Heart J 1996;17:1035-1039.

24. Hill JA, Friedman PL. Measurement of QT interval and QT dispersion. Lancet 1997;349:894-895.

25. Malik M, Camm AJ. Mystery o f QTc interval dispersion. Am J Cardiol 1997;79:785-787.

26. Molnar J, Weiss J, Zhang F, et al. Evaluation of five correction formulas using a software-assisted method of continuous QT measurement from 24-hour Holter recordings. Am J Cardiol 1996;78:920-926.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 27. Maison-Blanche P, Coumel P. Changes in repolarization dynamicity and the assessment of the arrhythmic risk. Pacing Clin Electrophysiol 1997;20(Part II):2614- 2624.

28. Zabel M, Woosley RL, Franz MR. Is dispersion of ventricular repolarization rate dependent? Pacing Clin Electrophysiol 1997;20(Part I):2495-2411.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Chapter 4

Heart Rate Variability

Introduction

The autonomic nervous system plays a major role in the control of the

cardiovascular system, particularly in the modulation of heart rate.M The degree of

variation in heart rate, or heart rate variability, can be used as an assessment of

autonomic influence on the heart.2'5 Increased sympathetic innervation decreases the

variability of RR intervals, while parasympathetic tone increases the degree of variation

in RR intervals. Sympathetic and parasympathetic tone not only has an impact on the

variation in heart rate, but can also affect heart rhythm. Autonomic balance influences

the occurrence and severity of ventricular arrhythmias, with sympathetic tone

exacerbating these arrhythmias, and parasympathetic tone having a protective effect.5

Reduced heart rate variability, associated with increased sympathetic tone and

withdrawal of parasympathetic tone, has been used to predict mortality in patients with

congestive heart failure and identify risk of sudden death in patients with ventricular

arrythmias.3’6"8 In veterinary medicine, analysis of heart rate variability has been used

to evaluate Doberman pinschers with dilated cardiomyopathy.9'11 Most dogs with

dilated cardiomyopathy develop congestive heart failure, with a smaller subset of

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. patients dying suddenly from cardiac arrhythmias.12,13 By contrast, Boxers manifest a

type of primary heart disease characterized by the presence of ventricular arrhythmias

and risk of sudden death, with a smaller subset of patients developing systolic

dysfunction and congestive heart failure.14 The purpose of this study was to (I)

measure heart rate variability in Boxers with and without ventricular arrhythmias, (2)

assess the ability of this method to identify Boxers based on disease severity, and (3)

use heart rate variability to determine if persistently elevated sympathetic tone is present

in affected dogs.

Materials and Methods

Patient selection

Boxer dogs were included in this study as part of a multi-phased research study

evaluating ventricular arrhythmias in Boxer dogs. All dogs were evaluated by physical

exam, standard in-hospital electrocardiography, 2-dimensional and Doppler

echocardiography, and 24-hour ambulatory electrocardiography (Holter monitoring).

Dogs were included in the study if they were over 2 years old with no evidence of

congenital heart disease.

Subject categorization

Dogs were classified according to the number of VPCs on Holter evaluation.

Dogs with <2 VPCs/24 horns were classified as unaffected (U). Dogs with >1000

VPCs/24 hours were classified as affected (A). A group of dogs with congestive heart

failure (HF) were also included to serve as reference group expected to have reduced

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. heart rate variability and increased sympathetic tone. A group of normal, non-Boxer

dogs (N) were also included to serve as a control population for the Boxers. These dogs

were evaluated by physical examination, standard in-hospital electrocardiography, and

Holter evaluation.

Ambulatory electrocardiography

Twenty-four hour ambulatory electrocardiograms3 were obtained using a seven

lead, three channel electrode system, and acquired using an analog tape recorder. Lead

were arranged to approximate the frontal leads I, II, and III (Figure 4.1). Recordings

were analyzed using a prospective software analysis algorithm with continuous user

interaction b by a trained cardiology research assistant under the guidance of a

veterinary cardiologist.

Heart rate variability analysis

Analysis of heart rate variability was performed on 24-hour Holter recordings

using a commercial laboratory6. Heart rate variability analysis was performed in the

time domain. Parameters that were calculated included the average normal R-R interval

(mean RR) and the standard deviation of the normal R-R intervals (SDNN) over the

entire recording period. Additional parameters were calculated by dividing the 24-hour

recording into 288 separate five minute periods. The mean and standard deviation (SD)

of R-R intervals for each five minute period were then calculated. The standard

deviation of the mean R-R interval for each five minute period (SDANN) and the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. average of the standard deviation of R-R intervals for each five minute period

(ASDNN) were calculated for the entire 24-hour ambulatory (Holter) recording.

Statistical evaluation

Data were analyzed by comparing the mean value for each time domain

parameter between groups of dogs using a one-way Analysis of Variance (ANOVA).

Differences were evaluated between the four groups of dogs, including normal, non-

Boxer dogs (N), affected (A) and unaffected (U) Boxers, and Boxers with heart failure

(HF). Significance was defined at a<0.05. If a statistically significant difference was

identified, a post-hoc analysis (Tukey pairwise comparison) was performed.

Results

A total of 34 dogs were included in the study. A total of 24 Boxers were

included, which is comprised of 10 dogs in the affected (A) group, 10 dogs in the

unaffected (U) group, and 4 in the heart failure (HF) group. Ten normal, non-Boxer (N)

dogs were also included. Of the Boxers, the mean age was 6.0 years, with 10 males

(42%) and 14 females (58%). Summary data for each individual group is presented in

Table 4.1, including the mean values for each parameter in each group of dogs. Results

of One-way ANOVA analysis identified differences between the HF group and all other

groups (N, A, and U). No differences between affected (A) and unaffected (U) Boxers

were identified in any parameter. Variable differences were identified between non-

CHF Boxers (A and U) and normal, non-Boxer (N) dogs. A summary of analysis of

variance findings is presented in Tables 4.2-4.5. Graphical representations of the

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. comparisons of heart rate variability parameters (mean RR, SDNN, ASDNN, SDANN)

and identification of significant differences between groups are presented in Figures

4.2-4.5.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Black Orange White Red Green FROM RIGHT FROM FROM LEFT Whi Figure 4.1 4.1 Figure ofelectrodes Position placement as seen both Holterthe and right sides. for left from There create bipolarelectrodes; are 7 to pair 3 3 (orchannels), and as to serve leads ground. 1 Channel 1 II. consiststheofbrown2 Channel electrodepositive (+) as black electrode(-), and the negative as Channel approximatingconsistsoflead orange electrode3 III. the the (+) positive blueas electrode and consists electrodepositive asred of the and the electrode white (+) as (-), negative approximating lead asnegative(-), approximating lead I. The green electrode serves ground. as Blue Brown

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 175.5 243.5 81.25 (msec) SDANN 280 171.9 283 335.1 (msec) ASDNN 151.8 121.8 292.8 327.9 411.6 SDNN (msec) 530 (msec) Mean RR RR Mean 0.6 913.8 0.7 734.1 VPC# 4188.7 734.9 2385.25 50% 4 30% 5.8 Group Age %Female Normal Normal (n=10) Affected (n=10) 8.2 60% Unaffected (n=10)Unaffected 3.6 60% Heart (n=4) Heart failure Table 4.1. Table data Summary age, gender, oftotal ventricularnumberprematurefor heart (VPC#), complexesrate and parameters includingvariability R-R mean the (mean standardinterval and durations deviation ofRR) RR interval over the entire overthe recoerding24 hour period; the mean value deviation the standard ofperiods formin 5 individual within the recording hour 24 and (ASDNN) the standard deviation ofmean the periods min of5 values individual within the 24 hour recordingthewithinhour 24 presented. (SDANN) are also

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.007 0.009 <0.001 - u HF NS - <0.001 0.002 U U Group A U NS N A MeanRR (msec) Group A ^ - 200 1000 Figure and 4.2. Figure Table of Comparison mean groups between dogs. of values RR Statistical significance is represented tabularin well asdesignation; as by form with notwere groups similarletter to letters found A=affected Boxers, heart with HF=Boxers failure. be significantlydifferent one-way by ANOVA. dogs, non-Boxer N=normal U=unaffected Boxers, VO Ov

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. HF HF 0.007 - NS 0.047 0.023 <0.001 - NS NS Group U SDNN (msec) SDNN U A N Group A U N 300 600 Figure and Table 4.3. and Table Figure Comparisonmeanof SDNN ofdogs. valuesbetween groups Statistical A=affectedBoxers, U=unaffected HF=BoxersBoxers, with heart failure. significance is significance represented is tabularas in well designation; as by form groups with similar letter letters were not to were not be differentsignificantlyfound byone-way ANOVA. dogs, non-Boxer N=normal vo

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.002 0.003 <0.001 - U HF NS - A Group U NS A N NS NS ASDNN (msec) Group Figure and Table 4.4. and4.4. Table Figure Comparison ofofdogs. ASDNN mean values between groups Statistical significance is represented significance represented in tabular as well is as designation; groups by form with similarletter letters be to not dilferentwere significantly found by one-way ANOVA. non-Boxerdogs, N=normal A=affected U=unaffected Boxers, with Boxers, HF=Boxers heart failure. 4 0 0$

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 0.002 <0.001 - NS 0.002 - A U HF SDANN SDANN (msec) U NS A N <0.001 0.001 Group N N U Group A HF -

0 ; 50 - 150 - 100 300 250 - 200 significance is represented is designation; significance tabularas as bywell form with similarin letter groups letters withBoxers, U=unaffected heartBoxers, A=affected HF=Boxers failure. Figure and Table 4.S. and Table 4.S. Figure of mean Comparison betweenof values dogs. SDANN groups Statistical were not todifferent were one-way befound significantly by ANOVA. dogs, N=normal non-Boxer

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Discussion

This study evaluated the analysis of heart rate variability (HRV) in Boxer dogs

with ventricular arrhythmias. The results of this study identified a reduction in heart

rate variability in a subpopulation of Boxers with congestive heart failure (CHF).

