Heart, Victor F. Froelicher, Jonathan Mayers, fifth edition

1600 John F. Kennedy Blvd. Ste 1800 Philadelphia, PA 19103-2899 EXERCISE AND THE HEART, Fifth Edition ISBN-13: 978-1-4160-0311-3 Copyright © 2006, Elsevier Inc. ISBN-10: 1-4160-0311-8 All rights reserved. No part of this publication may be reproduced or transmitted in any form or by any means, electronic or mechanical, including photocopying, recording, or any information storage and retrieval system, without permission in writing from the publisher (Elsevier, 1600 John F. Kennedy Boulevard, Suite 1800, Philadelphia, PA, 19103-2899). Notice Knowledge and best practice in this field are constantly changing. As new research and experience broaden our knowledge, changes in practice, treatment and drug therapy may become necessary or appro- priate. Readers are advised to check the most current information provided (i) on procedures featured or (ii) by the manufacturer of each product to be administered, to verify the recommended dose or formula, the method and duration of administration, and contraindications. It is the responsibility of the practi- tioners, relying on their own experience and knowledge of the patient, to make diagnoses, to determine dosages and the best treatment for each individual patient, and to take all appropriate safety precautions. To the fullest extent of the law, neither the publisher nor the authors assume any liability for any injury and/or damage to persons or property arising out of or related to any use of the material contained in this book. The Publisher Library of Congress Cataloging-in-Publication Data Froelicher, Victor F. Exercise and the heart/Victor F. Froelicher, Jonathan Myers.—5th ed. p.;cm Includes bibliographical references and index. ISBN 1-4160-0311-8 1. Exercise tests. 2. Heart function tests. 3. Heart—Diseases—Diagnosis. I. Myers, Jonathan, 1957- II. Title. [DNLM: 1. Heart Diseases—rehabilitation. 2. Exercise Test—methods. 3. Exercise Therapy—methods. 4. Exertion. WG 141.5.F9 F926e 2006] RC683.E.E94F76 2006 616.1′20754—dc22 2005051641 Editor: Susan F. Pioli Senior Editorial Assistant: Joan Ryan Publishing Services Manager: Joan Sinclair Project Manager: Mary Stermel Design Direction: Gene Harris Marketing Manager: Dana Butler Printed in the United States of America Last digit is the print number: 10 9 8 7 6 5 4 3 2 1

To Susan, my wife and best friend. VFF To my two older brothers, Chris and Tim, who are no longer with us. Their intellect eluded me but was always and continues to be a source of motivation and inspiration. JM









preface Welcome to the fifth edition of Exercise and in addition to balloon angioplasty are performed the Heart. Since the fourth edition, there have by interventionalists. been numerous important documents published, including an update of the American Heart AED (automated external defibrillator) and ICD Association (AHA)/American College of Cardiology (implantable cardiac defibrillator) are used for the (ACC) guidelines on exercise testing, the American new biphasic defibrillator products. Thoracic Society/American College of Chest Physicians Statement on Cardiopulmonary Exercise CRT (cardiac resynchronization therapy) is Testing, an AHA Scientific Statement on Exercise an implantable pacemaker for improving cardiac and Heart Failure, an AHA Scientific Statement function that is often combined with an ICD. on Physical Activity in the Prevention of Cardiovascular Disease, new editions of the ACS (acute coronary syndrome) is the term American Association of Cardiovascular and now widely used to describe the spectrum of con- Pulmonary Rehabilitation Guidelines, and the ditions associated with acute myocardial ischemia, American College of Sports Medicine Guidelines including unstable angina pectoris and non-Q on Exercise Testing and Prescription. Relevant wave MIs. information from these updated documents has been incorporated into this fifth edition. The In this edition, we’ve tried to incorporate necessity of practicing evidence-based medicine the influence of the remarkable advances in cardiol- makes it critical that all of us defer to the panels ogy in the subject matter. These advances are listed of experts who write these guidelines. In rare below (not in order of impact), because each by cases in which the guidelines are inconsistent or themselves has strongly influenced exercise testing, we offer an opinion or recommendation that dif- exercise training, and clinical exercise physiology. fers from the guidelines, we alert the reader. 1. Designation of ACSs As the field of cardiology has continued to 2. Biomarkers for ischemia and volume evolve, it is important to note some of the new or changed acronyms in medicine. overload/left ventricular dysfunction at point of contact (troponin and brain natri- HF has been recommended as the acronym to uretic peptide [BNP]) replace CHF, because CHF has confusingly repre- 3. Advances in percutaneous coronary inter- sented either chronic or congestive (acute) heart ventions (PCI) culminating in drug-eluting failure. stents that have greatly reduced stent failure 4. Evidence-based recommendations that PCI (percutaneous coronary intervention) has PCI is better than thrombolysis for acute replaced PTCA, because currently many techniques myocardial infarction 5. Medications that convincingly improve survival in patients with heart disease vii

viii Preface 6. Pacemakers for CRT patient and physician want or need individualized 7. Further advances in exercise training for prognostic information, there is no test more valuable than the standard exercise test. Several patients with heart failure and other groups recent studies have confirmed that exercise capac- previously excluded from rehabilitation ity alone has independent and significant prog- 8. The basic role of endothelial function in nostic power regardless of the patient’s clinical maintaining cardiovascular health and history. how it is affected by exercise training 9. Human genomics studies related to sudden Surprisingly, the next two items, which relate death (LQT1) and training (ACE) to patients with HF, have resulted in new ideas 10. Advances in cardiac defibrillators (implant- regarding cardiovascular physiology. First, HF able and portable units) results in major metabolic and cellular changes that can be improved by an exercise program. These These advances have actually interacted with alterations have provided interesting insights into one another, so it is best to address them in the exercise response, because changes in endothe- groupings that impact health care in a similar lial function appear to be a major contributor to fashion. We address those that impact the diag- these improvements. Second, implanted synchro- nostic use of exercise testing first. Many patients nous pacemakers have been shown to improve who required diagnostic exercise testing after both ventricular function and exercise capacity. the first appearance of symptoms now have the This is somewhat surprising, because previously diagnosis made based on an elevation of troponin. it was thought that myocyte damage was the pri- They frequently go straight to cardiac catheteriza- mary event leading to LV dysfunction and that tion. Many cardiologists believe that advances in conduction disturbances were a result of this. PCI make the noninvasive diagnosis of ischemic However, improvement in function resulting from chest pain moot, because angiography can be correction of dysynchrony suggests that damage used to make the diagnosis and treat the problem to the conduction system can be the cause of LV by averting all steps in between. The lowered dysfunction and impair exercise capacity. restenosis rate associated with drug-eluting stents has removed, in their minds, all the rea- There are two important bench findings sons not to diagnose and fix the problem all in one that are impacting our understanding of cardiac relatively low-risk procedure. However, it is pathophysiology. First, regular exercise can have important to keep in mind that health care costs a powerful effect on endothelial dysfunction, and continue to rise and fewer people are insured or now this mechanism is proposed as one of the able to afford this invasive approach. As clinicians major beneficial actions that exercise has on health. continue to deal with the problems of cost-efficacy, Second, we are only on the threshold of using we contend that the exercise test remains the human genomics to understand exercise and the most logical gatekeeper to more expensive and/or heart. Congenital diseases that cause exercise- invasive diagnostic tests. When a biomarker that related sudden death have been localized to specific can be measured at point of contact becomes val- genes (for instance, LQT1). The ACE gene appears idated as a way to increase the sensitivity of the to be an important determinant of the response to test along with multivariate scores, reasonable exercise training. clinicians apply the exercise test first. Although some would disagree, we contend that the “art” of Finally, advances in defibrillators have had medical decision making and the use of noninva- an important impact on exercise for the public. sive tests are currently more important than ever. Biphasic units resulting in lower energy needs for defibrillation, long-life lithium batteries, and Next, let us consider the advances in health care smart arrhythmia algorithms are the basis for that affect the prognostic use of exercise testing. these advances. Cardiopulmonary resuscitation PCI for acute myocardial infarction has been (CPR) has been improved by AEDs, and they are shown to be better for improving prognosis and now widely used by the public, resulting in better lessening myocardial damage than thrombolysis. outcomes following arrhythmic events. These The reason this is so is that it is more effective than devices are now ubiquitous and are mandatory at thrombolytic drugs in opening coronary arteries gyms and sporting events; importantly, the AHA blocked by thrombosis. Improved patency rates has developed guidelines for their use in health mean that follow-up exercise testing is less likely clubs. However, studies on the number of sudden to be needed routinely after MI to determine who cardiac deaths (50,000/year in the United States; needs coronary angiography. However, when the approximately one fifth of the number estimated initially and used as the impetus for AEDs) and

Preface ix their location (most sudden deaths occur ■ Exercise capacity should be reported in METs, at home and not in public places) have led to not minutes of exercise. some reassessment of their use. Randomized trials have demonstrated a survival benefit for ■ Hyperventilation before testing is not indi- ICDs in most patients with LV dysfunction. Now cated but can be used at another time if a false- patients with these devices must be dealt with in positive test is suspected. the context of cardiac rehabilitation and exercise laboratory settings. ■ ST measurements should be made at ST0 (J-junction), and ST depression should be consid- The following is our strongest variance from ered abnormal only if horizontal or downslop- the guidelines: Exercise testing should be used for ing; most clinically important ST depression screening healthy, asymptomatic individuals occurs in V5, particularly in patients with a along with risk factor assessment. We plan to normal resting ECG. lobby with our colleagues on this point for the following reasons: ■ Patients should be placed supine as soon as possible post exercise without a cool-down ■ A number of contemporary studies have walk in order for the test to have its greatest demonstrated remarkable risk ratios for the diagnostic value. combination of the standard exercise test responses and traditional risk factors. ■ The 2- to 4-minute recovery period is critical to include in analysis of the ST response. ■ Other modalities without the favorable test characteristics of the exercise test are being ■ Measurement of systolic blood pressure during promoted for screening. exercise is extremely important, and exer- tional hypotension is ominous; at this point, ■ Physical inactivity has reached epidemic only manual blood pressure measurement proportions, and the exercise test provides techniques are valid. an ideal way to make patients conscious of their deconditioning and to make physical ■ Age-predicted heart rate targets are largely activity recommendations. useless because of the wide scatter for any age; a relatively low heart rate can be maximal for a ■ Adjusting for age and other risk factors, given patient and submaximal for another. each MET increase in exercise capacity equates to a 10% to 25% improvement in ■ The Duke Treadmill Score should be calculated survival. automatically on every test except for the elderly. With this fifth edition, we once again have ■ Other predictive equations and heart rate recov- assumed the writing by ourselves. Though it is ery should be considered a standard part of the obvious which one of us was the main author for treadmill report. the various chapters, we collaborated on all of them and take both blame and credit. With the To ensure the safety of exercise testing and volume of studies on exercise testing and training reassure the noncardiologist performing the now all available on the world wide web, it is no test, the following list of the most dangerous cir- longer practical to review in detail as many individ- cumstances in the exercise testing lab should be ual studies, however important they are. Although considered: we have been careful to update our citations, we felt it necessary to keep the classic studies related ■ Testing patients with aortic valvular disease to particular issues. Wherever possible, we have or obstructive hypertrophic cardiomyopathy tried to summarize the major studies in tables, (ASH or IHSS) should be done with great followed by a comment and then our overall view care. Aortic stenosis can cause cardiovascu- or recommendation on a given issue. lar collapse, and these patients may be diffi- cult to resuscitate because of the outflow Once again we feel it is important to provide obstruction; IHSS can become unstable due the following precepts in the preface regarding to arrhythmia. Because of these conditions, methodology even though the details are in the a physical exam including assessment of chapters: systolic murmurs should be done before all exercise tests. If a significant murmur ■ The treadmill protocol should be adjusted to is heard, an echocardiogram should be the patient; one protocol is not appropriate for considered before performing the test. all patients. ■ When patients without diagnostic Q-waves on their resting ECG exhibit exercise-induced

x Preface ST segment elevation (i.e., transmural and many of them continue to collaborate with ischemia), the test should be stopped; this us. A few individuals in particular warrant men- can be associated with dangerous arrhyth- tioning here, because their contributions to this mias and infarction. This occurs in about 1 edition are significant. They include Paul Dubach of 1000 clinical tests. from Switzerland, Euan Angus Ashley from Scotland (now a Stanford cardiology fellow), and ■ A cool-down walk is advisable in the follow- Kari Saunamaki from Denmark. Takuya Yamazaki ing instances: was our most recent research fellow from Japan (of a list of many), and his desk is currently occu- 1. When a patient with an ischemic cardio- pied by Tan Swee Yaw from Singapore. Notable myopathy exhibits significant chest pain PhDs who keep a close eye on our science are due to ischemia, because the ischemia Barry Franklin, Paul Ribisl, and Bill Herbert. We can worsen in recovery have profited both personally and professionally by our association with all of these individuals 2. When a patient develops exertional hypo- and treasure the friendships that began through tension accompanied by ischemia (angina research collaboration. or ST depression) or when it occurs in a patient with a history of HF, cardio- Given this background, we are targeting this myopathy, or recent MI book as a reference for the clinical aspects of exer- cise testing and training. It is meant for the serious 3. When a patient with a history of sudden student, academic, or health care provider who death or collapse during exercise develops wants to have available much of the knowledge in PVCs that become frequent this field summarized in one source. Hopefully it will find an appropriate niche on the shelves in Appreciation of these circumstances can help many exercise labs, cardiac rehabilitation depart- avoid any complications in the exercise lab. ments, and educational training programs. We have tried to incorporate the latest available guide- As in previous editions, there are many pre- lines, position statements, and meta-analyses. Our medical and medical students, graduate students, love of the subject has led to the incorporation of residents, fellows, visiting professors, and inter- details that some could consider minutia yet we national medical graduates who have contributed might have missed some work considered impor- to the studies discussed in this book. They are tant by our colleagues. We hope you enjoy this too numerous to mention individually, but their book and that it is helpful to you. work is cited extensively in this edition. One of the most gratifying things about what we do is to Victor F. Froelicher have the opportunity to host these individuals Jonathan Myers and gain the friendships that result through the inevitable battles that occur in trying to answer a research question. Because of this, we have main- tained a wide range of contacts around the world,

CHAPTER one Basic Exercise Physiology Exercise physiology is the study of the physiologic the amount of oxygen consumed by the heart mus- responses and adaptations that occur as a result cle. Accurate measurement of myocardial oxygen of acute or chronic exercise. Exercise is the body’s consumption requires the placement of catheters most common physiologic stress, and it places in a coronary artery and in the coronary venous major demands on the cardiopulmonary system. sinus to measure oxygen content. The determi- For this reason, exercise can be considered the nants of myocardial oxygen uptake include most practical test of cardiac perfusion and func- intramyocardial wall tension (left ventricular tion. Exercise testing is a noninvasive tool to pressure × end-diastolic volume), contractility, evaluate the cardiovascular system’s response to and heart rate. It has been shown that myocardial exercise under carefully controlled conditions. oxygen uptake can be reasonably estimated by the The adaptations that occur during an exercise test product of heart rate and systolic blood pressure allow the body to increase its resting metabolic (double product). This information is valuable rate up to 20 times, during which time cardiac clinically because exercise-induced angina often output may increase as much as six times. The occurs at the same myocardial oxygen demand magnitude of these adjustments is dependent (double product) and thus is a useful physiologic upon age, gender, body size, type of exercise, fit- variable when evaluating therapy. When it is not ness, and the presence or absence of heart disease. the case, the influence of other factors should Although major adaptations are also required of be suspected, such as a recent meal, abnormal the endocrine, neuromotor, and thermoregula- ambient temperature, or coronary artery spasm. tory systems, the major focus of this chapter is on the cardiovascular response and adaptations The second principle of exercise physiology is of the heart to acute exercise. Cardiovascular one of pathophysiology: considerable interaction adaptations to chronic training in humans and takes place between the exercise test manifestations animals are reviewed in Chapter 12. of abnormalities in myocardial perfusion and function. The electrocardiographic response to exer- It is important to understand two basic cise and angina are closely related to myocardial principles of exercise physiology with regard to ischemia (coronary artery disease), whereas exer- exercise testing. The first is a physiologic princi- cise capacity, systolic blood pressure, and heart ple: total body oxygen uptake and myocardial rate responses to exercise can be determined by oxygen uptake are distinct in their determinants the presence of myocardial ischemia, myocardial and in the way they are measured or estimated dysfunction, or responses in the periphery. (Table 1-1). Total body or ventilatory oxygen Exercise-induced ischemia can cause cardiac dys- uptake (VO2) is the amount of oxygen that is function that results in exercise impairment and extracted from inspired air as the body performs an abnormal systolic blood pressure response. work. Conversely, myocardial oxygen uptake is Often it is difficult to separate the impact of 1

2 EXERCISE AND THE HEART TA B L E 1 - 1 . Two basic principles of exercise physiology Myocardial oxygen consumption ;Heart rate × systolic blood pressure Ventilatory oxygen consumption (VO2) (determinants include wall tension ≅ left ventricular pressure × volume; contractility; and heart rate) ;External work performed, or cardiac output × a-VO2 difference* *The arteriovenous O2 difference is approximately 15 to 17 vol% at maximal exercise in most individuals; therefore, VO2 max generally reflects the extent to which cardiac output increases. ischemia from the impact of left ventricular meter (Nm). One Nm is equal to one joule (J), dysfunction on exercise responses. An interaction which is another common expression of work. exists that complicates the interpretation of Because work is nearly always expressed per unit the exercise test findings. The variables affected of time (i.e., as a rate), an additional unit that by both myocardial ischemia and ventricular becomes important is power, the rate at which dysfunction (i.e., exercise capacity, maximal work is performed. The body’s metabolic equiva- heart rate, and systolic blood pressure) have the lent (MET) of power is energy. Therefore, it is greatest prognostic value. easy to think of work as anything with weight moving at some rate across time (which is often The severity of ischemia or the amount of analogous to distance). The common biologic myocardium in jeopardy is known clinically to measure of total body work is the oxygen uptake, be inversely related to the heart rate, blood which is usually expressed as a rate (making it pressure, and exercise level achieved. However, a measure of power) in liters per minute. MET is neither resting nor exercise ejection fraction nor a term commonly used clinically to express the a change in ejection fraction during exercise oxygen requirement of the work rate during an correlates well with measured or estimated maxi- exercise test on a treadmill or cycle ergometer. mal oxygen uptake, even in patients without signs One MET is equated with the resting metabolic or symptoms of ischemia.1,2 Moreover, exercise- rate (;3.5 mL of O2/kg/min), and a MET value induced markers of ischemia do not correlate well achieved from an exercise test is a multiple of with one another. Silent ischemia (i.e., markers of the resting metabolic rate, either measured ischemia presenting without angina) does not directly (as oxygen uptake) or estimated from the appear to affect exercise capacity in patients with maximal workload achieved using standardized coronary heart disease. Although not conclusive, equations.5 radionuclide studies support this position.3 Cardiac output is generally considered the most Energy and Muscular Contraction. Muscular important determinant of exercise capacity, but contraction is a complex mechanism involving studies suggest that in some patients with heart the interaction of the contractile proteins actin disease, the periphery plays an important role in and myosin in the presence of calcium. The limiting exercise capacity.1,4 British scientist A.F. Huxley proposed that the myosin and actin filaments in the muscle slid past Concepts of Work. Because exercise testing fun- one another as the muscle fibers shortened dur- damentally involves the measurement of work, ing contraction. Huxley won the Nobel Prize for there are several concepts regarding work that are this concept, which is still generally considered important to understand. Work is defined as force correct. The source of energy for this contraction moving through a given distance (W = F × D). If is supplied by adenosine triphosphate (ATP), muscle contraction results in mechanical move- which is produced in the mitochondria. ATP is ment, then work has been accomplished. Force is stored as two products, adenosine diphosphate equal to mass times acceleration (F = M × A). and phosphate, at specific binding sites on the Any weight, for example, is a force that is under- myosin heads. going the resistance provided by gravity. A great deal of any work that is performed involves over- The sequence of events that occurs when a coming the resistance provided by gravity. muscle contracts has three other major players: calcium and two inhibitory proteins, troponin The basic unit of force is the newton (N). It is and tropomyosin. Voluntary muscle contraction the force that, when applied to a 1-kg mass, gives begins with the arrival of electrical impulses it an acceleration of 1 m multiplied by sec−2. Since at the myoneural junction, initiating the release work is equal to force (in newtons) times distance of calcium ions. Calcium is released into the (in meters), another unit for work is the newton

C H A P T E R 1 Basic Exercise Physiology 3 sarcoplasmic reticulum that surrounds the oxygen supply. Under such conditions, glycolysis muscle filaments. The calcium binds to a special progresses in the cytoplasm much the same way protein, troponin-C, which is attached to as aerobic metabolism until pyruvate is formed. tropomyosin (another protein that inhibits the However, electrons released during glycolysis binding of actin and myosin), and actin. When cal- are taken up by pyruvate to form lactic acid. cium binds to troponin-C, the tropomyosin mole- Rapid diffusion of lactate from the cell inhibits cule is removed from its blocking position between any further steps in glycolysis. Thus, oxygen- actin and myosin. The myosin head then attaches independent glycolysis is inefficient; two ATP to actin, and muscular contraction occurs. molecules per glucose molecule is the total yield from this process. The main source of energy for muscular con- traction, ATP, is produced by oxidative phosphory- The fact that lactate accumulates in the blood lation. The major fuels for this process are during rapid glycolysis is an important concept in carbohydrates (glycogen and glucose) and free exercise science. The relative exercise intensity in fatty acids. At rest, roughly equal amounts of which lactate accumulation occurs is an impor- energy are derived from carbohydrates and fats. tant determinant of endurance performance. The Free fatty acids contribute greatly to the energy degree to which lactate accumulates in the blood supply during low levels of exercise, but greater is related to exercise intensity and the extent to amounts of energy are derived from carbohy- which fast-twitch (type IIB) fibers are recruited. drates as exercise progresses. Maximal work This subject is discussed further in Chapter 3. relies virtually entirely on carbohydrates. Because endurance performance is directly related to the Although lactate can contribute to fatigue by rate at which carbohydrate stores are depleted, increasing ventilation and inhibiting other major advantages exist for both: (1) having enzymes of glycolysis, it can also serve as an impor- greater glycogen stores in the muscle and tant energy source in muscles other than those in (2) deriving a relatively greater proportion of which it was formed, and it serves as an important energy from fat during prolonged exercise. Both precursor for liver glycogen during exercise.6-8 of these benefits are conferred with training. Muscle Fiber Types. The body’s muscle fiber Oxidative phosphorylation initially involves a types are classified on the basis of the speed with series of events that take place in the cytoplasm. which they contract, their color, and their mito- Glycogen and glucose are metabolized to pyruvate chondrial content. Type I, or slow-twitch fibers, through glycolysis. If oxygen is available, pyruvate are red in color and contain high concentrations enters the mitochondria from the sarcoplasm and of mitochondria. Type II, or fast-twitch fibers, are is oxidized to a compound known as acetyl CoA, white in color and have low concentrations of which then enters a cyclical series of reactions mitochondria. Fiber color is related to the known as the Krebs cycle. By-products of the degree of myoglobin, which is a protein that Krebs cycle are CO2 and hydrogen. Electrons from both stores oxygen in the muscle and carries hydrogen enter the electron transport chain, oxygen in the blood to the mitochondria. Not yielding energy for the binding of phosphate surprisingly, slow-twitch fibers with their high (phosphorylation) from adenosine diphosphate to myoglobin content are more resistant to fatigue; ATP. This process, oxidative phosphorylation, is the thus, a muscle with a high percentage of slow- greatest source of ATP for muscle contraction. A twitch fibers is well suited for endurance exercise. total of 36 ATP molecules per glucose molecule are However, slow-twitch fibers tend to be smaller formed in the mitochondria during this process. and produce less overall force than fast-twitch fibers. Fast-twitch fibers are generally larger and The mitochondria can produce ATP for muscle tend to produce more force, although they fatigue contraction only if oxygen is present. However, more easily. Research suggests that the speed at higher levels of exercise, total body oxygen of contraction for each fiber type is based largely demand may exceed the capacity of the cardio- on the activity of the enzyme myosin ATPase, vascular system to deliver oxygen. Historically, which sits in the myosin head and to which ATP “anaerobic” (without oxygen) glycolysis has been combines. the term used to describe the synthesis of ATP from glucose under these conditions. Many It is important to note that although the researchers have superseded this term with two fiber types can be separated by distinct char- more functional descriptions, such as “oxygen acteristics, both fibers function effectively for independent,” “nonoxidative,” or “rapid” glycoly- virtually all physical activities. Evidence also sis, because “anaerobic” incorrectly implies that suggests that slow-twitch and fast-twitch fibers glycolysis occurs only when there is an inadequate are not as dichotomous as previously thought.

4 EXERCISE AND THE HEART Myosin ATPase activity and speed of contraction and out of the lung and by the fraction of this of some slow-twitch fibers approximate those of ventilation that is extracted by the tissues: fast-twitch fibers. Moreover, type II (fast-twitch) fibers have been further divided into three sub- VO2 = VE × (FiO2 − FeO2) categories: type IIA, type IIB, and type IIC. The type IIA fiber mimics the type I fiber in that it where VE is minute ventilation, and FiO2 and has a high capacity for oxidative metabolism. FeO2 are the fractional amounts of oxygen in It has been suggested that the type IIA fiber the inspired and expired air, respectively. (For actually is a type II fiber that has been adapted for the moment, this equation is oversimplified, endurance exercise, and endurance athletes are as the measurement of VO2 also requires a deter- known to have a relatively large number of mination of expired CO2, as detailed in Chapter 3.) these fibers.9 The type IIB fiber is a “true” type II fiber in that it contains few mitochondria and Therefore, the cardiopulmonary limits (VO2 is better adapted for short bursts of activity. max) are defined by (1) a central component The type IIC fiber is poorly understood; it may (cardiac output) that describes the capacity of the represent an “uncommitted” fiber, capable of heart to function as a pump and (2) peripheral adapting into one of the other fiber types. factors (arteriovenous oxygen difference) that Historically, it has been thought that endurance describe the capacity of the lung to oxygenate athletes were obliged to be genetically endowed the blood delivered to it and the capacity of the with larger percentages of type I fibers, and that working muscle to extract this oxygen from the the opposite was true of sprinters or jumpers. blood. Figures 1-1 and 1-2 outline the many Numerous cross-sectional studies have con- factors affecting cardiac output and arteriovenous firmed these differences in fiber types between oxygen difference. An abnormality in one or endurance and sprint-type athletes since the more of these components often characterizes the advent of the muscle biopsy technique. However, presence and extent of some form of cardiovascu- fiber types may in fact represent a continuum, lar or pulmonary disease. In the following, these with some capable of adapting toward the models are reviewed in the context of the cardio- characteristics of another fiber. vascular response to exercise. ACUTE CARDIOPULMONARY Central Factors RESPONSE TO EXERCISE Figure 1-1 shows the central determinants of The cardiovascular system responds to acute maximal oxygen uptake. exercise with a series of adjustments that assure (1) active muscles receive blood supply appro- Heart Rate priate to their metabolic needs, (2) heat generated by the muscles is dissipated, and (3) blood supply Sympathetic and parasympathetic nervous system to the brain and heart is maintained. This response influences underlie the cardiovascular system’s requires a major redistribution of cardiac output first response to exercise, an increase in heart along with a number of local metabolic changes. rate. Sympathetic outflow to the heart and sys- temic blood vessels increases and vagal outflow The usual measure of the capacity of the body decreases. Of the two major components of cardiac to deliver and utilize oxygen is the maximal output, heart rate and stroke volume, heart rate is oxygen uptake (VO2 max). Thus, the limits of the responsible for most of the increase in cardiac cardiopulmonary system are historically defined output during exercise, particularly at higher by VO2 max, which can be expressed by the Fick levels. Heart rate increases linearly with workload principle: and oxygen uptake. Increases in heart rate occur primarily at the expense of diastolic, not systolic VO2 max = maximal cardiac output × maximal time. Thus, at very high heart rates, diastolic arteriovenous oxygen difference time may be so short as to preclude adequate ventricular filling. Cardiac output must closely match ventilation in the lung in order to deliver oxygen to the The heart rate response to exercise is working muscle. VO2 max is determined by the influenced by several factors, including age, maximal amount of ventilation (VE) moving into type of activity, body position, fitness, the pres- ence of heart disease, medications, blood volume,

C H A P T E R 1 Basic Exercise Physiology 5 ■ FIGURE 1-1 Central determinants of maximal oxygen uptake. (From Myers J, Froelicher VF: Hemodynamic determinants of exercise capacity in chronic heart failure. Ann Intern Med 1991;115:377-386.) and environment. Of these, the most important blood pressure will resist ventricular outflow factor is age; a decline in maximal heart rate (afterload) and result in a reduced stroke volume. occurs with increasing age.10 This decline appears During exercise, stroke volume increases up to to be due to intrinsic cardiac changes rather than approximately 50% to 60% of maximal capacity, to neural influences. It should be noted that there after which increases in cardiac output are due to is a great deal of variability around the regression further increases in heart rate. The extent to which line between maximal heart rate and age; thus, increases in stroke volume during exercise reflect age-related maximal heart rate is a relatively poor an increase in end-diastolic volume or a decrease index of maximal effort (see Chapter 5). Maximal in end-systolic volume, or both, is not entirely heart rate is unchanged or may be slightly clear but appears to depend upon ventricular func- reduced after a program of training. Resting heart tion, body position, and intensity of exercise. In rate is frequently reduced after training as a result healthy subjects, stroke volume increases at rest of enhanced parasympathetic tone. and during exercise after a period of exercise train- ing. Although the mechanisms have been debated, Stroke Volume. The product of stroke volume (the evidence suggests that this adaptation is due more volume of blood ejected per heartbeat) and heart to increases in preload—and possibly local adapta- rate determines cardiac output. Stroke volume is tions that reduce peripheral vascular resistance— equal to the difference between end-diastolic than to increases in myocardial contractility. and end-systolic volume. Thus, a greater diastolic filling (preload) will normally increase stroke In addition to heart rate, end-diastolic volume volume. Alternatively, factors that increase arterial is determined by two other factors: filling pressure and ventricular compliance. ■ FIGURE 1-2 Poxeyrgipehneurapltadkeete.rTmhienaa-nV–tOs 2of maximal difference is the difference between arterial and venous oxygen. Hb, hemoglobin; PAO2, partial pressure of alveolar oxygen; VE, minute ventilation. (From Myers J, Froelicher VF: Hemodynamic determinants of exercise capacity in chronic heart failure. Ann Intern Med 1991;115:377-386.)