However, affected Boxer dogs with greater than 1000 ventricular premature complexes

(VPCs) per 24 hours did not demonstrate any differences in HRV analysis compared to

unaffected dogs. In many instances, HRV parameters in normal non-Boxer dogs did

not differ from those of non-CHF Boxers, independent of their affected status. In every

case, however, Boxers with CHF differed in HRV analysis from all other dog groups,

including affected dogs, unaffected dogs, and normal non-Boxer dogs.

The results of this study suggest that HRV analysis in Boxers is similar to

studies performed in people, particularly in patients with CHF. In people, patients with

CHF have elevated sympathetic tone, causing a reduction in HRV.15'22 In fact, the

reduction of HRV in patients with CHF is predictive for mortality, especially with

dilated cardiomyopathy.6,8,23 The assessment of mortality in Boxers with CHF was not

performed, partly due to the number of dogs included (n=4), but also because Boxers

with CHF and systolic dysfunction are known to have a poor prognosis.14,24 The major

value of including the subset of Boxers with heart failure is to validate the technique

used to assess HRV; the significant reduction of HRV in this group of Boxers was

expected, and serves as a marker for increased sympathetic tone.

More intriguing, however, is the assessment of HRV analysis in patients with

ventricular arrhythmias in the absence of myocardial failure. The majority of Boxers

with ventricular arrhythmias have preserved systolic function, and the nature of the

100

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. arrhythmia is not well understood. The formation of ventricular arrhythmias is

complex, and is multifactorial in nature. Arrhythmias develop as a result of an

arrhythmic substrate that can be modified by several factors.25 The existence of an

arrhythmia may not occur despite the presence of a substrate if the necessary triggers

are not present. An important trigger for many arrhythmias is autonomic modulation,

particularly sympathetic tone.1 In Boxers with arrhythmic disease, many reports of

syncope or sudden death occur in association with increased activity or excitement,

suggestive of an increased catecholamine surge.26 The utility of HRV analysis in

people with ventricular arrhythmias has been shown.6,7 The finding in this study that

Boxers with frequent arrhythmias do not have reduced HRV may simply reflect a lack

of persistently increased sympathetic tone, despite the potential importance sympathetic

effects on arrhythmia development. It also appears that in some cases Boxers may have

different autonomic patterns relative to other breeds, as evidenced by the variable

differences between affected and unaffected Boxers compared to non-Boxer dogs.

These findings are supported by previous evaluations of HRV analysis in dogs, where

cardiomyopathic Dobermans with mild or moderate ventricular dysfunction did not

demonstrate reduced HRV relative to clinically normal dogs.10 Severe myocardial

dysfunction may be required to cause substantial elevations in sympathetic tone in order

to overcome the profound normal sinus arrhythmia found in dogs.10,11

The small number of dogs included in the study is a potential explanation of

why no differences were identified between affected and unaffected dogs. However, in

many instances, the values for HRV parameters were nearly identical in these groups,

and the magnitude of differences between both these groups and the normal non-Boxer

101

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. dogs and heart failure dogs was at times significant. The normal non-Boxer group was

always greater (even if not significantly so) than both the affected and unaffected group,

and the heart failure group was always less. Furthermore, the trend between the

affected and unaffected group was such that the unaffected group had lower values

compared to the affected group; a finding opposite of what would be expected. The low

number of heart failure dogs in the study did not prevent identification of significant

differences between this group and the other dogs.

The age of the affected group of Boxers was significantly greater than the

unaffected group of dogs. It is possible that the inability to identify a difference

between the two groups of dogs was because of the effect of age. However, it has been

shown that as Boxers get older the number of VPCs increases; therefore, this

relationship is expected.*1 Furthermore, the influence of age on HRV analysis in dogs is

unknown. Thus, the importance of the age difference between the affected and

unaffected dogs is not clear.

In conclusion, the analysis of HRV in Boxer dogs with ventricular arrhythmias

is most reduced in those dogs with concomitant congestive heart failure. It is in this

group of dogs that sympathetic tone is expected to be increased, particularly in a

persistent fashion. The inability to identify a difference between Boxers with and

without severe arrhythmias may be due to a lack of persistently elevated sympathetic

innervation associated with the presence of arrhythmias, or may reflect the insensitivity

of HRV analysis in this group of dogs. It is likely that sympathetic modulation is

important in the development or manifestation of ventricular arrhythmias, but may exist

in a more phasic or intermittent fashion. Development of other techniques for the

102

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. purpose of assessing sympathetic innervation to the heart may be necessary in the

evaluation of Boxer dogs with ventricular arrhythmias due to arrhythmogenic

cardiomyopathy.

103

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Footnotes

a Cardiocorder cassette recorder, DelMar, Irvine CA

b Accuplus Holter analysis system, DelMar, Irvine CA

c Biomedical systems, St. Louis MO

d Spier AW, Meurs KM, Lehmkuhl LB, Miller MW. Evaluation of ambulatory ECG monitoring in asymptomatic boxer dogs. 17th Annual Veterinary Medical Forum of the American College of Veterinary Internal Medicine, June 1999. (abstract)

104

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. Reference List

1. Schwartz PJ. The autonomic nervous system and sudden death.Eur Heart J 1998;19:F72-F80.

2. Malik M. Heart rate variability. Standards of measurement, physiological interpretation, and clinical use.Eur Heart J 1996;17:354-381.

3. Stein PK, Kleiger RE. Insights from the study of heart rate variability. Annu Rev Med 1999;50:249-261.

4. Stein PK, Bosner MS, Kleiger RE, Conger BM. Heart rate variability: A measure of cardiac autonomic tone.Am Heart J 1994;127:1376-1381.

5. Hohnloser SH, Klingenheben T, Zabel M, Li YG. Heart rate variability used as an arrhythmia risk stratifier after myocardial infarction. Pacing Clin Electrophysiol l997;20(Ptn):2594-2601.

6. Fauchier L, Babuty D, Cosnay P, Fauchier JP. Prognostic value of heart rate variability for sudden death and major arrhythmic events in patients with idiopathic dilated cardiomyopathy. J Am Coll Cardiol 1998;33:1203-1207.

7. Bikkina M, Alpert MA, Mukeiji R, Mulekar M, Cheng BY, Mukeiji V. Diminished short-term heartrate variability predicts inducible ventricular tachycardia. Chest 1998;113:312-316.

8. Fauchier L, Babuty D, Cosnay P, Autret ML, Fauchier JP. Heart rate variability in idiopathic dilated cardiomyopathy: Characteristics and prognostic value.J Am Coll Cardiol 1997;30:1009-1014.

9. Calvert CA, Wall TM. Correlations among time and frequency measures of heart rate variability recorded by use of a Holter monitor in overtly healthy Doberman pinschers with and without echocardiographic evidence of dilated cardiomyopathy. Am J Vet Res 2001 ;62:1787-1792.

10. Calvert CA, Jacobs GJ. Heart rate variability in Doberman pinschers with and without echocardiocardiographic evidence of dilated cardiomyopathy.Am J Vet Res 2000;61:506-511.

11. Minors SL, O'Grady MR. Heart rate variability in the dog: Is it too variable? Can J Vet Res 1997;61:134-144.

12. Calvert CA, Chapman WL, Toal RL. Congestive cardiomyopathy in Doberman pinscher dogs. J Am Vet Med Assoc 1982; 181:598-602.

105

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 13. Calvert CA, Hall G, Jacobs G, Pickus C. Clinical and pathologic findings in Doberman pinschers with occult cardiomyopathy that died suddenly or developed congestive heart failure.J Am Vet Med Assoc 1997;210:505-511.

14. Harpster NK. Boxer cardiomyopathy. In: Kirk RW, ed. Current Veterinary Therapy: Small Animal Practice. Philadelphia: WB Saunders, 1983:329-337.

15. Tygesen H, Rundqvist B, Waagstein F, Wennerblom B. Heart rate variability measurement correlates with cardiac norepinephrine spillover in congestive heart failure. Am J Cardiol 2001 ;87:1308-1311.

16. Jiang W, Hathaway WR, McNulty S, Larsen RL, Hansley KL, Zhang Y, O'Connor CM. Ability of heart rate variability to predict prognosis in patients with advanced congestive heart failure. Am J Cardiol 1997;80:808-811.

17. Ferrari R, Ceconi C, Curello S, Visioli O. The neuroendocrine and sympathetic nervous system in congestive heart failure. Eur Heart J1998; 19 (Suppl F):F45-F51.

18. Galinier M, Pathak A, Fourcade J, Androdias C, Cumier D, Vamous S, Massabuau P, Boveda S, Fauvel M, Senard JM, Bounhoure JP. Depressed low frequency power of heart rate variability as an independent predictor of sudden death in chronic heart failure. Eur Heart J 2000;21:475-482.

19. Bonaduce D, Petretta M, Marciano F, Vicario MLE, Apicella C, Rao MAE, Nicolai E, Volpe M. Independent and incremental prognostic value of heart rate variability in patients with chronic heart failure. Am Heart J 1999;138:273-284.

20. Boveda S, Galinier M, Pathak A, Fourcade J, Dongay B, Benchendikh D, Massabuau P, Fauvel JM, Senara JM, Bounhoure JP. Prognostic value of heart rate variability in time domain analysis in congestive heart failure. J Intervent Card Electrophysiol 2001;5:181-187.

21. Yoshikawa T, Baba A, Akaishi M, Mitamura H, Ogawa S, Suzuki M, Negishi K, Takahashi T, Murayama A. Neurohormaonal activations in congestive heart failure: Correlations with cardiac function, heart rate variability, and baroreceptor sensitivity. Am Heart J 1999;137:666-671.

22. Wijbenga JAM, Balk AHHM, Meij SH, Simoons ML, Malik M. Heart rate variability index in congestive heart failure: Relation to clinical variables and prognosis. Eur Heart J 1998;19:1719-1724.

23. Yi G, Goldman JH, Keeling PJ, Reardon M, McKenna WJ, Malik M. Heart rate variability in dilated cardiomyopathy: Relation to disease severity and prognosis. Heart 1997;77:108-114.

24. Harpster N. Boxer cardiomyopathy. A review of the long-term benefits of antiarrhydimic therapy. Vet Clin North Am Small Anim Pract 1991;21:989-1004. 106

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 25. Calvert CA. Heart rate variability. Vet Clin North Am Small Anim Pract 1998;28:1409-1427.

26. Goodwin, J. K. and Cattiny, G. Further characterization o f Boxer cardiomyopathy. Proceedings American College of Veterinary Internal Medicine Forum(13th), 300-302. 1995. Lake Buena Vista, FL. 1995.

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. SUMMARY AND CONCLUSIONS

The goal of this project was to examine the clinical utility of non-invasive

diagnostic tests in the evaluation of Boxers with ventricular arrhythmias. The impetus

behind such an endeavor arose from the current inability to adequately assess the

prognosis of affected individuals. Dogs presenting with systolic dysfunction and

congestive heart failure represent a minority of patients with the disease, and have a

natural history that is more straightforward. These dogs are more likely to suffer a

cardiac death, either as a result of sudden death or due to complications associated with

congestive heart failure. The more typical manifestation of the disease involves the

presence of arrhythmias that may or may not be associated with symptoms of syncope.

In this scenario, the course of disease can be profoundly variable, with some dogs

suffering from sudden death as a result of their ventricular arrhythmia, to dogs that live

a normal lifespan only to eventually succumb to non-cardiac related diseases. Holter

monitor evaluation may be valuable in documenting the presence and severity of

disease, but is often unable to allow the clinician to evaluate prognosis.

In human medicine, such evaluations are accomplished by use of

electrophysiologic testing. While these procedures are helpful in identifying individuals

at risk for complications or adverse events, they are inherently invasive, expensive, and

not without risk. In veterinary medicine, these procedures can be accomplished, but the

108

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. need for anesthesia, expense, and invasive nature has precluded them from widespread

use. As a result, identification of less expensive noninvasive diagnostic tests that do not

require anesthesia would improve the management of disease. Such tests currently

employed in human medicine include Holter monitoring, signal-averaged ECG,

assessment of QT dispersion, and analysis of heart rate variability.

Holter monitoring is frequently used to diagnose ventricular arrhythmias in

Boxers, and also serves to monitor disease progression and response to therapy. To

accurately assess either disease progression or antiarrhythmic effect, the inherent or

spontaneous variability of arrhythmia frequency should be known. Parameters exist in

human medicine to guide such evaluations, but the degree of variability in dogs has not

been reported. The evaluation of spontaneous variability of ventricular arrhythmias in

this study resulted in very similar findings to that in people. When frequent arrhythmias

are present, less than an 80% change in the number of VPCs may be within the

physiologic range of spontaneous variability. Dogs with less frequent arrhythmias may

have even more variability in the number of VPCs. However, the grade or complexity

of the arrhythmia appears to be less variable than the absolute number. Therefore, when

assessing response to therapy or monitoring for disease progression, a change of more

than 80% should be identified before documenting a drug response or true disease

progression.

The presence of ventricular arrhythmias depends on the existence of a substrate

for arrhythmia formation. Altered impulse conduction can create abnormal electrical

pathways, which may manifest as reentrant loops or circuits. Reentry is a common

mechanism of arrhythmia formation, and identification of abnormal conduction often

109

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. heralds the existence of ventricular arrhythmias. Noninvasive evaluation of abnormal

conduction can be achieved by the use of signal-averaged electrocardiography. SAECG

facilitates the identification of late potentials, which correlate to areas of abnormal

tissue depolarization. Documentation of late potentials is associated with increased risk

of mortality, and is useful in assessing prognosis and identifying at-risk individuals.

The evaluation of SAECG in Boxers proved to be useful in identifying more severely

affected individuals that were likely to suffer a cardiac death. However, many of the

dogs with late potentials were dogs with poor cardiac function in addition to ventricular

arrhythmias. Additional studies that include more dogs will be necessary to identify its

utility in dogs with arrhythmias in the absence of myocardial failure. Despite its

potential limitations, this technique may prove to be valuable in the evaluation of

Boxers with ventricular arrhythmias.

In addition to abnormal conduction, altered repolarization can also contribute to

reentry and facilitate arrhythmia formation. Repolarization of cardiac muscle may be

directly measured during an electrophysiologic study, but noninvasive assessment of

refractoriness can be performed from a surface ECG. Heterogeneity of repolarization

manifests on a surface ECG as a difference in the duration of QT duration between

leads for any given beat. This so-called QT dispersion correlates with a risk of

mortality in people, but did not appear to be as useful in Boxers with ventricular

arrhythmias. No correlation between QT parameters and disease severity was

identified, and most dogs did not exhibit any dispersion at all (QT dispersion=0). Dogs

that did exhibit some degree of dispersion were not dogs with severe arrhythmias, and

110

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. the magnitude of dispersion was modest. Therefore, evaluation of QT dispersion does

not appear to be a useful diagnostic test in Boxers with ventricular arrhythmias.

The arrhythmia substrate is only partially responsible for the manifestation of

arrhythmias. Other factors, including autonomic balance, have important modulating

effects on arrhythmia formation. Elevated sympathetic tone exacerbates the tendency

for arrhythmia development while parasympathetic tone tends to have a protective

effect. One way to assess the degree of autonomic modulation of the heart is to analyze

the variation of heart rate. As sympathetic tone increases, the variability in heart rate

decreases. Reduced heart rate variability has been used to identify cardiac patients at

risk for adverse events, especially patients with dilated cardiomyopathy, coronary artery

disease, and ventricular arrhythmias. Analysis of heart rate variability in Boxers

revealed that the only reduction in HRV was identified in those patients with congestive

heart failure. These patients were expected to have increased levels o f sympathetic

tone, and served to validate the technique. In contrast, there did not appear to be any

value of HRV analysis in Boxers with ventricular arrhythmias and preserved cardiac

function. Analysis of HRV in these dogs may not be sensitive enough to identify

alterations in autonomic balance, or these dogs may simply not have increased

sympathetic tone. Other diagnostics may be necessary to detect changes in autonomic

tone in these dogs.

In conclusion, it appears that the use of noninvasive testing to evaluate Boxers

with ventricular arrhythmias is limited. Those tests that did identify differences

between groups of dogs tended to reflect changes in only the most severely affected

individuals. Those dogs that are asymptomatic are not as likely to have abnormal tests,

111

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. and continue to represent a challenge in disease management. The use of more invasive

diagnostic testing may be required to fully evaluate these patients and enable the

veterinary clinician to adequately identify patients at risk of developing significant

complications of disease.

112

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. BIBLIOGRAPHY

1. Ajiki K, Murakawa Y, Yanagisawa-Miwa A. Autonomic nervous system activity in idiopathic dilated cardiomyopathy and in hypertrophic cardiomyopathy.Am J Cardiol 1993;71:1316-1320.

2. Amlie JP. QT dispersion and sudden cardiac death.Eur Heart J 1997;18:189-190.

3. Anastasiou-Nana MI, Karagounis LA, Kanakakis J, Kouvelas NE, Geramoutsos A, Chalkias K, Karelas J, Nanas JN. Correlation and stability of heart rate and ventricular ectopy variability in patients with congestive heart failure.Am J Cardiol 2001 ;88:175- 179.

4. Anderson G, Greenspan M. Ventricular ectopic beats with exit block: A retrospective holter monitor study.Electrocardiology 1983;16:133-140.

5. Antzelevitch C, Shimizu W, Yan GX, Sicouri S. Cellular basis for QT dispersion.J Electrocardiol 1997;30 (Suppl):168-175.

6. Antzelevitch C, Sicouri S. Clinical relevance of cardiac arrhythmias generated by afterdepolarizations: Role of M cells in the generation of U waves, triggered activity, and Torsade de Pointes. J Am Coll Cardiol 1994;23:259-277.