6 EXERCISE AND THE HEART Filling Pressure. The most important determi- ejected with each beat increases, owing to an nant of ventricular filling is venous pressure. The altered cross-bridge formation. Contractility is degree of venous pressure is a direct consequence commonly quantified by the ejection fraction, the of the amount of venous return. The Frank- percentage of blood ejected from the ventricle Starling mechanism dictates that, within limits, during systole using radionuclide, echocardio- all the blood returned to the heart will be ejected graphic, or angiographic techniques. Despite its during systole. As the tissues demand greater oxy- wide application as an index of myocardial con- gen during exercise, venous return increases, tractility, ejection fraction has been repeatedly which in turn increases end-diastolic fiber length shown to correlate poorly with exercise capacity. (preload), resulting in a more forceful contrac- tion. Venous pressure increases as exercise inten- Afterload. Afterload is a measure of the force sity increases. Over the course of a few beats, resisting the ejection of blood by the heart. cardiac output will equal venous return. Increased afterload (or aortic pressure, as is observed with chronic hypertension) results in a A number of other factors affect venous reduced ejection fraction and increased end- pressure, and therefore filling pressure, during diastolic and end-systolic volumes. During exercise. These factors include blood volume, dynamic exercise, the force resisting ejection in body position, and the pumping action of the the periphery (total peripheral resistance) is respiratory and skeletal muscles. A greater blood reduced by vasodilation, owing to the effect of local volume increases venous pressure and therefore metabolites on the skeletal muscle vasculature. end-diastolic volume by making more blood avail- Thus, despite even a fivefold increase in cardiac able to the heart. Because the effects of gravity are output among normal subjects during exercise, negated, filling pressure is greatest in the supine mean arterial pressure increases only moderately. position. In fact, stroke volume generally does not increase from rest to maximal exercise in the Volume Response to Exercise. Results of studies supine position. The intermittent mechanical evaluating the volume response to exercise have constriction and relaxation in the skeletal mus- varied greatly. Although the advent of radionu- cles during exercise also enhance venous return. clide techniques in the 1970s offered promise Finally, changes in intrathoracic pressure that for the noninvasive assessment of ventricular occur with breathing during exercise facilitate the volumes during exercise, the results have been return of blood to the heart. disappointing. Because of technical limitations, most of these studies have been performed in the Ventricular Compliance. Compliance is a mea- supine position. Early studies employing radio- sure of the capacity of the ventricle to stretch in nuclide or echocardiographic techniques during response to a given volume of blood. Specifically, supine exercise among normal subjects reported compliance is defined as the ratio of the change in that end-diastolic volume remained constant volume to the change in pressure. The diastolic or diminished slightly,11-14 increased in the pressure/volume relation is curvilinear; that is, at order of 27%,15 or varied greatly depending on low end-diastolic pressures, large changes in vol- the subject.16-18 Among patients with coronary ume are accompanied by small changes in pressure, artery disease exercised in the supine position, and vice versa. At the upper limits of end-diastolic increases in end-diastolic volume were observed pressure, ventricular compliance declines; that is, among patients with exercise-induced angina, the chamber stiffness increases as it fills. Because of whereas end-diastolic volume did not change in the difficulty in measuring end-diastolic pressure patients who were asymptomatic. Sharma et al19 during exercise, few data are available concerning and Jones et al20 reported increases in both ventricular compliance during exercise in humans. end-diastolic and end-systolic volumes in patients who developed angina during exercise. End-systolic volume is a function of two factors: Slutsky et al11 reported that end-diastolic volume contractility and afterload. remained unchanged in patients with coronary artery disease whether or not they developed Contractility. Contractility describes the forceful- angina. Manyeri and Kostuk21 reported large ness of the heart’s contraction. Increasing con- increases in both end-systolic and end-diastolic tractility reduces end-systolic volume, which volumes during supine exercise among 20 patients results in a greater stroke volume and thus greater with coronary artery disease, 13 of whom developed cardiac output. This process is precisely what angina during exercise. occurs with exercise in the normal individual; the percentage of blood in the ventricle that is

C H A P T E R 1 Basic Exercise Physiology 7 The ventricular volume response to upright increases in both end-systolic and end-diastolic exercise also varies greatly, even in similar popu- volumes from rest to peak exercise ranging lations. The results of some of the major studies between 10% and 20% in patients with left in this area are listed in Table 1-2. Among normal ventricular dysfunction. subjects, end-diastolic volume has been reported to increase greatly,15,21,22 increase moderately,23-27 The inconsistent results concerning the ven- or decrease slightly during upright exercise.28-31 tricular volume response to both supine and End-diastolic volume has been reported to upright exercise have led investigators to raise increase in the range of 8% to 56% among questions concerning the validity of radionuclide patients with coronary artery disease, and end- techniques for assessing ventricular function. systolic volume has been shown to increase in For example, Jensen et al41 studied the individual the range of 16% to 94% in response to upright variability of radionuclide ventriculography in exercise.21,23,32-37 Among normal subjects, end- patients with coronary artery disease with repeat systolic volume has generally been reported to testing for more than 1 year. Although differences decrease in response to maximal upright exercise in end-diastolic volume measurements between (range 4% to 79%).21,23-33,38 Higginbotham et al,22 initial and repeat testing were small, the standard however, observed a 48% increase in end-systolic deviations of the individual differences between volume among normal subjects; others have tests at rest and peak exercise were large, on the reported lesser increases. Less is known about order of 38 and 49 mL, respectively. Variability in the ventricular response to upright exercise in the ejection fraction and end-systolic volume patients with chronic heart failure. Sullivan et al,39 responses to exercise were of a similar magnitude. Tomai et al,31 and Delahaye et al40 all observed In light of the apparent shortcomings of the radionuclide techniques, investigators have TA B L E 1 - 2 . Ventricular volume response to upright exercise using radionuclide of echocardiographic techniques Investigator Population Technique Percent Percent Rerych et al 197823 Normals (n = 30) RN change EDV change ESV CAD (n = 20) RN Freeman et al 198134 Normals (n = 10) RN Increase 10 Decrease 35 CAD (n = 22) RN Increase 56 Increase 94 Wyns et al 198228 Normals (n = 10) RN Increase 25 Increase 10 Manyeri and Kostuk 198321 Normals (n = 22) RN Increase 30 Increase 38 Crawford et al 198333 CAD (n = 10) Echo Decrease 8 Decrease 65 CAD (n = 20) RN Increase 31 Decrease 22 Kalischer et al 198436 CAD (n = 18) RN Increase 8 Increase 22 CAD (n = 10) RN Increase 45 Increase 48 Hakki and Iskandrian 198543 Mixed (n = 117) RN Increase 27 Increase 48 Shen et al 198535 Normals (n = 17) RN Increase 24 Increase 38 CAD (n = 14) RN Increase 15 — Higginbotham et al 198622 Normals (n = 24) RN Increase 22 Increase 27 Iskandrian and Hakki 198624 Normals (n = 41) RN Increase 26 Increase 29 Plotnick et al 198627 Normals (n = 30) RN Increase 45 Increase 48 Renlund et al 198729 Normals (n = 13) RN Increase 6 Decrease 35 Sullivan et al 198839 CHF (n = 20) RN Increase 4 Decrease 50 Ginzton et al 198938 Normals (n = 14) Echo Decrease 3 Decrease 79 Younis et al 199025 Normals (n = 9) RN Increase 20 Increase 20 Goodman et al 199126 Normals (n = 15) RN Decrease 26 Decrease 48 Myers et al 199137 CAD (n = 8) Echo Increase 17 Decrease 4 Schairer et al 199230 Normals (n = 15) Echo Increase 19 Decrease 14 Tomai et al 199232 Normals (n = 12) RN Increase 16 Increase 16 Tomai et al 199331 Normals (n = 10) RN Decrease 4 Decrease 52 CHF (n = 10) RN Decrease 8 Decrease 42 Delahaye et al 199740 CHF (n = 13) RN Decrease 8 Decrease 43 Lapa-Bula et al 200241 CHF (n = 10) Echo Increase 12 Increase 14 Increase 15 Increase 23 Increase 4 Decrease 5 CAD, coronary artery disease; CHF, chronic heart failure; Echo, echocardiography; EDV, end-diastolic volume or end-diastolic volume index; ESV, end-systolic volume or end-systolic volume index; RN, radionuclide ventriculography.

8 EXERCISE AND THE HEART employed alternative methods for quantifying ven- to maximal exercise during upright exercise tricular function during exercise. Crawford et al33 among patients with chronic heart failure, evaluated the feasibility and reproducibility of whereas Lapu-Bula et al42 reported that volumes two-dimensional echocardiography for assessing changed minimally during exercise. Few other left ventricular function during exercise. A 9% data are available for this group in the upright test-retest difference in end-diastolic volume position. was demonstrated. End-diastolic volume was reported unchanged from rest to peak exercise in Peripheral Factors (a-VO2 patients with coronary disease, but it increased Difference) significantly (20%) from rest to peak exercise in normal subjects. Ginzton et al38 compared Figure 1-2 shows the peripheral determinants of athletes with sedentary subjects during upright maximal oxygen uptake. Oxygen extraction by the exercise using two-dimensional echocardio- tissues during exercise reflects the difference graphy. After a slight increase in end-diastolic between the oxygen content of the arteries (gen- volume submaximally in both groups, end- erally 18 to 20 mL O2/100 mL at rest) and oxygen diastolic volume decreased 39% and 35% at peak content in the veins (generally 13 to 15 mL exercise among athletes and sedentary subjects, O2/100 mL at rest, yielding a typical a-VO2 differ- respectively. Although both groups decreased ence at rest of 4 to 5 mL O2/100 mL, ;23% end-systolic volume progressively during exercise, extraction). During exercise, this difference the reduction was greater among the athletes widens as the working tissues extract greater (70% versus 52%). amounts of oxygen; venous oxygen content reaches very low levels and a-VO2 difference may Thus, the ventricular volume response to be as high as 16 to 18 mL O2/100 mL with exhaus- exercise is not entirely clear, but it appears to tive exercise (exceeding 85% extraction of oxygen depend upon the type of disease, method of mea- from the blood at VO2 max). Some oxygenated surement (radionuclide or echocardiographic), blood always returns to the heart, however, as type of exercise (supine versus upright), and exer- smaller amounts of blood continue to flow cise intensity (submaximal versus maximal). through metabolically less active tissues that do Much of the disagreement on this issue can no not fully extract oxygen. Generally, a-VO2 differ- doubt be attributed to differences in the exercise ence does not explain differences in VO2 max level at which measurements were taken. With between subjects who are relatively homogenous. this in mind, some rough generalizations may be That is, a-VO2 difference is generally considered to made concerning changes in ventricular volume widen by a relatively “fixed” amount during in response to upright exercise. exercise, and differences in VO2 max have been historically explained by differences in cardiac In normal subjects, the response from upright output. However, some patients with cardiovascu- rest to a moderate level of exercise is an increase lar or pulmonary disease exhibit reduced VO2 max in both end-diastolic and end-systolic volumes values that can be attributed to a combination of of about 15% and 30%, respectively. As exercise central and peripheral factors. progresses to a higher intensity, end-diastolic volume probably does not increase further,27 but Determinants of Arterial Oxygen Content. Arterial end-systolic volume decreases progressively. At oxygen content is related to the partial pressure peak exercise, end-diastolic volume may even of arterial oxygen, which is determined in the decline somewhat, while stroke volume is main- lung by alveolar ventilation and pulmonary diffu- tained by a progressively decreasing end-systolic sion capacity, and in the blood by hemoglobin volume. Based on six studies that have quantified content. In the absence of pulmonary disease, the volume response of patients with coronary arterial oxygen content and saturation are usu- artery disease in the upright position,21,23,34-37 ally normal throughout exercise, even at very end-diastolic volume has been reported to high levels. This is true even for patients with increase 16% to 56% during exercise. The increase severe coronary disease or chronic heart failure. in end-systolic volume has been reported to However, often patients with symptomatic pul- range from 16% to 48%. An exception, however, monary disease neither ventilate the alveoli is a study performed by Rerych et al23 that adequately nor diffuse oxygen from the lung reported a 94% increase in end-systolic volume. into the bloodstream normally, and a decrease in Sullivan et al,39 Tomai et al,31 and Delahaye et al40 reported approximately 20% increases in both end-systolic and end-diastolic volumes from rest