7. Armoundas AA, Rosenbaum DS, Ruskin JN, Garan H, Cohen RJ. Prognostic significance of electrical altemans versus signal averaged electrocardiography in predicting the outcome of electrophysiological testing and arrythmia-free survival,br heart j1998;80:251-256.

8. Aronson D, Burger AJ. Effect of beta-blockade on heart rate variability in decompensated heart failure.IntJ Cardiol 2001;79:31-39.

9. Barr CS, Naas A, Freeman M, et al. QT dispersion and sudden unexpected death in chronic heart failure. Lancet 1994;343:327-329.

10. Barron HV, Lesh MD. Autonomic nervous system and sudden cardiac death.J Am Coll Cardiol 1996;27:1053-1060.

11. Bazett H. An analysis of the time-relations of electrocardiograms, br heart j 1920;7:353-370. 113

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 12. Benn M, Hansen PS, Pedersen AK. QT Dispersion in Patients with Arrhythmogenic Right Ventricular Dysplasia.Eur Heart J 1999;20:764-770.

13. Bhargava V. Limitation of signal-averaged ECGs and measurement of late potentials in the presence of noise.JElectrocardiol 1994;27:353-355.

14. Bikkina M, Alpert MA, Mukeiji R, Mulekar M, Cheng BY, Mukeiji V. Diminished short-term heartrate variability predicts inducible ventricular tachycardia. Chest 1998;113:312-316.

15. Bleifer S, Bleifer D, Hansmann D. Diagnosis of occult arrhythmias by holter electrocardiography. Prog Cardiovasc Dis 1974;26:569-599.

16. Bogun F, Chan KK, Harvey M, Goyal R, Castellani M, Niebauer M, Daoud E, Man KC, Strickberger A, Moraday F. QT dispersion in nonsustained ventricular tachycardia and coronary artery disease. Am J Cardiol 1996;77:256-259.

17. Bonaduce D, Petretta M, Marciano F, Vicario MLE, Apicella C, Rao MAE, Nicolai E, Volpe M. Independent and incremental prognostic value of heart rate variability in patients with chronic heart failure.Am Heart J 1999;138:273-284.

18. Bonagura JD, O'Grady MR, Herring DS. Echocardiography: Principles of interpretation. Vet Clin North Am Small Anim Pract 1985; 15:1177-1194.

19. Boveda S, Galinier M, Pathak A, Fourcade J, Dongay B, Benchendikh D, Massabuau P, Fauvel JM, Senara JM, Bounhoure JP. Prognostic value of heart rate variability in time domain analysis in congestive heart failure. J Intervent Card Electrophysiol 2001 ;5:181 -187.

20. Breithardt G, Wichter T, Fetsch T, Seifert T, Borggrefe M, Shenasa M, Reinhardt L. The signal-averaged ECG: Time-domain analysis. Eur Heart J 1993;14 (Suppl. E):27- 32.

21. Brendorp B, Elming H, Jun L, Kober L, Malik M, Jensen GB, Torp-pedersen C. QT Dispersion has no prognostic Information for Patients with Advanced Dongestive Heart Failure and REduced Left Ventricular Systolic Function.Circulation 2001; 103:831- 835.

22. Bright JM, Cali JV. Clinical usefulness of Cardiac Event Recording in Dogs and Cats examined because of syncope, episodic collapse, or intermittent weakness: 60 cases (1997-1999). JAVMA 2000;216:1110-1114.

23. Brodsky M, Wu D, Denes P. Arrhythmias documented by 24 hour continous electrocardiographic minitoring in 50 male medical students without apparent heart disease. Am J Cardiol 1977;39:390-395.

114

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 24. Buckingham TA, Thessen CM, Hertweck D, Janosik DL, Kennedy HL. Signal- averaged electrocardiography in the time and frequency domains.Am J Cardiol 1989;63:820-825.

25. Calvert C, Jacobs G, Pickus C. Bradycardia-associated episodic weakness, syncope, and aborted sudden death in cardiomyopathic doberman pinschers. J Vet Intern Med 1996;10:88-93.

26. Calvert CA, Chapman WL, Toal RL. Congestive cardiomyopathy in Doberman pinscher dogs. J Am Vet Med Assoc 1982; 181:598-602.

27. Calvert CA, Hall G, Jacobs G, Pickus C. Clinical and pathologic findings in Doberman pinschers with occult cardiomyopathy that died suddenly or developed congestive heart failure.J Am Vet Med Assoc 1997;210:505-511.

28. Calvert CA, Jacobs G, Pickus CW, Smith DD. Results of Ambulatory Electrocardiography in Overtly healthy Doberman Pinschers with Echocardiographic Abnormalities. J Am Vet Med Assoc 2000;217:1328-1332.

29. Calvert CA, Jacobs GJ, Kraus M, Brown J. Signal-averaged electrocardiograms in normal Doberman pinschers. J Vet Intern Med 1998;12:355-364.

30. Calvert CA, Jacobs GJ, Kraus MS. Failure of spectro-temporal mapping to detect ventricular late potentials in dogs.Am J Vet Res 1999;60:396-401.

31. Calvert CA, Jacobs GJ, Kraus MS. Possible ventricular late potentials in Doberman Pinschers with occult cardiomyopathy.J Am Vet Med Assoc 1998;213:235-239.

32. Calvert CA, Jacobs GJ, Pickus CW, Smith DD. Results of ambulatory electrocardiography in overtly healthy Doberman Pinschers with echocardiographic abnormalities. J Am Vet Med Assoc 2000;217:1328-1332.

33. Calvert CA, Jacobs GJ, Pickus CW. Efficacy and toxicity of tocainide for the treatment of ventricular tachyarrhythmias in Doberman Pinschers with occult cardiomyopathy. J Vet Intern Medl996; 10:325-240.

34. Calvert CA, Jacobs GJ, Smith DD, Tathbun SL, Pickus CW. Association between results of ambulatory electrocardiography and development of cardiomyopathy during long-term follow-up of doberman pinschers. J Am Vet Med Assoc 2000;216:34-39.

35. Calvert CA, Jacobs GJ. Heart rate variability in Doberman pinschers with and without echocardiographic evidence of dilated cardiomyopathy.Am J Vet Res 2000;61:506-511.

36. Calvert CA, Wall M. Effect of severity of myocardial failure on heart rate variability in doberman pinschers with and without echocardiographic evidence of dilated cardiomyopathy. J Am Vet Med Assoc 2001;219:1084-1088. 115

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 37. Calvert CA, Wall M. Results of ambulatory electrocardiography in overtly healthy Doberman Pinschers with equivocal echocardiographic evidence of dilated cardiomyopathy. J Am Vet Med Assoc 2001;219:782-784.

38. Calvert CA, Wall TM. Correlations among time and frequency measures of heart rate variability recorded by use of a Holter monitor in overtly healthy Doberman pinschers with and without echocardiographic evidence of dilated cardiomyopathy.Am J Vet Res 2001 ;62:1787-1792.

39. Calvert CA. Heart rate variability. Vet Clin North Am Small Anim Pract 1998;28:1409-1427.

40. Calvert, C. A. High Resolution Electrocardiography (Signal Averaged Electrocardiography) in Normal Doberman Pinschers and Doberman Pinschers with Occult Cardiomyopathy, proc 9th AC VIM Forum, 693-695. 1991. New Orleans.

41. Cheng TO. Decreased Heart Rate Variability as a Predictor for sudden Death was known in China in the third Century A.D. Eur Heart J. 2000;21:2081-2082.

42. Cohen TJ, Goldner B, Merkatz K, Jadonath R, Adler H, Ehrlich JC. A simple electrocardiographic algorithm for detecting ventricular tachycardia.Pacing Clin Electrophysiol 1997;20(Part I):2412-2418.

43. Compton SJ, Lux RL, Ramsey MR, Strelich KR, Sanguinetti MC, Green LS, Keating MT, Mason JW. Genetically defined therapy of inherited long-QT syndrome. Correction of abnormal repolarization by potassium. Circulation 1996;94:1018-1022.

44. Constable PD, Hinchcliff KW, Olson JL, Stepien RL. Effects of Endurance Training on Standard and signal-averaged Electrocardiograms of Sled Dogs. Am J vet Res 2000;61:582-588.

45. Copie X, Eiou M, Lavergne T, Guize L, Le Heuzey JY. Measurement of QT Interval, cardiac electrophysiology review 1997;3:357-359.

46. Coumel P. Diagnostic and prognostic values and limitations of Holter monitoring. Heart Journal 1989; 10 (Suppl E): 19-30.

47. Dabrowski A, Kramarz E, Piotrowicz R, Kubik L. Predictive power of increased qt dispersion in ventricular extrasystoles and in sinus beats for risk stratification after myocardial infarction. Circulation 2000; 101 :1693.

48. Davies LC, Francis DP, Ponikowski P, Piepoli MF, Coats AJ. Relation of heart rate and blood pressure turbulence following premature ventricular complexes to baroreflex sensitivity in chronic congestive heart failure. Am J Cardiol 2001;87:737-742.

49. Day C, McComb J, Matthews J, Campbell RWF. Reduction in QT dispersion by sotalol following myocardial infarction. Br Heart J 1991;12:423-427. 116

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 50. Day CP, McComb JM, Campbell RWF. QT dispersion: An indication of arrhythmia risk in patients with long QT intervals. Br Heart J 1990;63:342-344.