C H A P T E R 1 Basic Exercise Physiology 9 arterial oxygen saturation during exercise is one physical performance in athletes and for studying of the hallmarks of this disorder. Arterial hemo- the normal and abnormal physiology of other globin content is also usually normal throughout organ systems. exercise. Naturally, a condition such as anemia would reduce the oxygen-carrying capacity of A major increase and redistribution of cardiac the blood, along with any condition that would output underlies a series of adjustments that shift the O2 dissociation curve leftward, such as allow the body to increase its resting metabolic reduced 2, 3-diphosphoglycerate, PCO2, or elevated rate as much as 10 to 20 times with exercise. The temperature. capacity of the body to deliver and utilize oxygen is expressed as the maximal oxygen uptake. Determinants of Venous Oxygen Content. Maximal oxygen uptake is defined as the product Venous oxygen content reflects the capacity to of maximal cardiac output and maximal arterio- extract oxygen from the blood as it flows through venous oxygen difference. Thus, the cardio- the muscle. It is determined by the amount of pulmonary limits are defined by (1) a central blood directed to the muscle (regional flow) and component (cardiac output) that describes the capillary density. Muscle blood flow increases in capacity of the heart to function as a pump and proportion to the increase in work rate and thus (2) peripheral factors (arteriovenous oxygen the oxygen requirement. The increase in blood difference) that describe the capacity of the lung flow is brought about not only by the increase to oxygenate the blood delivered to it and the in cardiac output, but also by a preferential capacity of the working muscle to extract this redistribution of the cardiac output to the exercis- oxygen from the blood. Hemodynamic responses ing muscle. A reduction in local vascular resist- to exercise are greatly affected by the type of exer- ance facilitates the greater skeletal muscle flow. cise being performed, by whether or not disease In turn, locally produced vasodilatory mecha- is present, and by the age, gender, and fitness of nisms, along with neurogenic dilatation resulting the individual. from higher sympathetic activity, mediate the greater skeletal muscle blood flow. A marked Coronary artery disease is characterized by increase in the number of open capillaries reduces reduced myocardial oxygen supply, which, in the diffusion distances, increases capillary blood vol- presence of an increased myocardial oxygen ume, and increases mean transit time, facilitating demand, can lead to myocardial ischemia and oxygen delivery to the muscle. reduced cardiac performance. Despite years of study, a number of dilemmas remain with regard Cross-sectionally, fit individuals have a greater to the response to exercise clinically. Although skeletal muscle capillary density than sedentary myocardial perfusion and function are intuitively subjects. In addition, fit subjects may have a linked, it is often difficult to separate the impact greater capacity to redistribute blood flow toward of ischemia from that of left ventricular dysfunc- the working muscle and away from nonexercising tion on exercise responses. Indices of ventricular tissue. The converse is true in many patients function and exercise capacity are poorly related. with cardiovascular disease. For example, one Cardiac output is considered the most important of the characteristics of the patient with chronic determinant of exercise capacity in normal sub- heart failure is an “exaggeration” of the decondi- jects and in most patients with cardiovascular or tioning response. These patients exhibit a reduced pulmonary disease. However, among patients capacity to redistribute blood, a reduced capacity with disease, abnormalities in one or several of the to vasodilate in response to exercise or following links in the chain that defines oxygen uptake con- ischemia, and a reduced capillary-to-fiber ratio. tribute to the determination of exercise capacity. SUMMARY The transport of oxygen from the air to the mitochondria of the working muscle cell requires The major cardiopulmonary adaptations that are the coupling of blood flow and ventilation to required of acute exercise make exercise testing a cellular metabolism. Energy for muscular con- very practical test of cardiac perfusion and func- traction is provided by three sources: stored tion. The rather remarkable physiologic adapta- phosphates (ATP and creatine phosphate), oxygen- tions that occur with exercise have made exercise independent glycolysis, and oxidative metabolism. a valuable research medium not just for the study Oxidative metabolism provides the greatest of cardiovascular disease, but also for studying source of ATP for muscular contraction. Muscular contraction is accomplished by three fiber types that differ in their contraction speed, color, and mitochondrial content. The duration and intensity

10 E X E R C I S E A N D T H E H E A R T of activity determine the extent to which these heart disease. Evaluation by radionuclide angiography. Ann Surg fuel sources and fiber types are called upon. 1978;187:449-464. 24. Iskandrian AS, Hakki AH: Determinants of the changes in left REFERENCES ventricular end-diastolic volume during upright exercise in patients with coronary artery disease. Am Heart J 1986;112: 1. Myers J, Froelicher VF: Hemodynamic determinants of exercise 441-446. capacity in chronic heart failure. Ann Intern Med 1991;115: 25. Younis LT, Melin JA, Robert AR, Detry JMR: Influence of age 377-386. and sex on left ventricular volumes and ejection fraction during upright exercise in normal subjects. Eur Heart J 1990; 2. McKirnan MD, Sullivan M, Jensen D, Froelicher VF: Treadmill 11:916-924. performance and cardiac function in selected patients with 26. Goodman JM, Lefkowitz CA, Liu PP, et al: Left ventricular func- coronary heart disease. J Am Coll Cardiol 1984;3:253-261. tional response to moderate and intense exercise. Can J Sport Sci 1991;16:204-209. 3. Hammond HK, Kelley TL, Froelicher VF: Noninvasive testing in 27. Plotnick GD, Becker L, Fisher ML, et al: Use of the Frank–Starling the evaluation of myocardial ischemia: Agreement among tests. mechanism during submaximal versus maximal upright exercise. J Am Coll Cardiol 1985;5:59-69. Am J Physiol 1986;251:H1101-H1105. 28. Wyns W, Melin JA, Vanbutsele RJ, et al: Assessment of right and 4. Clark AL, Poole-Wilson PA, Coats AJ: Exercise limitation in left ventricular volumes during upright exercise in normal men. chronic heart failure: Central role of the periphery. J Am Coll Eur Heart J 1982;3:529-536. Cardiol 1996;28:1092-1102. 29. Renlund DG, Lakatta EG, Fleg JL, et al: Prolonged decrease in cardiac volumes after maximal upright bicycle exercise. J Appl 5. American College of Sports Medicine: Guidelines for Exercise Physiol 1987;63:1947-1955. Testing and Prescription, 6th ed. Philadelphia, Lea & Febiger, 1999. 30. Schairer JR, Stein PD, Keteyian S, et al: Left ventricular response to submaximal exercise in endurance-trained athletes and seden- 6. Brooks GA: Intra- and extra-cellular lactate shuttles. Med Sci tary adults. Am J Cardiol 1992;70:930-933. Sports Exerc 2000;32:790-799. 31. Tomai F, Ciavolella M, Crea F, et al: Left ventricular volumes during exercise in normal subjects and patients with dilated 7. Brooks GA: Lactate shuttles in nature. Biochem Soc Trans 2002; cardiomyopathy assessed by first-pass radionuclide angiography. 30:258-264. Am J Cardiol 1993;72:1167-1171. 32. Tomai F, Ciavolella M, Gaspardone A, et al: Peak exercise left 8. Myers J, Ashley E: Dangerous curves: A perspective on exercise, ventricular performance in normal subjects and in athletes lactate, and the anaerobic threshold. Chest 1997;111:787-795. assessed by first-pass radionuclide angiography. Am J Cardiol 1992;70:531-535. 9. Saltin B, Henricksson J, Hugaard E, Andersen P: Fiber types and 33. Crawford MH, Amon KW, Vance WS: Exercise 2-dimensional metabolic potentials of skeletal muscles in sedentary man and echocardiography. Quantitation of left ventricular performance endurance runners. Ann NY Acad Sci 1977;301:3-29. in patients with severe angina pectoris. Am J Cardiol 1983; 51:1-6. 10. Hammond K. Froelicher VF: Normal and abnormal heart rate 34. Freeman MR, Berman DS, Staniloff H, et al: Comparison of responses to exercise. Prog Cardiovasc Dis 1985;27:271-296. upright and supine bicycle exercise in the detection and evaluation of extent of coronary artery disease by equilibrium radionuclide 11. Slutsky R, Karliner J, Ricci D, et al: Response of left ventricular ventriculography. Am Heart J 1981;102:182-189. volume to exercise in man assessed by radionuclide equilibrium 35. Shen WF, Roubin GS, Choong CY-P, et al: Left ventricular angiography. Circulation 1979;60:565. response to exercise in coronary artery disease: Relation to myocardial ischemia and effects of nifedipine. Eur Heart J 1985; 12. Cotsamire DL, Sullivan MJ, Bashore TM, Leier CV: Position as a 6:1025-1031. variable for cardiovascular responses during exercise. Clin Cardiol 36. Kalisher AL, Johnson LL, Johnson YE, et al: Effects of propranolol 1987;10:137-142. and timolol on left ventricular volumes during exercise in patients with coronary artery disease. J Am Coll Cardiol 1984; 13. Stein RA, Michelli D, Fox EL, Krasnow N: Continuous ventricular 3:210-218. dimensions in man during supine exercise and recovery. Am J 37. Myers J, Wallis J, Lehmann K, et al: Hemodynamic determinants Cardiol 1978;41:655-660. of maximal ventilatory oxygen uptake in patients with coronary artery disease. Circulation 1991;84:II-150. 14. Bevegard BS, Shepherd JT: Regulation of circulation during 38. Ginzton LE, Conant R, Brizendine M, Laks MM: Effect of long- exercise in man. Physiol Rev 1967;47:178-213. term high-intensity aerobic training on left ventricular volume during maximal upright exercise. J Am Coll Cardiol 1989;14: 15. Poliner LR, Dehmer GJ, Lewis SE, et al: Left ventricular per- 364-371. formance in normal subjects: A comparison of the responses to 39. Sullivan MJ, Higginbotham MB, Cobb FR: Exercise training in exercise in the upright and supine positions. Circulation patients with severe left ventricular dysfunction. Hemodynamic 1980;62:528-534. and metabolic effects. Circulation 1988;78:506-515. 40. Delahaye N, Cohen-Solal A, Faraggi M, et al: Comparison of left 16. Bristow JD, Klosten FE, Farrahi C, et al: The effects of supine ventricular responses to the six-minute walk test, stair climbing, exercise on left ventricular volume in heart disease. Am Heart J and maximal upright bicycle exercise in patients with congestive 1966;71:319-329. heart failure due to idiopathic dilated cardiomyopathy. Am J Cardiol 1997;80:65-70. 17. Adams KF, Vincent LM, McAllister SM, et al: The influence of age 41. Lapu-Bula R, Robert A, Van Craeynest D, et al: Contribution of and gender on left ventricular response to supine exercise in exercise-induced mitral regurgitation to exercise stroke volume asymptomatic normal subjects. Am Heart J 1987;113:732-742. and exercise capacity in patients with left ventricular systolic dysfunction. Circulation 2002;106:1342-1348. 18. Granath A, Jonsson B, Strandall T: Circulation in healthy old men, 42. Jensen DG, Genter F, Froelicher VF, et al: Individual variability of studied by right heart catheterization at rest and during radionuclide ventriculography in stable coronary artery disease exercise in supine and sitting position. Acta Med Scand 1964; patients over one year. Cardiology 1984;71:255-265. 176:425-446. 43. Hakki AH, Iskandrian AS: Determinants of exercise capacity in patients with coronary artery disease: Clinical implications. 19. Sharma B, Goodwin JF, Raphael MJ, et al: Left ventricular J Cardiac Rehabil 1985;5:341-348. angiography on exercise: A new method of assessing left ventricular function in ischemic heart disease. Br Heart J 1976; 38:59-70. 20. Jones R, McEwan P, Newman G, et al: Accuracy of diagnosis of coronary artery disease by radionuclide measurement of left ventricular function during rest and exercise. Circulation 1981;64:586-601. 21. Manyeri DE, Kostuk WJ: Right and left ventricular function at rest and during bicycle exercise in the supine and sitting positions in normal subjects and patients with coronary artery disease. Assessment by radionuclide ventriculography. Am J Cardiol 1983; 51:36-42. 22. Higginbotham MB, Morris KG, Williams RS, et al: Regulation of stroke volume during submaximal and maximal upright exercise in normal man. Circ Res 1986;58:281-291. 23. Rerych SK, Scholz PM, Newman GE, et al: Cardiac function at rest and during exercise in normals and in patients with coronary

CHAPTER two Exercise Testing Methodology Despite the many advances in technology related instances not enough data are yet available to to the diagnosis and treatment of cardiovascular validate them, so they should be used judiciously. disease, the exercise test remains an important Also, what about the various ancillary tests diagnostic modality. Its numerous applications, and the nonexercise stress modalities? In this widespread availability, and high yield of clini- chapter, we will address basic methodology cally useful information continue to make it and comment on the impact these advances in an important gatekeeper for more expensive and technology have had. We start by listing the invasive procedures. However, the many different advantages and disadvantages of exercise ECG approaches to the exercise test have been a draw- testing. These considerations are important back to its proper application. Excellent guide- because the health care provider must evaluate lines have been updated by organizations such the suitability of the various testing modalities in as the American Heart Association, American each situation. Association of Cardiovascular and Pulmonary Rehabilitation, and American College of Sports ADVANTAGES AND Medicine. These guidelines are based on a multi- DISADVANTAGES tude of research studies over the last 30 years OF EXERCISE and have led to greater uniformity in methods. ECG TESTING Nevertheless, in many laboratories, methodology remains based on tradition, convenience, equip- ADVANTAGES OF THE STANDARD ment, or personnel available. EXERCISE ECG TEST New technology, while adding convenience, 1. Low cost has also raised new questions with regard to 2. Availability of trained personnel methodology. For example, all commercially 3. Exercise capacity determined available systems today depend upon computers. 4. Patient acceptability Do computer-averaged exercise electrocardio- 5. Takes less than an hour to accomplish grams (ECGs) improve test accuracy, and should 6. Convenience the practitioner rely on this processed informa- 7. Availability tion or on the raw data? What about the many 8. Long history of use, validation of responses, computerized exercise scores that now can so easily be calculated? Technology has changed the application of multivariate scores exercise-testing laboratory environment, and concerns such as these have arisen. Though many 11 of these techniques are attractive, in many

12 E X E R C I S E A N D T H E H E A R T DISADVANTAGES OF EXERCISE safe, the population referred for this procedure ECG TESTING usually is at high risk for coronary events. Irving and Bruce5 have reported an association between 1. Limited sensitivity and specificity exercise-induced hypotension and ventricular 2. Inability to localize ischemia or coronary lesions. fibrillation. Shepard6 has hypothesized the follow- 3. No estimate of left ventricular (LV) function ing risk levels for exercise: (1) three or four 4. Not suitable for certain patients. times normal in a cross-country foot race, (2) 6 to 5. Requires cooperation and the ability to walk or pedal 12 times normal when patients at risk for coro- nary artery disease (CAD) are performing unac- a cycle ergometer. customed exercise, and (3) as high as 60 times normal when patients with existing CAD are per- SAFETY PRECAUTIONS forming exercise in a stressful environment, such AND RISKS as a physician’s office. Cobb and Weaver7 esti- mated the risk to be over 100 times in the latter The safety precautions outlined by the American situation and point out the dangers of the recov- Heart Association are very explicit in regard to the ery period. The risk of exercise testing in patients requirements for exercise testing. Everything with CAD cannot be disregarded even with its necessary for cardiopulmonary resuscitation excellent safety record. Studies documenting the must be available, and regular drills should be risks of exercise training are presented in more performed to ascertain that both personnel and detail in Chapter 12. equipment are prepared for a cardiac emergency. The classic survey of clinical exercise facilities by Indications to stop an exercise test, in addition Rochmis and Blackburn in 19711 showed exercise to the factors to consider in assessing the degree testing to be a safe procedure, with approximately of exertion, are outlined in Table 2-1. Most prob- only one death and five nonfatal complications lems can be avoided by having an experienced per 10,000 tests. Perhaps because of an expanded physician, nurse, or exercise physiologist stand- knowledge concerning indications, contraindica- ing next to the patient, measuring blood pressure, tions, and endpoints, data suggest that maximal and assessing patient appearance during the test. exercise testing is safer today than 30 years ago. The exercise technician should operate the recorder In 1989, Gibbons et al2 reported the safety of and treadmill, take the appropriate tracings, enter exercise testing in 71,914 tests conducted over data on a form, and alert the physician to any a 16-year period. The complication rate was abnormalities that may appear on the monitor 0.8 per 10,000 tests. In a recent survey of 71 exer- scope. If the patient’s appearance is worrisome, if cise testing laboratories throughout the Veterans systolic blood pressure drops or plateaus, if there Administration Health Care System including are alarming ECG abnormalities, if chest pain 75,828 tests, we observed an event rate of 1.2 per occurs and becomes worse than the patient’s usual 10,000 tests.3 The fact that the event rate was pain, or if the patient wants to stop the test for any similar between a clinically referred population reason, the test should be stopped, even at a sub- (the Veterans Administration, a higher risk maximal level. In most instances, a symptom- group), and a generally healthier population2 limited maximal test is preferred, but it is usually underscores the fact that the test is extremely advisable to stop if 0.2 mV of additional ST-segment safe. Gibbons et al2 suggested that the low compli- elevation occurs, or if 0.2 mV of flat or downsloping cation rate in their study was due to the inclusion ST-segment depression occurs. In some patients of a cool-down walk, but we have observed a low estimated to be at high risk because of their clinical rate of ventricular tachycardia,4 and a low overall history, it may be appropriate to stop at a submaxi- complication rate3 despite having patients assume mal level, as it is not unusual for severe ST-segment a supine position immediately after the test and depression, dysrhythmias, or both to occur in the despite exercising higher risk patients. This issue is postexercise period. If the measurement of maximal addressed in more detail in Chapter 13, and a sum- exercise capacity or other information is needed, it mary of these studies is presented in Table 13-6. may be preferable to repeat the test later, once the patient has demonstrated a safe performance of a However, it is important to note that there submaximal workload. have been reports of complications, including acute infarctions and deaths, associated with Exercise testing should be an extension of the exercise testing. Although the test is remarkably history and physical examination. A physician obtains the most information by being present to talk with, observe, and examine the patient in