51. de Backer G, Jacobs D, Prineas R. Ventricular premature contractions: A randomized non-drug intervention trial in normal men. Circulation 1979;59:762-768.

52. de Bruyne MC, Hoes A, Kors J, Hoffman A, van Bemmel J, Grobbee D. QTc dispersion predicts cardiac mortality in the elderly. Circulation 1998;97:467-472.

53. Dekker JM, Crow RS, Folsom AR, Hannan PJ, Liao D, Swenne CA, Schouten EG. Low Heart rate Variability in a 2-Minute Rhythm Strip Predicts Risk of Coronary Heart Disease and Mortality from Several Causes. Circulation 2000; 102:1239-?

54. Denes P. The importance of derived 12-lead electrocardiography in the interpretation of arrhythmias detected by holter recording. Am Heart J 1992; 124:905- 911.

55. Di Marco J, Philbrick J. Use of ambulatory eclectocardiographic (Holter) monitoring. Ann Intern Med 1990; 113:53-68.

56. Dickinson D, Scott O. Ambulatory elecrocardiographic monitoring in 100 healthy teenage boys. Br Heart J 1984;51:179-183.

57. Doven O, Sayin T, Giildal M, Karaoguz R, Oral D. Heart rate variability in hypertrophic obstructive cardiomyopathy: Association with functional classification and left ventricular outflow gradient.IntJ Cardiol 2001;77:281-286.

58. Ducceschi V, Sarubbi B, D'Andrea A, et al. Increased QT dispersion and other repolarization abnormalities as a possible cause of electrical instability in isolated aortic stenosis. IntJ Cardiol 1998;64:57-62.

59. Engler R, Ryan W, LeWinter M, Bluestein H, Karliner JS. Assessment of long-term antiarrhythmic therapy: studies on the long-term efficacy and toxicity of tocainide. Am J Cardiol 1979;43:612-618.

60. Fauchier JP, Fauchier L, Babuty D. Time-domain signal averaged electrocardiogram in nonischemic ventricular tachycardia. Pacing Clin Electrophysiol 1996;19:231-244.

61. Fauchier L, Babuty D, Cosnay P, Autret ML, Fauchier JP. Heart rate variability in idiopathic dilated cardiomyopathy: Characteristics and prognostic value.J Am Coll Cardiol 1997;30:1009-1014.

62. Fauchier L, Babuty D, Cosnay P, Fauchier JP. Prognostic value of heart rate variability for sudden death and major arrhythmic events in patients with idiopathic dilated cardiomyopathy. J Am Coll Cardiol 1998;33:1203-1207.

117

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 63. Fauchier L, Babuty D, Cosnay P, Poret P, Rouesnel P, Fauchier JP. Long-term prognostic value of time domain analysis of signal-averaged electrocardiography in idiopathic dilated cardiomyopathy.Am J Cardiol 2000;85:618-623.

64. Fei L, Goldman JH, Prasad K, et al. QT dispersion and RR variations on 12-lead ECGs in patients with congestive heart failure secondary to idiopathic dilated cardiomyopathy. Eur Heart J1996; 17:258-263.

65. Ferrari R, Ceconi C, Curello S, Visioli O. The neuroendocrine and sympathetic nervous system in congestive heart failure. Eur Heart J1998; 19 (Suppl F):F45-F51.

66. Fetcher G, Cantwell J. Continuous ambulatory electroardiographic monitoring. Chest 1977;71:27-32.

67. Fontaine J. Evaluation of patients with complex ventricular arrhythmias: Current noninvasive and invasive methods. Am Heart J 1992; 123:1123-1129.

68. Forslund L, Bjorkander I, Ericson M, Held C, Kahan T, Rehnqvist N, Hjemdahl P. Prognostic implications of autonomic function assessed by analyses of catecholamines and heart rate variability in stable angina pectoris. Heart 2002;87:415-422.

69. Freitas J, Lago P, Puig J. Circadian heart rate variability rhythm in shift workers. J Electrocardiol 1997;30:39-44.

70. Fu GS, Meissner A, Simon R. Repolarization dispersion and sudden cardiac death in patients with impaired left ventricular function.Eur Heart J 1997;18:281-289.

71. Galinier M, Pathak A, Fourcade J, Androdias C, Cumier D, Vamous S, Massabuau P, Boveda S, Fauvel M, Senard JM, Bounhoure JP. Depressed low frequency power of heart rate variability as an independent predictor of sudden death in chronic heart failure. Eur Heart J 2000;21:475-482.

72. Glancy JM, Weston PJ, Bhullar HK, et al. Reproducibility and automatic measurement of QT dispersion. Eur Heart J1996; 17:1035-1039.

73. Goldberger JJ, Challapalli S, Waligora M, Kadish AH, Johnson DA, Ahmed MW, Inbar S. Uncertainty principle of signal-averaged electrocardiography. Circulation 2000;101:2909-2915.

74. Goldner B, Brandspiegel HZ, Horwitz L, Jadonath R, Cohen TJ. Utility of QT dispersion combined with the signal-averaged electrocardiogram in detecting patients susceptible to ventricular tachyarrhythmia.Am J Cardiol 1995;76:1192-1193.

75. Gomes J, Winters S. Identification of patients with hight risk of arrhythmic mortality. Cardiol Clin 1993;11:55-63.

118

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 76. Goodwin, J. K. and Cattiny, G. Further characterization of Boxer cardiomyopathy. Proceedings American College of Veterinary Internal Medicine Forum(13th), 300-302. 1995. Lake Buena Vista, FL. 1995.

77. Graboys T. Appropriate indications for ambulatory electrocardiographic monitoring. Cardiol Clin 1992;10:551-554.

78. Grimm W, Glaveris C, Hoffmann J, Menz V, Muller H-H, Hufnagel G, Maisch B. Arrhythmia risk stratification in idiopathic dilated cardiomyopathy based on electrocardiography and 12-lead, signal-averaged, and 24-hours Holter electrocardiography. Am Heart J 2000;140:43-51.

79. Grimm W, Steder U, Menz V, Hoffman J, Grote F, Maisch B. Clinical Significance of increased QT dispersion in the 12-lead standard ECG for arrhythmia risk prediction in Dilated Cardiomyopathy. Pacing Clin Electrophysiol 1996;19(part II): 1886-1889.

80. Grimm W, Steder U, Menz V, Hoffman J, Maisch B. QT dispersion and arrhythmic events in idiopathic dilated cardiomyopathy. Am J Cardiol 1996;78:458-461.

81. Hall L, Dunn J, Delaney M. Ambulatory electrocardiography in dogs. Vet Rec 1991;129:213-216.

82. Hamdan M, Scheinman M. Current approaches in patients with ventricular tachyarrhythmias. Contemp Issues in Cardiol 1995;79:1097-1120.

83. Han J, Moe GK. Nonuniform recovery of excitability in ventricular muscle. Circ Res 1964;14:44-60.

84. Hanton G, Nahas K, Priou C, Rabemampianina Y, Baneux P. The QT interval in the dog ecg:importance in preclinical toxicology and relationship to heart rate.Toxicology Methods 2000; 11:21 -40.

85. Harpster N. Boxer cardiomyopathy. A review of the long-term benefits of antiarrhythmic therapy. Vet Clin North Am Small Anim Pract 1991;21:989-1004.

86. Harpster NK. Boxer cardiomyopathy. In: Kirk RW, ed. Current Veterinary Therapy: Small Animal Practice. Philadelphia: WB Saunders, 1983:329-337.

87. Harrison D, Fitzgerald J, Winkle R. Ambulatory electrocardiography for diagnosis and treatment of cardiac arrhythmias. N Engl J Med 1976;294:373-380.

88. Harrison D, Fitzgerald J, Winkle R. Contribution of ambulatory eclectocardiographic monitoring to antiarrhythmic management. Am J Cardiol 1978;41:996-1004.

119

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 89. Hermida J, Minassian A, Jarry G. Familial incidence of late ventricular potentials and electrocardiographic abnormalities in arrhythmogenic right ventricular dysplasia.J Am Coll Cardiol 1997;79:1375-1380.

90. Higham PD, Fumiss SS, Campbell RWF. QT dispersion and components of the QT interval in ischaemia and infarction. Br Heart J 1995;73:32-36.

91. Hill JA, Friedman PL. Measurement of QT interval and QT dispersion. Lancet 1997;349:894-895.

92. Hinkle L, Carver S, Argyros D. The prognostic significance of ventricular premature contractions in healthy people and in people with coronary heart disease. Acta Cardiol 1974;suppl. 18:5-32.

93. Hinkle L, Carver S, Stevens M. The frequency of Asymptomatic disturbances of cardiac rhythm and conduction in middle-aged men. Am J Cardiol 1969;24:629-650.

94. Hoffman J, Grimm W, Menz V, Muller HH, Maisch B. Heart rate variability and baroreflex sensitivity in idiopathic dilated cardiomyopathy. Heart 2000;83:531-536.

95. Hohnloser SH, Klingenheben T, Zabel M, Li YG. Heart rate variability used as an arrhythmia risk stratifier after myocardial infarction. Pacing Clin Electrophysiol 1997;20 (Pt II):2594-2601.