C H A P T E R 2 Exercise Testing Methodology 13 TA B L E 2 - 1 . Indications for terminating an exercise test and assessment of maximal effort Absolute Reasons or Indications to Terminate Acute myocardial infarction Severe angina—chest pain score of 4 out of 4 Exertional hypotension—a drop in systolic blood pressure of ≥ 10 mmHg, or drop below the value obtained in the standing position prior to testing, particularly in patients who have heart failure, have had a prior myocardial infarction, or are exhibiting signs or symptoms of ischemia ≥ 1.0 mm ST elevation in leads without diagnostic Q waves Serious arrhythmias—ventricular tachycardia, third-degree heart block Poor perfusion as judged by skin temperature and cyanosis Neurologic signs—confusion, lightheadedness, vertigo Technical problems—inability to interpret the ECG pattern; any malfunction of the recording or monitoring device; inability to measure the systolic blood pressure Patient’s request to terminate Relative Reasons or Indications to Terminate The following indications may be superseded if done so in the context of good clinical judgment. Increasing chest pain—chest pain score of 3 out of 4 ≥ 2.0 mm horizontal or downsloping ST depression Pronounced fatigue or shortness of breath Wheezing Leg pain or claudication Increase in systolic blood pressure to 250 mmHg or increase in diastolic blood pressure to 115 mmHg Less serious arrhythmias than those in preceding list (frequent or mutifocal premature ventricular contractions, supraventricular tachycardia, bradyarrhythmias) Bundle branch block or another rate-dependent intraventricular conduction defect that cannot be distinguished from ventricular tachycardia Assessment of Maximal Effort As no single marker of effort is usually specifically indicative of a maximal effort, it is best to consider multiple responses. Borg scale 17-20 Signs of fatigue, profound shortness of breath, or exhaustion Age-predicted maximal heart rate, with a population-specific regression equation Expired gas measurements, including respiratory exchange ratio (>1.10) conjunction with the test. A brief physical exami- supervision of a physician. However, the physician nation should always be performed to rule out any must be in the immediate vicinity or on the prem- contraindications that exist. Accordingly, indi- ises or the floor and available for emergencies.8,9 viduals who supervise exercise tests must have In situations where the patient is deemed to be at the cognitive and technical skills necessary to be higher risk for an adverse event during exercise competent to do so. The American College testing, the physician should be physically pres- of Cardiology, American Heart Association, and the ent in the exercise testing room to personally American College of Physicians, with broad supervise the test. Such cases include, but are not involvement from other professional organizations limited to, patients with recent acute coronary involved with exercise testing, such as the syndrome or myocardial infarction (within 7 to American College of Sports Medicine, have out- 10 days), severe LV dysfunction, severe valvular lined the cognitive skills needed to competently stenosis (e.g., aortic stenosis), or known complex supervise exercise tests.8 These skills include arrhythmias. The physician’s reaction to signs or knowledge of appropriate indications and con- symptoms should be moderated by the informa- traindications to testing, an understanding of risk tion the patient gives regarding his or her usual assessment, the ability to recognize and treat com- activity. If abnormal findings occur at levels of plications, and knowledge of basic cardiovascular exercise that the patient usually performs, then it and exercise physiology, along with the ability to may not be necessary to stop the test for them. interpret the test in different patient populations. Also, the patient’s activity history should help determine appropriate work rates for testing. The need for physician presence during exer- cise testing has been the subject of a great deal CONTRAINDICATIONS of discussion in the past. In many cases, exercise tests can be supervised by properly trained and Table 2-2 lists the absolute and relative con- competent exercise physiologists, physical thera- traindications to performing an exercise test. pists, nurses, physician assistants, or medical technicians who are working under the direct

14 E X E R C I S E A N D T H E H E A R T TA B L E 2 - 2 . Contraindications to exercise testing Absolute Acute myocardial infarction (within 2 days) Unstable angina not stabilized by medical therapy Uncontrolled cardiac arrhythmias causing symptoms or hemodynamic compromise Symptomatic severe aortic stenosis Uncontrolled symptomatic heart failure Acute pulmonary embolus or pulmonary infarction Acute myocarditis or pericarditis Relative* Left main coronary stenosis or its equivalent Moderate stenotic valvular heart disease Electrolyte abnormalities Uncontrolled arterial hypertension† Tachyarrhythmias or bradyarrhythmias Hypertrophic cardiomyopathy and other forms of outflow tract obstruction Mental or physical impairment leading to inability to exercise adequately High-degree atrioventricular block *Relative contraindications can be superseded if benefits outweigh risks of exercise. †In the absence of definitive evidence, a systolic blood pressure of 200 mmHg and a diastolic blood pressure of 110 mmHg are reasonable criteria. Good clinical judgment should be foremost The tapering of beta-blockers should be over- in deciding the indications and contraindica- seen by a physician. tions for exercise testing. In selected cases 4. If the reason for the exercise test is not appar- with relative contraindications, testing can ent, the referring physician should be contacted provide valuable information even if performed such that this gets clarified. submaximally. 5. A 12-lead ECG should be obtained in both the supine and standing positions. The latter is an PATIENT PREPARATION important rule, particularly for patients with known heart disease, since an abnormality Preparations for exercise testing include the may prohibit testing. On rare occasions, a following: patient referred for an exercise test will instead be admitted to the coronary care unit. 1. The patient should be instructed not to eat or 6. The patient should receive careful explana- smoke at least 2 to 3 hours prior to the test and tions of why the test is being performed and of to come dressed for exercise. the testing procedure, including its risks and possible complications. A demonstration 2. A brief history and physical examination (par- should be provided of how to get on and off the ticularly for patients with systolic murmurs) treadmill and how to walk on it. The patient should be performed to rule out any contra- should be told that he or she can hold on to the indications to testing (see Table 2-2). handrails initially but then should use the rails only for balance (discussed in the following 3. Specific questioning should determine which section). drugs are being taken, and potential electro- lyte abnormalities should be considered. TREADMILL The labeled medication bottles should be brought along so that they can be identified The treadmill should have front and side rails for and recorded. It is generally no longer con- patients to steady themselves, and some patients sidered necessary for most patients to stop may benefit from the helping hand of the person taking their beta-blockers prior to testing. administering the test. The treadmill should be If it is considered necessary to do so in calibrated at least monthly. Some models can be selected patients, they should be stopped greatly affected by the weight of the patient and gradually in order to avoid the “rebound” phenomenon, which can be dangerous.

C H A P T E R 2 Exercise Testing Methodology 15 will not deliver the appropriate workload to heavy After the test, responsibility rests with the physi- patients. An emergency stop button should be cian for prompt interpretation and consideration readily available to the staff only. A small platform of the implications of the test. Communication of or stepping area at the level of the belt is advisable these results to the patient is necessary—with so that the patient can start the test by “pedaling” advice concerning adjustments in lifestyle—and the belt with one foot prior to stepping on. After this should be done immediately after the test is they become accustomed to the treadmill, performed. patients should not grasp the front or side rails, as this decreases the work performed and thus the The second consideration should be adherence oxygen uptake, which increases exercise time, to proper standards of care during performance of resulting in an overestimation of exercise capac- the test. Exercise testing should be carried out ity. Gripping the handrails also increases ECG only by persons thoroughly trained in its admin- muscle artifact. For patients who have difficulty istration and in the prompt recognition of prob- letting go of the handrails, it is helpful to have lems that may arise. A physician trained in them take their hands off the rails, close their exercise testing and resuscitation should be read- fists, and extend one finger on each hand, touch- ily available during the test to make judgments ing the rails only with those fingers in order to concerning test termination. Resuscitative equip- maintain balance while walking. Some patients ment should always be available. As mentioned may require a few moments before they feel com- above, an updated joint position statement from fortable enough to let go of the handrails, but we several professional organizations was published strongly discourage grasping the handrails after in 2000, outlining the standards for physician the first minute of exercise. competence for performing exercise testing.8 LEGAL IMPLICATIONS OF BLOOD PRESSURE MEASUREMENT EXERCISE TESTING Although numerous clever devices have been In any procedure with a risk of complications, it is developed to automate blood pressure measure- advisable to make certain the patient understands ment during exercise, none can be recommended. the situation and acknowledges the risks. Some The time-proven method of the physician hold- physicians feel that informing patients of the ing the patient’s arm with a stethoscope placed risks involved will occasionally make them overly over the brachial artery remains the most reliable anxious or discourage them from performing the method to obtain the blood pressure. The test. Because of this, and the fact that a signed patient’s arm should be free of the handrails so consent form does not necessarily protect a that noise is not transmitted up the arm. It is physician from legal action, there has been less sometimes helpful to mark the brachial artery. insistence on consent forms. However, a great An anesthesiologist’s auscultatory piece or an deal of case law exists suggesting that a written electronic microphone can be fastened to the informed consent before the exercise test is arm. A device that inflates and deflates the cuff on important to protect the patient, physician, and the push of a button can also be helpful. If systolic institution. blood pressure appears to be increasing sluggishly or decreasing, it should be taken again immedi- Establishment of physician-patient communi- ately. If a drop in systolic blood pressure of 10 to cation before and after performance of the exer- 20 mmHg or more occurs, or if it drops below cise test should be the first legal consideration. the value obtained in the standing position prior A test should not be performed without first to testing, the test should be stopped. This is obtaining the patient’s informed consent, after particularly important in patients who have the patient is made aware of the potential risks heart failure, a prior myocardial infarction, or are and benefits of the procedure. A physician may be exhibiting signs or symptoms of ischemia. An held responsible in the event of a major untoward increase in systolic blood pressure to 250 mmHg event, even if the test is carefully performed, in or an increase in diastolic blood pressure to the absence of informed consent. The argument 115 mmHg are also indications to stop the can be made that the patient would not have test. The clinical implications of abnormal undergone the procedure had he or she been blood pressure responses to the exercise test are made aware of the risks associated with the test. discussed in detail in Chapter 5.

16 E X E R C I S E A N D T H E H E A R T ECG RECORDING INSTRUMENTS ECG signal produced may not be a true represen- tation of the actual waveform and in fact may be Many technologic advances in ECG recorders dangerously misleading. Also, the instruments have taken place. The medical instrumentation that make computer ST-segment measurements industry has promptly complied with specifica- cannot be totally reliable as they are based on tions set forth by various professional groups. imperfect algorithms. For instance, the algorithm Machines with high-input impedance ensure that measures QRS end at 70 or 80 msec after that the voltage recorded graphically is equivalent the peak of the R wave can hardly be valid, partic- to that on the surface of the body despite the ularly with a changing heart rate. high natural impedance of the skin. There remains some concern about mismatching lead Because of physician insistence on having impedance, which can result in distortion. exercise tracings as clean as resting tracings, Optically isolated buffer amplifiers have ensured manufacturers have taken some worrisome steps patient safety, and machines with a frequency with filtering and ECG presentation. One such response from 0 to 100 Hz are commercially approach is “linked medians,” in which averages available. The 0 Hz lower end is possible because are connected together at the same R-R interval DC coupling is technically feasible. as raw data. Even though these tracings are appropriately labeled, and often presented with a Some ECG equipment has monitoring and channel of raw data as well, most physicians do diagnostic modes, particularly equipment used in not realize that they are dealing with created coronary care units. The diagnostic mode follows waveforms instead of raw data. diagnostic instrument specifications with a fre- quency response from 0.05 to 100 Hz. In the mon- ECG PAPER RECORDERS itor mode, there can be distortion of the ECG. The monitor mode is available to lessen the effects of For some patients it is advantageous to have electrical interference, motion, and respiration a recorder with a slow paper speed option such as on the ECG and should not be used for exercise 5 mm/sec. This speed makes it possible to record testing. The type of distortion is affected by the an entire exercise test and reduces the likelihood ECG waveform that is presented. If the ECG wave- of missing any dysrhythmias when specifically form is a tall R wave without an S wave, the evaluating patients with these problems. A faster ST-segment distortion can be different than if paper speed of 50 mm/sec can be helpful for mak- there is an R wave followed by a large S wave. In ing accurate ST-segment slope measurements. general, an inadequate low-frequency response Many different types of ECG paper can be used. can greatly decrease the Q- and R-wave amplitude Wax-treated paper is known to retain an ECG image and create S waves. Alteration of the 25 to 45 Hz for 20 years or longer; however, it is pressure- frequency response is the most common cause of sensitive and easily marked. Thermochemically ST-segment distortion found in tracings with treated paper is sturdy and resists marking. There abnormal ST segments. Some of the newer filter- are many different types of thermochemically ing techniques delay the appearance of the ECG treated paper, and the life expectancy of images signal on the monitor screen by several seconds. recorded on them is usually adequate. However, at least one instance of ECG paper losing a WAVEFORM AVERAGING recorded image resulted in legal action by a hos- pital against a manufacturer. Ceramic-coated Digital averaging techniques have made it possi- paper is very sturdy and comparable in price to ble to average ECG signals to remove noise. There other ECG papers. It has a hard finish with a high is a need for consumer awareness in these areas, contrast, which makes it durable and easy to since most manufacturers do not specify how the interpret. Untreated paper is the cheapest ECG use of such procedures modifies the ECG. Signal paper, but the ink-jet and carbon-transfer tech- averaging can actually distort the ECG signal. niques characteristically produce fuzzy images on These techniques are attractive because they can untreated paper. The ink-jet and carbon-transfer produce a clean tracing in spite of poor skin recorders are available with six channels and are preparation. However, the common expression expensive, but they do have an excellent upper- used by computer scientists, “Garbage in, garbage frequency response for phonocardiography. The out,” has never been more applicable than to the ceramic paper also requires an ink-jet rather than computerized ECG. The clean-looking exercise a heat stylus. Ink-jet recorders require more