96. Hohnloser SH, Klingenheben T, Zabel M, Li YG. Heart rate variability used as an arrhythmia risk stratifier after myocardial infarction. Pacing Clin Electrophysiol 1997;20(PtII):2594-2601.

97. Holter N. New methods for heart studies. Science 1961;134:1214-1220.

98. Horowitz L. Ventricular arrhythmias: control of therapy by holter monitoring. Eur Heart J 1989;10:53-60.

99. Iellamo F, Legramante JM, Massaro M, Raimondi G, Galante A. Effects of a Residential Exercise Training on Baroreflex Sensivity and Heart Rate Variability in Patients with coronary Artery Disease: A Randomized, Controlled Study. Circulation 2000;102:2588-2592.

100. Jansson K, Hagerman I, Ostlund R, Karlberg KE, Nylander E, Nyquist O, Dahlstrom U. The effects of metoprolol and captopril on heart rate variability in patients with idiopathic dilated cardiomyopathy. Clin Cardiol 1998;22:397-402.

101. Jiang W, Hathaway WR, McNulty S, Larsen RL, Hansley KL, Zhang Y, O'Connor CM. Ability o f heart rate variability to predict prognosis in patients with advanced congestive heart failure. Am J Cardiol 1997;80:808-811.

120

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 102. Kapoor WN, Karpf M, Wieand S, Peterson JR, Levey GS. A Prospective Evaluation and Follow-up of Patients with Syncope.The New England Journal o f Medicine 1983;309:197-204.

103. Karagounis LA, Anderson JL, Morena FL, Sorenson SG. Multivariate associates of QT dispersion in patients with acute myocardial infarction: primacy of patency status of the infarct-related artery. Am Heart J 1998;135:1027-1035.

104. Kautzner J, Malik M. QT interval dispersion and its clinical utility.Pacing Clin Electrophysiol 1997;20(Part II):2625-2640.

105. Kazmierczak J, de Sutter J, Tavemier R, Cuvelier C, Dimmer C, Jordaens L. Electrocardiographic and morphometric features in patients with ventricular tachycardia of right ventricular origin. Heart 1998;79:388-393.

106. Kennedy H, Chandra V, Caralis D. Effectiveness of increasing hours of continuous ambulatory electrocardiography in detecting maximal ventricular ectopy. Am J Cardiol 1978;42:925-930.

107. Kennedy H, Underhill S. Frequent or complex ventricular ectopy in apparently healthy subjects. Am J Cardiol 1976;38:141-148.

108. Kennedy HL, Whitlock JA, Sprague MK, Kennedy LJ, Buckingham TA, Goldberg RJ. Long-term follow-up of asymptomatic healthy subjects with frequent and complex ventricular ectopy.New Engl J Med 1985;12:193-197.

109. Kessler DK, Kessler KM, Myerburg RJ. Ambulatory Electrocardiography: A Cost per Management Decision Analysis. Archives o f Internal Medicine 1995;155:165-169.

110. Kinlay S, Leitch J, Neil A. Cardiac event recorders yield more diagnoses and are more cost-effective than 48-hour holter moitoring in patients with palpitations.Ann Intern Med 1996;124:16-20.

111. Kinoshita O, Fontaine G, Rosas F. Optimal high-pass filter setting of the signal- averaged electrocardiogram in patients with arrhythmogenic right ventricular dysplasia. Am J Cardiol 1994;74:1074-1075.

112. Kinoshita O, Rosas F, Elias J. Time-and Frequency-domain analyses of the signal- averaged ECG in patients with arrhythmogenic right ventricular dysplasia.Circulation 1995;91:715-721.

113. Kleiger RE, Miller JP, Bigger JT, Jr., Moss AJ. Decreased herat rate variability and its association with increased mortality after acute myocardial infarction. Am J Cardiol 1987;59:256-262.

121

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 114. Knight BP, Pelosi F, Michaud GF, Strickberger SA, Morady F. Clinical Consequences of Electrocardiographic Artifact Mimicking Ventricular Tachycardia. The New England Journal o f Medicine 1999;341:1270-1274.

115. Kristal-Boneh E, Raifel M, Froom P, Ribak J. Heart rate variability in health and disease. Scand J Work Environ Health 1995;21:85-95.

116. Krupienicz A, Czamecki R, Adamus J. QT dispersion magnitude is related to the redpiratory phase in healthy subjects.Am J Cardiol 1998.

117. Kuchar DL, Rosenbaum DS. Noninvasive recording of late potentials: Current state of the art. Pacing Clin Electrophysiol 1989;12:1538-1551.

118. Kulan K, Komsuoglu B, Tuncer C, Kulan C. Significance of QT dispersion on ventricular arrhythmias in mitral valve prolapse.IntJ Cardiol 1996;54:251-257.

119. Kushner LI, Calvert CA, Boyle CR. Effects of acepromazine and buprenorphine on measured indices of the signal-averaged electrocardiogram in healthy dogs. Am J vet Res 1996;57:1511-1514.

120. Lander P, Beribari J. Use of high-pass filtering to detect late potentials in the signal-averaged ECG. JElectrocardiol 1989;22 (Suppl.):7-12.

121. Lantz DA. Efficacy of holter monitors. Ann Intern Med 1996; 125:697-698.

122. Lee KW, Okin PM, Kligfield P, Stein KM, Lerman BB. Precordial QT dispersion and inducible ventricular tachycardia. Am Heart J 1997;134:1005-1013.

123. Lehmkuhl LB, Bonagura JD. Comparison of transducer placement sites for Doppler echocardiography in dogs with subaortic stenosis.Am J Vet Res 1994;55:192- 198.

124. Lewis B, Antman E, Graboys T. Detailed analysis of 24 hour ambulatory electrocardiographic recordings during ventricular fibrillation or torsade de pointes.J Am Coll Cardiol 1983;2:426-436.

125. Lipski J, Cohen L, Espinoza J. Value of holter monitoring in assessing cardiac arrhythmias in symptomatic patients.Am J Cardiol 1976;37:102-107.

126. Lombardi F. Chaos Theory, Heart Rate variability, and Arrhythmic Mortality. Circulation 2000; 101:8.

127. Lopera GA, Huikuri HV, Makikallio TH, Tapanainen J, Chakko S, Mitrani RD, Interian A, Castellanos A, Myerburg RJ. Is abnormal heart rate variability a specific feature of congestive heart failure?Am J Cardiol 2001;87:1211-1213.

122

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 128. Lopes M, Runge P, Harrison D. Comparison of 24 versus 12 hours of ambularoty ECG monitoring. Chest 1975;67:269-273.

129. Lown B, Calvert C, Armington R. Monitoring for serious arrhythmias and high risk of sudden death. Circulation 1975;51-52 (Suppl 3):IIM 89-111-198.

130. Lown B, Tykocinski M, Garfein A. Sleep and ventricular premature beats. Circulation 1973;108:691-700.

131. Lown B, Wolf M. Approaches to Sudden Death from Coronary Heart Disease. Circulation 1971;447:130-142.

132. Lunney J, Ettinger SJ. Mexiletine administration for management of ventricular arrhythmia in 22 dogs. J Am Anim Hosp Assoc 1991;27:597-600.

133. Macallum GE, Houston BJ. Characterization of Cardiac Alterations in Nonsedated Cynomolgus Monkeys. Am J vet Res 1993;54:327-332.

134. Maison-Blanche P, Coumel P. Changes in repolarization dynamicity and the assessment of the arrhythmic risk. Pacing Clin Electrophysiol 1997;20(Part II):2614- 2624.

135. Makikallio TH, Huikuri HV, Hintze U, Videbaek J, Moller M. Fractal analysis and time- and frequency-domain measures of heart rate variability as predictors of mortality in patients with heart failure.Am J Cardiol 2001 ;87:178-182.

136. Malik M, Camm AJ. Mystery of QTc interval dispersion. Am J Cardiol 1997;79:785-787.

137. Malik M. Heart rate variability. Standards o f measurement, physiological interpretation, and clinical use.Eur Heart J 1996;17:354-381.

138. Manttari M, Oikarinen L, Manninen V, Viitasalo M. Qt dispersion as a risk factor for sudden cardiac death and fatal myocardial infarction in a coronary risk population. Heart 1997;78:268-272.

139. Marino D, Matthiesen D, Fox PR. Ventricular Arrhythmia in dogs undergoing splenectomy: a prospective study.Vet Surg 1994;23:101-106.

140. Martin GJ, Adams SL, Martin HG, Mathews J, Zull D, Scanlon PJ. Prospective Evaluation of Syncope.Annals o f Emergency Medicine 1984;13:499-504.

141. Massin MM, Derkenne B, von Bemuth G. Correlations between indices of heart rate variability in healthy children and children with congenital heart disease. Cardiology 1999;91:109-113.

123

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 142. Mehta D, Camm AJ. Signal-averaged electrocardiography and the significance of late potentials in patients with "idiopathic” ventricular tachycardia: A review.Clin Cardiol 1989;12:307-312.

143. Mehta D, Goldman M, David O. Value of quantitative measurement of signal- averaged electrocardiographic variables in arrhythmogenic right ventricular dysplasia: Correlation with echocardiographic right ventricular cavity dimensions.J Am Coll Cardiol 1996;28:713-719.