C H A P T E R 2 Exercise Testing Methodology 17 maintenance and have largely been replaced by slightly more work on a treadmill. This is because thermal head printers. Copying can be a problem a greater muscle mass is involved, in addition to as some photographic reproduction machines the fact that most subjects are more familiar with poorly copy reds and blues. walking than cycling. Thermal head printers have nearly totally Numerous modalities have been used to replaced all other types of printers. These provide dynamic exercise for exercise testing, recorders are remarkable in that they can use including steps, escalators, and ladder mills. blank thermal paper and write out the grid as well Today, however, the bicycle ergometer and the as the ECG, vector loops, and alphanumerics. treadmill are the most commonly used dynamic They can record graphs and figures as well as exercise devices. The bicycle ergometer is usually tables and typed reports. They are totally digitally cheaper, takes up less space, and makes less noise. driven and can produce very high resolution Upper body motion is usually reduced, but care records. The paper price is comparable to that for must be taken so that the arms do not perform paper used with other recorders, and these isometric exercise. The workload administered by devices are themselves reasonably priced and very the simple bicycle ergometers is not well cali- durable, particularly because no stylus is needed. brated and is dependent upon pedaling speed. It Some exercise systems use a laser printer, but this can be easy for a patient to slow pedaling speed is not suitable for the exercise environment, during exercise testing and decrease the adminis- where recording the ECG is delayed by the 5 to tered workload. More modern electronically 20 seconds required for the printing to occur. braked bicycle ergometers keep the workload at a specified level over a wide range of pedaling Z-fold paper has the advantage over roll paper speeds. Electrically braked ergometers are partic- in that it is easily folded, and the study can be ularly needed for supine exercise testing. interpreted in a manner similar to paging through a book. Exercise ECGs can be micro- ARM ERGOMETRY filmed on rolls, cartridges, or in fiche cards for storage. They can also be stored in digital or Alternative methods of exercise testing are neces- analog format on magnetic media or optical sary for patients with vascular, orthopedic, or discs. The latest technology involves recording on neurologic conditions that prevent them from CD-ROM discs that are erasable and have fast performing leg exercise. Arm ergometry is one access and transfer times. These devices can be such alternative. For a given submaximal work- easily interfaced with microcomputers and can load, arm exercise is performed at a greater phys- store gigabytes of digital information. iologic cost than is leg exercise. However, at maximal effort, physiologic responses are gener- EXERCISE TEST MODALITIES ally significantly greater in leg exercise than in arm exercise. At a given power output (expressed Three types of exercise can be used to stress the as kilopond meters per minute [kpm/min] or cardiovascular system: isometric, dynamic, and a watts), heart rate, systolic and diastolic blood combination of the two. Isometric exercise, pressure, the product of heart rate and systolic defined as constant muscular contraction without blood pressure, minute ventilation, and blood or with minimal external movement (such as lactate concentration are higher during arm exer- handgrip), imposes a disproportionate pressure cise. In contrast, stroke volume and the venti- load on the left ventricle relative to the body’s abil- latory threshold (the latter expressed as a ity to supply oxygen. Dynamic exercise is defined percentage of aerobic capacity) are lower during as rhythmic muscular activity resulting in move- arm exercise than during leg exercise.10-13 ment, and it initiates a more appropriate increase Because cardiac output is nearly the same in arm in cardiac output and oxygen exchange. Since a and leg exercise at a given oxygen uptake,14 the delivered workload can be accurately calibrated elevated blood pressure during arm exercise is and the physiologic response easily measured, due to increased peripheral vascular resistance. dynamic exercise is preferred for clinical testing. With progressive workloads of dynamic exercise, This difference in cardiopulmonary and hemo- patients with CAD can be protected from rapidly dynamic responses to arm exercise as compared increasing myocardial oxygen demand. Although with leg exercise at identical workloads appears bicycling is a dynamic exercise and appropriate to be due to several factors. First, mechanical for exercise testing, most individuals perform efficiency is lower during arm exercise than

18 E X E R C I S E A N D T H E H E A R T leg exercise. This lower efficiency may reflect the ergometry in women, expressed as arm VO2 max/leg VO2 max, was 79%, compared to the involvement of smaller muscle groups and the mean value of 72% derived from seven separate static effort by the torso muscles to stabilize studies on men. A summary of the studies com- the shoulder required for arm work.15 Both fac- paring maximal heart rate responses to arm and tors could increase oxygen requirements yet not leg exercise is presented in Table 2-3. affect the external work performed by the arms. Fardy et al22 reported a greater arm VO2 max The higher rate-pressure product and estimated than leg VO2 max when aerobic capacity was expressed per milliliter of limb volume. Because myocardial oxygen demand at a given external the arms were approximately one-third the vol- workload for arm work as compared to leg work ume of the legs, and the VO2 max for the arms was may be due to increased sympathetic tone during two thirds that of the legs, VO2 max for arm exercise was twice that of the legs. The twofold arm exercise (owing to reduced stroke volume increase in oxygen use per unit of arm volume with compensatory tachycardia), isometric con- may be spurious, because arm ergometry also traction of torso muscles, or vasoconstriction in the nonexercising leg muscles.16-18 utilizes muscles of the back, shoulders, and Maximal oxygen uptake (VO2 max) during arm chest. ergometry in men generally varies between 64% Several investigators have examined the and 80% of leg ergometry VO2 max. Similarly, maximal cardiac output is lower during arm exer- ability of leg or arm exercise testing to conversely cise than during leg exercise, whereas maximal predict performance capacity of the other extrem- heart rate, systolic blood pressure, and rate- ities in able-bodied subjects. Asmussen and pressure product are comparable19 or slightly Hemmingsen23 showed that it was not possible to lower20 during arm exercise. Although women have estimate leg VO2 max from experiments with arm a lower arm VO2 max than men, it appears that work, and vice versa. Franklin et al24 found weak their aerobic capacity for arm work is not dis- proportionately inferior to men’s. Vander et al21 correlations between maximal power output found that the relationship between arm and leg (kpm/min) or VO2 max (metabolic equivalents; TA B L E 2 - 3 . Comparison of the maximal heart rate (HR max) in response to arm and leg exercise in men and women HR max difference HR max (beats/min) Ratio legs/arms beats/min Investigator Arms Legs Ratio arms/legs (%) Men (normal) 177 190 13 93 Astrand et al. (1968) 178 188 10 95 Stenberg et al. (1967) 173 195 22 89 Bar-Or and Zwiren (1975) 176 189 13 93 Bergh et al. (1976) 184 193 9 95 Davis et al. (1976) 174 185 11 94 Fardy et al. (1977) 174 195 21 89 Magel et al. (1978) 183 186 3 98 Bouchard et al. (1979) 167 190 23 88 De Boer et al. (1982) 169 179 10 94 Sawka et al. (1982) 172 184 12 93 Franklin et al. (1983) 160 160 0 100 Balady et al. (1986) 122 129 7 95 Men (cardiac patients) 142 145 3 98 Schwade et al. (1977) 101 109 8 93 DeBusk et al. (1978) Balady et al. (1985) 169 177 8 95 Women (normal) 164 174 11 94 Vander et al. (1984)21 Mean Results

C H A P T E R 2 Exercise Testing Methodology 19 METs) for arm and leg exercise. Schwade et al25 supine bicycle exercise has been observed and has also reported a poor correlation (r = 0.37) been used to separate heart disease patients from between peak workloads during arm and leg normal subjects. Exercise factor, or the increase exercise in patients with ischemic heart disease. in cardiac output for a given increase in oxygen uptake, is based on studies of normal subjects. To determine the sensitivity of arm exercise in For every 100 mL increase in oxygen consump- detecting CAD, Balady et al26, 27 tested 30 patients tion, cardiac output should increase by 500 mL. with angina pectoris using both arm ergometry Left ventricular filling pressure does not increase and a treadmill before coronary angiography. All in proportion to work in normal persons, but it patients had at least 70% diameter reduction in often increases in patients with heart disease. one or more major coronary arteries. Ischemic ST Radionuclide imaging has shown that the ejection depression (≥0.1 mV) or angina occurred more fraction usually increases in normal subjects frequently with leg exercise (86%, 26 patients) but can decrease during exercise in patients with than with arm exercise (40%, 12 patients). No sig- ischemia or left ventricular dysfunction. However, nificant difference in peak rate-pressure product many patients with heart disease demonstrate dis- was seen between tests, although peak VO2 was cordance between their disease and ventricular greater during leg exercise than during arm function and can respond normally to exercise. exercise (18 versus 13 mL/kg/min). For concor- dantly positive tests, oxygen uptake at the onset BICYCLE ERGOMETER VERSUS of ischemia was significantly lower during arm TREADMILL testing than during leg testing (12 versus 17 mL/ kg/min). No significant difference in heart rate In most studies comparing the upright cycle was seen between tests at the onset of ischemia. ergometer with treadmill exercise, maximal heart Thus, arm exercise testing is a reasonable, but not rate values have been demonstrated to be roughly equivalent, alternative to leg exercise testing in similar, whereas maximal oxygen uptake has been patients who cannot perform leg exercise. shown to be 6% to 25% greater during treadmill exercise.28-31 Early hemodynamic studies by SUPINE VERSUS UPRIGHT Niederberger et al32 concluded that bicycle exer- EXERCISE TESTING cise constitutes a greater stress on the cardio- vascular system for any given oxygen uptake than A great deal of the information available on hemo- does treadmill exercise. The clinical importance dynamic responses to exercise has come from of these findings in relation to patients with car- supine exercise, mostly because cardiac catheter- diovascular disease undergoing exercise testing is ization is required to obtain much of this infor- that slightly higher maximal oxygen uptakes are mation. However, there are marked differences achieved with slightly less hemodynamic stress between the body’s response to acute exercise in when treadmill exercise is used. Wicks et al33 the supine versus upright position. During supine reported similar ECG changes with treadmill test- bicycle exercise, stroke volume and end-diastolic ing as compared to bicycle testing in coronary volume do not change much from values obtained patients. Rather than for any clinical reason, how- at rest, whereas in the upright position, these ever, the treadmill is the most commonly used values increase during mild work and then dynamic testing modality in the United States plateau. Naturally, exercise capacity is markedly because patients are more familiar with walking lower in the supine position than during upright than they are with bicycling. Patients are more cycling. In patients with heart disease, left ven- likely to give the muscular effort necessary to tricular filling pressure is more likely to increase adequately increase myocardial oxygen demand during exercise in the supine position than in the by walking than by cycling. upright position. When patients with angina per- form identical submaximal bicycle workloads in EXERCISE WITH INTRACARDIAC supine and upright positions, heart rate is higher CATHETERS and angina will develop at a lower double product while the patient is supine. ST-segment depres- Exercise testing with intracardiac catheters has sion is often greater in the supine position significant advantages over alternative diagnostic because of the greater left ventricular volume. methods for: (1) separation of cardiac from As with upright exercise, a linear relationship between cardiac output and oxygen uptake during

20 E X E R C I S E A N D T H E H E A R T pulmonary dyspnea, (2) separation of LF systolic exercise suggests that valve disease rather than from diastolic dysfunction, and (3) quantitative concomitant pulmonary disease is the cause of evaluation of the clinical significance of valvular clinical symptoms. disease. EXERCISE PROTOCOLS Cardiac versus Pulmonary Dyspnea. Patients with severe chronic obstructive pulmonary The many different exercise protocols in use have disease (COPD) have clinical findings that make led to some confusion regarding how physicians the assessment of left ventricular function compare tests between patients and serial tests on extremely difficult. Many patients with COPD the same patient. The most common protocols, have left-sided heart disease secondary to CAD, their stages, and the predicted oxygen cost of each hypertension, or left-sided valvular disease. In stage are illustrated in Figure 2-1. When treadmill patients with left-sided heart disease, the common and cycle ergometer testing were first introduced denominator for cardiac dyspnea is elevation of left into clinical practice, practitioners adopted pro- atrial pressure. This leads to elevation of the pul- tocols used by major researchers such as Balke monary wedge pressure, which leads to increased and Ware,34 Astrand and Rodahl,35 Bruce,36 and pulmonary interstitial fluid, decreased pulmonary Ellestad37 and their coworkers. In 1980, Stuart compliance, and dyspnea. In contrast, significant and Ellestad38 surveyed 1375 exercise laboratories elevation of left atrial or pulmonary wedge pres- in North America and reported that of those sure is unusual in uncomplicated COPD cases. performing treadmill testing, 65.5% use Measurement of rest/exercise wedge pressure the Bruce protocol for routine clinical testing. allows one to distinguish the pathophysiology of A survey published in 2001 among 71 exercise COPD from left-sided heart disease. In the former laboratories at Veterans Administration Medical case, pulmonary artery pressure may rise markedly, Centers indicated that the percentage of exercise but pulmonary wedge pressure will remain below laboratories primarily using the Bruce protocol 20 mmHg even with maximal supine exercise. is similar to that 20 years earlier.3 Thus, this pro- In left-sided heart disease, a pulmonary wedge tocol remains widely used. A disadvantage of the pressure greater than 25 mmHg often occurs at Bruce protocol is that it uses relatively large maximal exercise. and unequal 2 to 3 MET increments in work every 3 minutes. Large and uneven work increments Left Ventricular Systolic versus Diastolic such as these have been shown to result in a ten- Dysfunction. Left ventricular systolic dysfunction dency to overestimate exercise capacity.9,30,31,39 with a resultant increase in left ventricular volume In part for this reason, many investigators, along leads to an increase in diastolic filling pressure. with exercise testing guidelines, have since rec- The patient with heart failure after a myocardial ommended protocols with smaller and more infarction is the classic example of systolic dysfunc- equal increments.9,30,31,39-42 tion. In hypertrophic cardiomyopathy, systolic or contractile function can be normal or even better In a classic study, Redwood et al42 performed than normal, but a thick, noncompliant ventricle serial testing in patients with angina and reported that cannot readily fill leads to an increased that work rate increments that were too rapid pulmonary wedge pressure. Diastolic dysfunction resulted in a reduced exercise capacity and could is characterized by a normal cardiac output for a not be reliably used for studying the effects of given workload, but this output comes at the therapy. When excessive work rates were used, the expense of an elevated filling pressure. The dis- reduction in myocardial oxygen demand as a tinction between systolic and diastolic function result of nitroglycerin administration was minor, requires the measurement of cardiac output. suggesting that protocols placing heavy and abrupt demands on the patient may mask a poten- Quantitation of Valvular Disease. Patients whose tial salutary effect of an intervention. These inves- symptoms seem out of proportion to their valvular tigators recommended that the protocol be disease can be assessed using these invasive tech- individualized for each patient to elicit angina niques. In the case of significant valvular lesions, within 3 to 6 minutes. Smokler et al43 reported exercise leads to an increase in pulmonary wedge that among 40 pairs of treadmill tests conducted pressure. Forward output may be maintained until within a 6-month period, tests that were less than late in their course. Elevation of exercise pul- 10 minutes in duration showed a much greater monary wedge pressure at symptom-limited percentage of variation than those that were