144. Messineo FC. Ventricular Ectopic Activity: Prevalence and Risk. Am J Cardiol 1989;64:53j-56j.

145. Meurs KM, Spier AW, Miller MW, et al. Familial ventricular arrhythmias in Boxers. J Vet Intern Med 1999;13:437-439.

146. Meurs KM, Spier AW, Wright NA, Atkins CE, DeFrancesco TC, Gordon SG, Hamlin RL, Keene BW, Miller MW, Moise NS. Comparison of the effects of four antiarrhythmic treatments for familial ventricular arrhythmias in Boxers. J Am Vet Med Assoc 2002;221.

147. Meurs KM, Spier AW, Wright NA, Hamlin RL. Comparison of in-hospital versus 24-hour ambulatory electrocardiography for detection of ventricular premature complexes in mature Boxers.J Am Vet Med Assoc 2001;218:222-224.

148. Meurs KM, Spier AW, Wright NA, Hamlin RL. Use of ambulatory electrocardiography for detection of ventricular premature complexes in healthy dogs.J Am Vet Med Assoc 2000; 218:1291-1292.

149. Michelson E, Morganroth J. Spontaneous variability of complex bentricular arrhythmias deteted by ong-term electrocardiographic recording. Circulation 1980;61:690-695.

150. Miller R, Lehmkuhl LB, Bonagura JD, Beall MJ. Retrospective analysis of the Clinical utility of Ambulatory Electrocardiographc (Holter) Recordings in Syncopal Dogs: 44 Cases (1991 -1995). J Vet Intern Med 1999; 13:111 -122.

151. Minors SL, O’Grady MR. Heart rate variability in the dog: Is it too variable? Can J Vet Res 1997;61:134-144.

152. Mkparu FO. QT dispersion: proposed technique for optimal risk stratification.Am J Cardiol 1998.

153. Moise NS, Gilmour RF, Riccio M. Diagnosis of inherited ventricular tachycardia in German Shepherd dogs. J Am Vet Med Assoc 1997;210:403-410.

124

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 154. Molnar J, Weiss J, Zhang F, Rosenthal JE. Evaluation of five correction formulas using a software-assisted method of continuous QT measurement from 24-hour Holter recordings. Am J Cardiol 1996;78:920-926.

155. Morganroth J, Michelson E, Horowitz L. Limitations of routine long-term electrocardiographic moitoring to assess ventricular ectopic frequency.Circulation 1978;58:408-414.

156. Morganroth J. Computer recognition of cardiac arrhythmias and statistical approaches to arhythmia analysis.Ann N Y Acad Sci 1984;432:128.

157. Moss AJ, Schnitzler R, Green R, Decamilla J. Ventricular Arrhythmias 3 weeks after acute Myocardial Infarction. Ann Intern Med 1971;75:837-841.

158. Moss AJ. Clinical Management of Patients with the Long QT Syndrome: Drugs, Devices, and Gene-Specific Therapy.Pacing Clin Electrophysiol 1997;20:2058-2060.

159. Nademanee K, Christenson P, Intarachot V. Variability o f indexes for myocardial ischemia: A comparison of exercise treadmill test, ambulatory eclectrocardiographic monitoring and symptoms of myocardial ischemia. J Am Coll Cardiol 1989;13:574-579.

160. Nava A, Folino AF, Turrini P, Buja GF, Daliento L, Thiene G. Signal-averaged electrocardiogram in patients with arrhythmogenic right ventricular cardiomyopathy and ventricular arrhythmias. Eur Heart J 2000;21:58-65.

161. O'Grady, M. R. DCM in Doberman pinschers: Lessons learned in the first decade of study. 20. 2002. Dallas, Tx. Proceedings of the 20th annual ACVIM forum.

162. Oikarinen L, Viitasalo M, Toivonen L. Dispersions of the QT interval in postmyocardial infarction patients presenting with ventricular tachycardia or with ventricular fibrillation. Am J Cardiol 1998;81:691-697.

163. Okin PM, Devereux RB, Howard BV, Fabsitz RR, Lee ET, Welty T. Assessment of QT Interval and QT Dispersion for Prediction of All-Cause and Cardiovascular Mortality in American Indians. Circulation 2000;101:61.

164. Olhausen KV, Schwarz F, Apfelbach J, Rohrig N, Kramer B, Kubler W. Determinants of the Incidence and Severity of Ventricular Arrhythmias in Aortic Valve Disease. AJC 1983;51:1109.

165. Oselladore L, Nava A, Buja G, Turrini P, Daliento L, Livolsi B, Thiene G. Signal- Averaged Electrocardiography in Familial Form of Arrhythmogenic Right Ventricular Cardiomyopathy. AJC 1995;75:1038-1041.

166. Pantano J, Oriel R. Prevalence and nature of ardiac arrhythmias in apparently normal well-trained runners.Am Heart J 1982;104:762-768.

125

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 167. Patterson D, Detweiler D, Hubben K. Spontaneous abnormal cardiac arrhythmias and conduction disturbances in the dog.Am J Vet Res 1961;355-369.

168. Peeters HA, Sippensgroenewegen A, Wever EFD, et al. Electrocardiographic identification of abnormal ventricular depolarization and repolarization in patients with idiopathic ventricular fibrillation. J Am Coll Cardiol 1998;31:1406-1413.

169. Pieper SJ, Hammill SC. Heart rate variability: Technique and investigational applications in cardiovascular medicine.Mayo Clin Proc 1995;70:955-964.

170. Pitzalis M, Mastropaspua F, Massari F. Holter-guided identification of premature ventricular contractions susceptible to suppression by -blockers.Am Heart J 1996;131:508-515.

171. Ponikowski P, Anker SD, Chua TP. Depressed Heart Rate Variability as an Independent Predictor of Death in Chronic Congestive Heart Failure Secondary to Ischemic or Idiopathic Dilated Cardiomyopathy.American Journal o f Cardiology 1997;79:1645-1650.

172. Puddu PE, Jouve R, Mariotti S, et al. Evaluation of 10 QT prediction formulas in 881 middle-aged men from the seven countries study: Emphasis on the cubic root Fridericia's equation. JElectrocardiol 1988;21:219-229.

173. Puljevic D, Smalcelj A, Durakovic Z, Goldner V. QT dispersion, daily variations, QT interval adaptation and late potentials as risk markers for ventricular tachycardia. Eur Heart J 1997;18:1343-1349.

174. Raekallio M. Long term ECG recording with Holter monitoring in clinically healthy horses. Acta Vet Scand 1992;33:71-75.

175. Ravenswaaij C, Kollee L, Hopman J. Heart rate variability. Ann Intern Med 1993;118:436-447.

176. Ryan M, Lown B, Horn H. Comparison of ventricular ectopic activity during 24- hour monitoring and exercise testing in patients with coronary heart disease.N Engl J Med 1975;292:224-229.

177. Sami M, Kraemer H, Harrison D. A new method for evaluating antiarrhythmic drug efficacy. Circulation 1980;62:1172-1179.

178. Samson RA, Lee H-C. Delayed afterdepolarizations and triggered arrhythmias in hypertrophic cardiomyopathic hearts.Journal o f Laboratory and Clinical Medicine 1994;124:242-247.

179. Savelieva I, Yi G, Guo X, et al. Agreement and reproducibility of automatic versus manual measurement of QT interval and QT dispersion.Am J Cardiol 1998;81:471-477. 126

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 180. Scalvini S, Volterrani M, Zanelli E, Pagani M, Mazzuero G, Coats AJ, Giordano A. Is heart rate variability a reliable method to assess autonomic modulation in left ventricular dysfunction and heart failure? Assessment of autonomic modulation with heart rate variability. IntJ Cardiol 1998;67:9-17.

181. Schactman M, Greene JS. Signal-averaged electrocardiography: A new technique for determining which patients may be at risk for sudden cardiac death. Focus on Critical Care 1991;18:202-210.

182. Schwartz PJ, Stramba-Badiale M, Segantini A, Austoni P, Bosi G, Girogetti R, Grancini F, Mami ED, Perticone F, Rosti D, Salice P. Prolongation o f the QT interval and the sudden infant death syndrome. N Engl J Med 1998;338:1709-1714.

183. Schwartz PJ. The autonomic nervous system and sudden death.Eur Heart J 1998;19:F72-F80.

184. Silberstein MD, Singer DE, Mulley AG, Thibault GE, Barnett O. Patients with Syncope Admitted to Medical Intensive Care Units. JAMA 1982;248:1185-1189.

185. Simson MB. Noninvasive identification of patients at high risk for sudden cardiac death: Signal-averaged electrocardiography. Circulation 1992;85(Suppl 1)^1-145-1-151.

186. Singer DH, Martin GJ, Magid N, Weiss JS, Schaad JW, Kehoe R, Zheutlin T, Fintel DJ, Hsieh AM, Lesch M. Low heart rate variability and sudden cardiac death. J Electrocardiol 1988;Suppl:S46-S55.

187. Snyder PS, Cooke KL, Murphy sT, Shaw NG, Lewis DD, Lanz OI. Electrocardiographic findings in dogs with motor vehicle related trauma.J Am Anim Hosp Assoc 2001;37:55-62.

188. Somsen GA, Szabo BM, van Veldhuisen DJ, de Milliano PA, de Groot CA, Lie KI. Imaging and diagnostic testing: Comparision between iodine 123 metaiodobenzylguanidine scintigraphy and heart rate variabliity for the assesssment of cardiac sympathetic activity in mild to maderate heart failure.AHJ 1997;134:456-458.