Functional C H A P T E R 2 Exercise Testing Methodology 21ClinicalO2 costBicycle class status ■ FIGURE 2-1 ml/kg/min METS ergometer Treadmill protocols METS NormalThe oxygen cost per stage for most of the commonly used treadmill protocols. and Bruce Balke-Ware USAFSAM “Slow” McHenry Stanford ACIP CHF I USAFSAM II Healthy, dependent on age, activity 1 watt = 6.1 3 min % grade %% III Sedentary health IV Limited Kpm/min stages at Symptomatic MPH %GR 3.3 MPH grade grade 5.5 20 1 min at at 56.0 16 For 70 kg stages MPH %GR 3 MPH 2 MPH 16 52.5 body 5.0 18 MPH %GR 49.0 weight 26 45.5 42.0 15 Kpm/min 25 3.3 25 MPH %GR 3.4 24.0 15 38.5 24 35.0 31.5 14 1500 4.2 16 23 3.3 21 3.1 24.0 14 28.0 13 22 24.5 21 3.3 20 22.5 13 21.0 20 17.5 12 1350 19 3.3 18 20.0 3.0 21.0 MPH %GR 12 14.0 18 10.5 11 1200 3.4 14 17 3.3 15 MPH %GR 3.3 15 17.5 3.0 17.5 3.4 14.0 11 10 1050 16 2 25 3.3 12 15.0 10 7.0 15 12.5 3.0 14.0 3.0 15.0 3.5 9 900 14 3.3 10 2 20 3.3 9 10.0 3.0 12.5 9 8 13 8 7 750 2.5 12 12 17.5 3.0 10.5 3.0 10.0 11 7 10 6 9 5 4 6 600 8 3.3 5 2 15 3.3 6 7.5 14.0 3.0 7.0 3.0 7.5 3 7 2 10 2 5 450 1.7 10 6 3.3 0 5.0 10.5 2.0 10.5 5 2.0 0 25 4 300 1.7 5 4 2.0 3 2.5 7.0 3.0 3.0 2.0 7.0 3 2 20 0 3.5 2.5 2.0 2.0 3.5 1 3 150 1.7 0 2.0 0.0 1.5 0.0 2 1 1.0 0.0 1 USAFSAM = United States Air Force School of Aerospace Medicine ACIP = asymptomatic cardiac ischemia pilot CHF = congestive heart failure (modified Naughton) Kpm/min = Kilopond meters/minute %GR = percent grade MPH = miles per hour

22 E X E R C I S E A N D T H E H E A R T greater than 10 minutes in duration. Buchfuhrer (ramp rate) to be individualized, allowing a given et al30 performed repeated maximal exercise test- test duration to be targeted. ing in five normal subjects while varying the work rate increment. Maximal oxygen uptake varied Our laboratory compared ramp treadmill and with the increment in work; the highest values bicycle tests to protocols more commonly were observed when intermediate increments used clinically.31 Ten patients with chronic heart were used. These investigators suggested that an failure, 10 with CAD who were limited by angina exercise test with work increments individualized during exercise, 10 with CAD who were asympto- to yield a duration of approximately 10 minutes matic during exercise, and 10 age-matched was optimal for assessing cardiopulmonary func- normal subjects performed three bicycle tests tion. Lipkin et al,44 on the other hand, observed (25 W/2-minute stage, 50 W/2-minute stage, and that among patients with chronic heart failure, ramp) and three treadmill tests (Bruce, Balke, small work increments yielding a long test dura- and ramp) in randomized order on different days. tion (mean 31 ± 15 minutes) resulted in reduced For the ramp tests, ramp rates on the bicycle values for maximal oxygen uptake, minute venti- and treadmill were individualized to yield a test lation, and arterial lactate compared with tests duration of approximately 10 minutes for each using more standard increments. These observa- subject. Maximal oxygen uptake was significantly tions have led a number of investigators to sug- higher (18%) on the treadmill protocols than on gest that protocols should be individualized for the bicycle protocols collectively, confirming each patient such that test duration falls within previous observations. However, only minor dif- the range of 8 to 12 minutes, and this recommen- ferences in maximal oxygen uptake were observed dation is reflected in the various exercise testing when the treadmill protocols were compared guidelines published over the last 2 decades. with one another and when the cycle ergometer protocols were compared with one another. RAMP TESTING The relationships between oxygen uptake and An approach to exercise testing that has gained work rate (predicted oxygen uptake), defined interest is the ramp protocol, in which work as a slope for each protocol, are illustrated in increases constantly and continuously (Fig. 2-2). Table 2-4. These relationships, which reflect the The popular call for “optimizing” exercise testing degree of change in oxygen uptake for a given would appear to be facilitated by the ramp increase in work (a slope of unity would suggest approach, because: (1) work increments are small that the cardiopulmonary system is adapting in and (2) this protocol allows for increases in work direct accordance with the demands of the work), were highest for the ramp tests and lowest for the protocols containing the greatest increments in work. Further, the variance about the slope 50 40 VO2 (ml/kg/min) 30 20 10 0 ■ FIGURE 2-2 0 1 2 3 4 5 6 7 8 9 10 The ramp treadmill test. Following a Time (min) 1-minute warmup at 2.0 mph/0% grade, the rate of change in speed and grade VO2MAX Warm up Ramp is individualized to yield a work rate (x axis) corresponding to an estimated exercise capacity (y axis) in approximately 10 minutes.

C H A P T E R 2 Exercise Testing Methodology 23 TA B L E 2 - 4 . Slopes in oxygen uptake versus work rate for 40 subjects performing six exercise protocols Treadmills Bicycles 25 W Bruce Balke Ramp 50 W Ramp Slope 0.62 0.79 0.80 0.69 0.59 0.78 SEE 4.0 3.4 2.5 2.3 2.8 1.7 Note: Each slope ≥ 0.78 was significantly different from each slope ≥ 0.69 (p <0.05 except Balke versus 25 W, p = 0.07). If the change in ventilatory oxygen uptake were equal to change in work rate, the slope would be 1.0. SEE, standard error of the estimate (mL O2/kg/min); 25 W, 25 w/stage; 50 W = 50 w/stage. (standard error of the estimate in oxygen uptake, A number of equipment manufacturers have mL/kg/min) was largest for the tests with the developed treadmills that can perform ramping. largest increments between stages (Bruce tread- The ramp appears to offer a better method of test- mill and 50 w/stage bicycle) and smallest for ing patients than the traditional, single-protocol the ramp tests. These observations suggest that: approaches. (1) oxygen uptake is overestimated from tests that contain large increments in work and (2) the vari- WALKING TESTS ability in estimating oxygen uptake from work rate is markedly greater on these tests than for Guyatt et al45 point out that bike and treadmill an individualized ramp test. exercise tests can be difficult for many patients with heart failure and may not reflect their It is also interesting how oxygen uptake kinet- capacity to undertake day-to-day activities as ics were influenced by the presence of disease. accurately as a walking test will. Walking tests The oxygen uptake slopes were generally have proven useful as measures of outcomes for steeper (closer to unity) among normal subjects patients with chronic heart failure and pulmonary regardless of the protocol used (Table 2-5). disease. They can be simply done in a hospital Patients with heart disease had reduced slopes or clinic by determining how far a patient compared with normal subjects, confirming walks in 6 minutes over a measured corridor (the previous investigations. However, we observed a “6-minute walk test”). To investigate the potential pronounced improvement in the slope of the value of the 6-minute walk as an objective mea- oxygen uptake/work rate relation in these sure of exercise capacity in patients with chronic patients when using an individualized protocol. heart failure, the test was administered six times In fact, the response of patients with heart failure over 12 weeks to 18 patients with chronic heart was similar to that of normal subjects when failure and 25 patients with chronic lung disease. both groups performed ramp treadmill tests The subjects also underwent bike testing, and (Fig. 2-3). their functional status was evaluated by conven- tional measures. The walking test proved highly Because this approach appears to offer several acceptable to the patients, and reproducible advantages, we now perform all our clinical and results were achieved after the first two walks. research testing using the ramp. However, this approach is empirical and more data from other laboratories are necessary to confirm its utility. TA B L E 2 - 5 . Slopes in oxygen uptake versus work rate for each patient subgroup performing each of six exercise protocols Slope CAD Angina CHF Normal SEE 0.51 0.53 0.53 0.71* 2.6 3.1 2.8 4.2 *P < 0.001 versus other groups. CAD, coronary artery disease; CHF, chronic heart failure; SEE, standard error of the estimate (mL O2/kg/min). If the change in ventilatory oxygen uptake were equal to the change in work rate, the slope would be 1.0.

24 E X E R C I S E A N D T H E H E A R T Measured oxygen uptake25 rate response up to a submaximal level.49 When (ml/kg/min) used clinically, one problem with submaximal Unity tests is that the most vulnerable patients can be stressed to a relatively greater extent, whereas Ramp–CHF the less impaired may be limited by submaximal target heart rates. A submaximal test is clinically 19 indicated for patients in the predischarge post- myocardial infarction or post-bypass surgery 13 Bruce–CHF period, and these tests have been shown to be important in risk stratification for such 7 patients.50,51 Submaximal tests in the early post- 7 13 19 25 myocardial infarction or postsurgery period Predicted oxygen uptake are also useful for making appropriate activity (ml/kg/min) recommendations, for modifying the medical regimen, and for identifying the need for further ■ FIGURE 2-3 interventions. A submaximal test is also an appro- Relation between the change in measured and predicted priate option for patients with a high probability oxygen uptake for the ramp and Bruce treadmill protocols of serious arrhythmias. among patients with chronic heart failure (CHF). The unity line is achieved when predicted oxygen uptake is equal to The testing endpoints for submaximal, predis- measured oxygen uptake. The regression equation was charge testing have traditionally been arbitrary, y = 0.80x + 2.0 for the ramp test and y = 0.54x + 3.8 but should always be based on clinical judgment. (p < 0.01 for slope) for the Bruce test. A heart rate limit of 140 beats/min and a MET level of 7 are often used for patients younger than The results correlated with the conventional mea- age 40 years, and limits of 130 beats/min and a sures of functional status and exercise capacity. The MET level of 5 are often used for patients older “6-minute” or other walking tests are now fre- than 40 years. For those using beta-blockers, a quently incorporated into pharmaceutical trials as Borg perceived exertion level in the range of 7 to an additional measure of efficacy among patients 8 (1-to-10 scale) or 15 to 16 (6-to-20 scale) are with heart disease. In patients with heart failure, conservative endpoints. The initial onset of the 6-minute walk test has been demonstrated by symptoms, including fatigue, shortness of breath, several groups to have prognostic value.46-48 or angina, is also an indication to stop the test. A However, as pointed out in Chapter 5, we and low-level protocol should be used, that is, one other investigators have observed that 6-minute that uses no more than 1-MET increments per walk performance correlates only modestly with stage. The Naughton protocol is commonly used exercise tolerance on the treadmill (see Table 5-1). for submaximal testing. Ramp testing is also ideal for this purpose because the ramp rate (such as SUBMAXIMAL EXERCISE TESTING 5 METs achieved over a 10-minute duration) can be individualized depending on the patient tested. Submaximal exercise testing is useful clinically for predischarge post-myocardial infarction eval- PERCEIVED EXERTION uations (see Chapter 9), and versions of submaxi- mal tests are commonly used for fitness screening Rather than use heart rate to clinically determine evaluations, such as in health clubs. Such tests the intensity of exercise, it is preferable to use are rather limited for the latter purpose as most of either the 6-to-20 Borg scale or the later, non- them use an extrapolation of the heart rate linear 1-to-10 scale of perceived exertion.52,53 The response to estimate fitness. A common submaxi- 6-to-20 scale was developed by noting that young mal cycle ergometer test is outlined by the YMCA, men could approximate their exercise heart rate if in which work is incremented based on the heart a scale ranging from 60 to 200 was aligned with labels ranging from very, very light for 60 to very, very hard for 200. The last digit was dropped, and the scale was used for all ages. Because sensory perception of pain or exertion is nonlinear, Borg then developed the 1-to-10 scale. Since the devel- opment of the Borg scales, many studies have

C H A P T E R 2 Exercise Testing Methodology 25 associated ratings of perceived exertion with compared to older designs that required direct physiologic responses to exercise, and these scales metal-to-skin contact. Many disposable electrodes have been widely used to estimate effort during have been shown to perform well for exercise test- exercise testing.53 ing. Silver plate or silver-silver chloride crystal pellets are the best electrode materials. Platinum SKIN PREPARATION is too expensive, and the frequently used German silver is actually an alloy. If electrodes of different Proper skin preparation is essential for the per- types of metals are used together, an offset voltage formance of an exercise test. Because noise can be generated that makes it impossible to increases with the square of resistance, it is record an ECG. The disposable electrodes have extremely important to lower the resistance at the advantages of quick application and no need the skin-electrode interface during exercise for cleansing for reuse. They are more expensive and thereby improve the signal-to-noise ratio. to use than the older nondisposable electrodes, It is often difficult to ensure that technicians but have replaced them completely. consistently prepare the skin properly, because doing so may cause the patient discomfort and Developing suitable connecting cables between minor skin irritation. Yet, the performance of an the electrodes and the recorder has been a exercise test with an ECG signal that cannot be problem in gathering exercise ECG data. The continuously monitored and accurately inter- earliest versions of these cables were subject to preted because of artifact is worthless and can wire-continuity problems, frequent failures, and even be dangerous. motion artifact; they were improperly shielded, they used inadequate connectors, and many did The general areas for electrode placement not perform well in metropolitan areas or near should be cleansed with an alcohol-saturated x-ray or other high-voltage equipment. Several gauze pad, then the exact areas for electrode commercial companies have concentrated on application may be marked with a felt-tip pen. solving these problems, and currently there are The mark serves as a guide for removal of the exercise cables available that are constructed to superficial layer of skin. The electrodes are placed avoid these problems. Buffer amplifiers carried by using anatomic landmarks that are best found the patient are no longer advantageous. Cables with the patient supine. Some individuals with develop continuity problems over time with use loose skin can have a considerable shift of elec- and require replacement rather than repair. trode positions when they assume an upright We find replacement to often be necessary after position. The next step is to remove the super- roughly 500 tests. Some systems have used ficial layer of skin using light abrasion with fine- analog-to-digital converters carried by the patient grain emery paper designed for this purpose. Skin in the electrode junction box. Digital signals are resistance should be reduced to 5000 Ω or less, relatively impervious to noise, so the patient cable which can be verified prior to the exercise test can be unshielded and is very light. Several with an inexpensive AC-impedance meter driven companies have developed systems that transmit at 10 Hz. A DC meter should not be used, since it the ECG signal in wireless format, obviating the can polarize the electrodes. Each electrode can need for cables. be tested against a common electrode with an ohmmeter, and when 5000 Ω or less is not LEAD SYSTEMS achieved, the electrode must be removed and skin preparation repeated. Clever devices have been Electrodes have been placed in a variety of ways designed for this purpose. This maneuver saves using many different lead systems. This has com- time by obviating the need to interrupt a test plicated making comparisons of the ST-segment because of noisy tracings. response to exercise. The four major exercise ECG lead systems are the bipolar, the Mason- ELECTRODES AND CABLES Likar 12-lead,54 a simulation of Wilson’s central terminal, and the three-dimensional (either The only suitable electrodes are constructed with orthogonal or nonorthogonal systems). a metal interface, sunken to create a column that can be filled with either an electrolyte solu- Bipolar lead systems have been used because tion or a saturated sponge. These fluid-column of the relatively short time required for place- electrodes markedly decrease motion artifact as ment, the relative freedom from motion artifact, and the ease with which noise problems can be located. Figure 2-4 illustrates the electrode