189. Stein I, Plunkett J, Troy M. Compatison of techniques for examining long-term ECG recordings. Med Instrument 1980;14:69-72.

190. Stein KM, Borer J, Hochreiter C. Prognostic value and physiological correlates of heart rate variability in chronic severe mitral regurgitation. Circulation 1993;88:127- 135.

191. Stein PK, Bosner MS, Kleiger RE, Conger BM. Heart rate variability: A measure of cardiac autonomic tone. Am Heart J 1994;127:1376-1381.

192. Stein PK, Kleiger RE. Insights from the study of heart rate variability. Annu Rev Med 1999;50:249-261. 127

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 193. Steinbach K, NUmberg M. Present and future role of ambulatory Holter monitoring for arrhythmia risk stratification. Pacing Clin Electrophysiol 1997;20 (Pt II):2587-2593.

194. Steinberg JS, Bigger JT, Jr. Importance of the endpoint of noise reduction in analysis of the signal-averaged electrocardiogram. Am J Cardiol 1989;63:556-560.

195. Struthers AD. Two new culprits in cardiovascular disease:QT dispersion and aldosterone. J Hum Hypertens 1995;9:659-661.

196. Stys A, Stys T. Current applications of heart rate variability. Clin Cardiol 1998;21:719-724.

197. Takada H, Mikawa T, Murayama M. Range of ventricular ectopic complexes in healthy subjects studied with repeated ambulatory electrocardiographic recordings.Am J Cardiol 1989;63:184-186.

198. Teerlink JR, Jalaluddin M, Anderson SJ, Kukin ML, Eichhom EJ, Francis G, Packer M, Massie BM. Ambulatory Ventricular Arrhythmias in Patients with Heart Failure do not Specifically Predict an Increased Risk of Sudden Death.Circulation 2000; 101:40.

199. The ESVEM trial: impact on treatment of ventricular tachyarrhythmias. Pacing Clin Electrophysiol 1997;20:468-477.

200. Thiene G, Basso C, Danieli G. Arrhythmogenic right ventricular cardiomyopathy. 7 0 / 1997;7n3:84-91.

201. Tielemen RG, Crijns HJGM, Wiesfeld ACP, et al. Increased dispersion of refractoriness in the absence of QT prolongation in patients with mitral valve prolapse and ventricular arrhythmias. Br Heart J 1995;73:37-40.

202. Tocco DJ, DeLuna FA, Duncan AEW, Vassil TC, Ulm EH. The physiological Disposition and Metabolism of Enalapril Maleate in Laboratory Animals. Drug Metabolism and Disposition 1982; 10:15-19.

203. Toivonen L. Spontaneous variability in the fequency of ventricular premature complexes over prolonged intervals and implications for antiarrhythmic treatment.Am J Cardiol 1987;60:608-612.

204. Towbin JA, Friedman RA. Prolongation of the QT interval and sudden infant death syndrome. The New England Journal o f Medicine 1998;338:1760-1761.

205. Tresch D. Diagnosic and prognostic value of ambulatory electrocardiographic monitoring in older patients.JAGS 1995;43:66-70.

128

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 206. Tsumabuki S, Ito M, Arita M. Day-to-day variation of the frequency of ventricular pemature contractions depends on variation of heart rates.Jpn Circulation J 1988;52:1231-1239.

207. Turrini P, Angelini A, Thiene G, Buja G, Daliento L, Rizzoli G, Nava A. Late potentials and ventricular arrhythmias in arrhythmogenic right ventricular cardiomyopathy. Am J Cardiol 1999;83:1214-1219.

208. Tygesen H, Rundqvist B, Waagstein F, Wennerblom B. Heart rate variability measurement correlates with cardiac norepinephrine spillover in congestive heart failure. Am J Cardiol 2001 ;87:1308-1311.

209. Ulloa H, Adams W, Houston B, Altrogge D. Arrhythmias prevalence during ambulatory eclectrodardiographic monitoring of beagles. Am J Vet Res 1995;56:275- 281.

210. Vincent GM. The Molecular Genetics of the Long QT Syndrome: Genes Causing Fainting and Sudden Death. Annu.Rev.Med. 1998;49:263-274.

211. Vogel AP, Jaax GP, Tezak-reid TM, Bartholomew JL. Ambulatory electrocardiography (Holter monitoring ) in caged monkeys. Laboratory Animals 1991;25:16-20.

212. Vuille C, Weyman AE. Left Ventricle I: General considerations, assessment of chamber size and function. In: Weyman AE, ed. Principles and Practice of Echocardiography. Philadelphia: Lea& Febiger, 1994:575-624.

213. Walter PF, Reid SD, Wenger NK. Transient Cerebral Ischemia Due to Arrhythmia. Ann Intern Med 1970;72:471-474.

214. Ware WA. Practical Use of Holter Monitoring. Comp Conti Educ Pract Vet 1998;20:167-177.

215. Ware WA. Twenty-Four-Hour Ambulatory Electrocardiography in Normal Cats.J Vet Intern Med 1999;13:175-180.

216. Wijbenga JAM, Balk AHHM, Meij SH, Simoons ML, Malik M. Heart rate variability index in congestive heart failure: Relation to clinical variables and prognosis. Eur Heart /1998;19:1719-1724.

217. Winkle R, Derrington D, Schroeder J. Characteristics of ventricular Tachycardia in ambulatory patients. Am J Cardiol 1977;39:487-492.

218. Winkle R, Gradman A, Fitzgerald J. Antiarrhymic drug effect assessed from ventricular arrhymia reduction in the ambulatory electrocardiogram and treadmill test: comparison of propranolol, procainamide and quinidine.Am J Cardiol 1978;42:473- 480. 129

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 219. Winkle R, Jalal S, Kates R. Clinical pharmacology and antiarrhythmic efficacy of N-Acetylprocainamide. Am J Cardiol 1981;47:123-130.

220. Winkle R, Rodriguez I, Bragg-Remschel D. Technological status and problems of ambulatory electrocardiographic monitoring. Ann N Y Acad Sci 1997; 108-116.

221. Winkle R. Ambulatory electocardiography and the diagnosis, evaluation, and treatment of chroic ventricular arrhythmias. Prog Cardiovasc Dis 1980;23:99-128.

222. Winkle R. Ambulatory electrocardiography. Mod Concepts Cardiovasc Dis 1980;1009:7-12.

223. Winkle R. Antiarrhythmic drug effect mimicked by spontaneous variability of ventricular ectopy.Circulation 1978;57:1116-1121.

224. Winkle R. Current status of ambulatory electocardiography.Am Heart J 1981;102:757-770.

225. Winkle RA, Peterson M, Hall R. Characterization o f ventricular tacharrhythmias on ambulatory ECG recordings in post-myocardial infarction patients: Arrhythmia detection and duration of recording, relationship between arrhythmia frequency and complexity, and day-to-day reproducibility. Am Heart J 1981; 102:162-169.

226. Winkle RA. Current Status of Ambulatory Electrocardiography.AHJ 1981;102:757-770.

227. Yi G, Goldman JH, Keeling PJ, Reardon M, McKenna WJ, Malik M. Heart rate variability in dilated cardiomyopathy: Relation to disease severity and prognosis. Heart 1997;77:108-114.

228. Yoshikawa T, Baba A, Akaishi M, Mitamura H, Ogawa S, Suzuki M, Negishi K, Takahashi T, Murayama A. Neurohormaonal activations in congestive heart failure: Correlations with cardiac function, heart rate variability, and baroreceptor sensitivity. Am Heart J 1999;137:666-671.

229. Yoshikawa T, Baba A, Akaishi M, Mitamura H, Ogawa S, Suzuki M, Negishi K, Takahashi T, Murayama A. Neurohormaonal activations in congestive heart failure: Correlations with cardiac function, heart rate variability, and baroreceptor sensitivity. Am Heart J 1999;137:666-671.

230. Zabel M, Klingenheben T, Franz MR, et al. Assessment of QT dispersion for prediction of mortality or arrhythmic events after myocardial infarction. Circulation 1998;97:2543-2550.

231. Zabel M, Portnoy S, Franz MR. Electrocardiographic indexes of dispersion of ventricular repolarization: An isolated heart validation study.J Am Coll Cardiol 1995;25:746-752. 130

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission. 232. Zabel M, Woosley RL, Franz MR. Is dispersion of ventricular repolarization rate dependent? Pacing Clin Electrophysiol 1997;20(Part 1)^2495-2411.

233. Zaputovic L, Mavric Z, Zaninovic-Juijevic T, Matana A, Bradic N. Relationship between QT dispersion and the incidence of early ventricular arrhythmias in patients with acute myocardial infarction. IntJ Cardiol 1997;62:211-216.

234. Zimetbaum PJ, Josephson ME. The evolving role of ambulatory arrhythmia monitoring in general clinical practice. Ann Intern Med 1999;130:848-856.

235. Zipes DP. Genesis of cardiac arrhythmias: Electrophysiological considerations. In: Braunwald E, ed. Heart Disease: A textbook of cardiovascular medicine. Philadelphia: W.B. Saunders Company, 1997:548-592.

236. Zipes DP. Sudden cardiac death: Future approaches.Circulation 1992;85(Suppl I):I-160-I-166

131

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.