26 E X E R C I S E A N D T H E H E A R T This difference in the leads most likely explains why investigators using CM5 have reported an inadequate ST slope to be as serious as horizontal H depression. Specific ST-segment responses rela- tive to test sensitivity and different lead systems are presented in more detail in Chapter 6. S M Vector Leads B A number of three-dimensional or vectorcar- R diographic (VCG) lead systems can be used during exercise. The corrected Frank-lead system XC has the advantage that the electrical activity of the heart is orthogonally represented in the three ■ FIGURE 2-4 derived signals. The relative ease of placement of its electrodes and the fact that there are only The common bipolar ECG leads used during exercise testing. seven of them have made the Frank system the most popular orthogonal lead system. Care placements for most of the bipolar lead systems. should be taken so that the X and Z electrodes The usual positive reference is an electrode placed are placed as described by Frank in his original the same as the positive reference for V5. The paper, at the fifth intercostal space level and the negative reference for V5 is Wilson’s central termi- sternum. The VCG approach makes it possible to nal, which consists of connecting the limb evaluate the spatial changes of the ST-segment electrodes—right arm (RA), left arm (LA), and left vector. The Frank X is a left precordial lead but is leg (LL). The only other no-table bipolar lead about 25% smaller in amplitude than V5 because system is the roving bipolar lead, which was intro- of the Frank network resistance, which is an duced by McHenry. In this system, beginning with attempt to electrically move the heart to the cen- a CC5 placement, the electrodes are moved ter of the chest. However, ST-segment criteria around to obtain the maximal R wave with a small have not been adjusted for this. In both the S wave. 12-lead and Frank systems, several electrodes can be shared. V4 and V6 are I and A, respectively, LF The problem with comparing the results of can also be F, and in this way 14 electrodes can be ST-segment analysis if different leads are used used to obtain both systems. has been demonstrated by a computer analysis study.55 ST-segment depression and slope The Dalhousie square is a simple way to assist measurements were made on signals gathered with the proper and reproducible placement of simultaneously from CC5, CM5, and V5. A common the Frank electrodes and of the Wilson precordial positive reference electrode was used. CM5 consis- electrodes.56 It is a simple right-angle device that tently had a more negative J-junction and a is held to the chest. Proper placement is necessary more positive slope than did V5 and CC5, whereas for the application of ECG/VCG interpretive crite- V5 and CC5 were essentially identical on the ria. Reproducible placement is essential to assess basis of standard analysis but differed statistically serial changes. when computer measurements were compared. Mason–Likar Electrode Placement Because a 12-lead ECG could not be obtained accurately during exercise with electrodes placed on the wrists and ankles, in 1966 Mason and Likar54 first suggested that adhesive electrodes be placed in the base of the limbs for exercise testing. In addition to providing a noise–free exercise tracing, their modified placement apparently showed no differences in ECG configuration as compared with the standard limb lead placement.

C H A P T E R 2 Exercise Testing Methodology 27 However, others have found that the Mason–Likar negative reference to be a combination of three placement causes amplitude changes and axis additional electrodes rather than the single elec- shifts when compared with standard placement, trode used as the negative reference for bipolar which could lead to diagnostic changes. leads. Simulation of Wilson’s central terminal by Consequently, it has been recommended that the other combinations of electrodes has not been modified exercise electrode placement not be validated, and therefore such alternate configura- used for recording a resting ECG. The pre- tions should be avoided. exercise test ECG has been further complicated by the recommendation that it be obtained The University of California, standing, as that is the same position maintained San Diego (UCSD) Electrode during exercise. This recommendation is made Placement Study more difficult to follow by the common practice of moving the limb electrodes onto the chest It is clinically important to obtain an accurate in order to minimize motion artifact. pre-exercise ECG, which should be compared with previous tracings in order to determine if Figure 2-5 illustrates the Mason–Likar torso- any changes have occurred, and because it is mounted limb lead system. The conventional needed as a baseline for testing. We hypothesized ankle and wrist electrodes are replaced by that much of the confusion regarding distortion electrodes mounted on the torso at the base of the of the pre-exercise ECG was due to: (1) misplace- limbs. In this way, the artifact introduced by ment of limb electrodes medially on the torso and movement of the limbs is avoided. The standard (2) obtaining the ECG in the standing position. precordial leads use Wilson’s central terminal as Therefore, we compared 12-lead ECGs utilizing their negative reference, which is formed by the standard limb placement (electrodes on wrists connecting the right arm, left arm, and left leg. and ankles) against two modified exercise place- This triangular configuration around the heart ments in the supine and standing positions in the results in a zero voltage reference through the same patients.57 cardiac cycle. The use of Wilson’s central terminal for the precordial leads (V leads) requires the Right and left arm electrodes should be V4 should be placed in the fifth intercostal placed far laterally on the font edge of space on the midclavicular line the shoulders (anterior deltoids) LA RA V1 and V2 are Anterior V3 lies halfway positioned in the axillary line between V2 and V4 fourth intercostal space on the right V1 V2 and left border of V3 V4 the sternum V5 V6 V4, V5, and V6 should be placed along a horizontal line; this line does not follow the curve of the intercostal space RL Midaxillary line The right leg electrode (green, LL The left leg electrode (red, ground lead) should be placed on bottom of Einthoven’s the lower back over the back bone triangle) should be placed just below the umbilicus ■ FIGURE 2-5 The Mason–Likar-simulated standard 12-lead ECG electrode placement for exercise testing.

28 E X E R C I S E A N D T H E H E A R T Prior to exercise testing, 104 male patients USD exercise ** ** UCSD exercise with stable coronary heart disease were studied. Misplace Included were 30 men with ECG criteria for an (Located on back) * inferior myocardial infarction, 13 with an ante- Ground for all * Umbilicus rior myocardial infarction, five with diagnostic except standard Q waves in multiple locations, six with right bun- limb placement * Left leg for all dle branch block (three with diagnostic Q waves), except standard 33 with other abnormalities, and 17 with normal Limb ECGs. Just prior to a treadmill test, each patient * limb placement had 12-lead ECGs recorded with lead placements as illustrated in Figure 2-6. The four electrode Limb placements were as follows: * ■ Placement 1—the standard limb lead electrode placement on the wrists and ankles; supine Ground for standard * * Left leg for standard (“standard”). limb placement only limb placement only ■ Placement 2—arm electrodes placed medially ■ FIGURE 2-6 on the torso, 2 cm below the midpoint of the Electrode placement of the University of California, clavicle and leg electrodes below the umbilicus San Diego (UCSD) study of the effects of limb lead (“misplaced”). placement and standing on the routine ECG. ■ Placement 3—the “correct” Mason–Likar reveal any “serial” changes in the other three placement with the arm electrodes placed at tracings. the base of the shoulders against the deltoid border 2 cm below the clavicle, and the leg Differences between the standard limb lead electrodes the same as in placement 2 with ECG and the other ECGs were grouped into three the patient supine (“exercise supine”). categories: (1) diagnostic changes, (2) clinically important changes, and (3) other obvious ■ Placement 4—the same as placement 3 except changes (Table 2-6). The category “diagnostic that the patient is standing (“exercise standing”). changes” contained tracings with waveforms that had changed when the lead placement was In addition, the Frank X, Y, Z leads were recorded altered or the patient stood in such a way that the at the same time as the exercise-supine and diagnosis was different than for the supine limb exercise-standing ECGs. lead ECG. In all cases except one, the change in diagnosis was either the loss or appearance of The tracings were read by two blinded an inferior infarction. The one exception was a observers looking for definite diagnostic standing tracing showing that the change in changes that might be clinically important position had caused an anterior infarct pattern (including “new” Q waves in aVL or III) and other obvious changes. Q waves were considered diagnostic if they were 25% or greater of the following R-wave amplitude and 40 msec or longer in duration. Visual analyses of each of the 104 patients’ four ECGs were performed inde- pendently and by consensus of two observers. The tracings were interpreted separately and then compared with the standard limb lead ECG to TA B L E 2 - 6 . Differences noted by standard visual interpretation between exercise test electrode placements and standard supine electrocardiogram Diagnostic Changes Misplaced Exercise-standing Exercise-supine Important Changes Other Obvious Changes 6 12 3 Total Changes 19 12 7 0 3 6 10 28 30

C H A P T E R 2 Exercise Testing Methodology 29 I II III AVR AVL AVF V1 V2 V3 V4 V5 V6 “Misplaced” Standing Loss of Gain of inferior Q waves Loss of Q wave R wave Q wave extension Exercise supine Wrist/ankles I II III AVR AVL AVF “Misplaced” “Misplaced” New Q wave Standing Standing Exercise supine Q wave inferiority Exercise supine Wrist/ankles Wrist/ankles ■ FIGURE 2-7 Examples of the artifact seen in the pre-exercise test 12-lead ECG study. to disappear. The “exercise standing” placement depression, and one instance of a Q wave appear- had a total of 12 diagnostic changes: seven where a new diagnosis of inferior infarction was made, ing in V6. The “misplaced” placement had various four where an inferior infarction diagnosis was important changes: eight where a new Q wave lost, and the previously mentioned exception of losing an anterior infarction diagnosis. The “mis- appeared in aVL; seven where a Q wave disap- placed” electrode placement had six diagnostic peared in only III; three where a Q wave appeared changes: one where the criterion for an inferior infarction was reached and five where it was lost. in III; two ST-segment or T-wave changes; The “exercise supine” placement had three diag- nostic changes, all showing a loss of the criteria and one where a Q wave appeared in aVL. The for an inferior infarction compared with the “exercise supine” placement had seven important standard ECG. changes: four where a Q wave disappeared in III, The category “important changes” consisted of changes that might be clinically important but one where it appeared, and two where a Q wave did not of themselves alter the ECG diagnosis. Such changes included significant Q waves in appeared in aVL. Figure 2-7 illustrates the III or aVL alone, ST-segment and T-wave changes changes seen in three patients; in the bottom such as flipped or flattened T waves or ST two tracings, changes occurred only in the limb leads. QRS frontal axis means, standard deviations, and differences analyzed by computer for the four electrode placements are given in Table 2-7. When compared with the standard electrode placement, “misplaced” showed an average of 26 degrees of

30 E X E R C I S E A N D T H E H E A R T TA B L E 2 - 7 . Computer analysis of QRS axis in frontal plane as mean values, standard deviation, mean of differences, and standard deviation of difference Standard Supine-exercise Misplaced Standing-exercise Mean 18 27 44 18 Standard Deviation 43 48 48 53 Mean Difference 9 26 −3 Standard Deviation of Differences 36 41 48 Significant Difference NS P < 0.01 NS deviation to the right (p < 0.01), “supine exercise” maintained during exercise. Other studies have showed an average of 9 degrees of rightward also shown differences between ECGs taken deviation (not significant), and “standing exer- supine versus sitting or standing. Sigler61 studied cise” showed an average of 3 degrees of deviation 100 patients and found a tendency for the QRS to the left (not significant). “Standing exercise” axis to shift to the left for abnormal tracings and showed the greatest amount of variability, with to the right for normal tracings when the patient a standard deviation of 53 degrees. changed from a supine to standing position. Dougherty62 correlated changes in the frontal plane Other Electrode Placement QRS axis with changes in heart position measured Studies with a chest x-ray brought about by moving the patient from a supine to a standing position. He Kleiner et al58 compared ECGs gathered on found that every degree of positional heart change 75 patients using the standard wrist and ankle caused a 3-degree shift in QRS frontal axis through- placement against the Mason–Likar placement. out the normal range in the same direction. Fifty of the 75 patients had a rightward axis Bruce et al63 found a higher prevalence of false- shift of 30 degrees or more on the modified positive polar cardiographic criteria for myocardial ECG compared with the standard. In addition, infarction in 72 normal men and women. 11 of these patients had a rightward shift in axis on their modified ECG that resulted in Q- and Another complicating factor is the effect of T-wave inversion in lead aVL, without prior his- respiration on inferior Q waves. Because it has tory of myocardial infarction by ECG. Seventeen been suggested that inspiration caused Q waves in patients had diagnostic criteria for an old inferior lead III to diminish or disappear in normal sub- myocardial infarction, and seven of them (41%) jects and to persist in patients with inferior had these criteria erased by the rightward axis myocardial infarctions, Mimbs et al64 studied shift on the modified placement. These investiga- the effect of respiration on the Q wave in lead III tors cautioned that the modified exercise place- in normal subjects and in patients with docu- ment of Mason and Likar should not be mented inferior myocardial infarctions. They considered interchangeable with the standard found that the Q wave in III decreased on inspira- electrode placement. However, it was not stated tion in 82% of patients with a recent myocardial where the shoulder electrodes were placed or infarction and in 44% of patients with an old if the modified ECG was recorded supine or infarction. Among eight normal subjects with a Q standing. Rautaharju et al59 came to a similar wave in III, only one showed a decrease in ampli- conclusion as Kleiner et al.58 tude; others showed no change. The researchers concluded that the effect of inspiration on the The pre-exercise ECG is further complicated ECG was variable, although they did not report by positional differences. Shapiro et al60 studied specific amplitude changes. the differences between supine and sitting Frank- lead VCGs in 59 adult male patients with sus- As part of a detailed study of 194 patients, pected CAD. They observed that QRS spatial and Reikkinen and Rautaharju65 analyzed the effects R-wave amplitudes in lead Z were significantly of respiration and sitting on the VCG. Because of higher and R-wave amplitudes in lead Y lower for the great variability of the changes, though rarely sitting than for supine positions. They concluded with any significant differences in the means, they that the pre-exercise ECG should be obtained reported the percentage of patients who had with the patient in the same position as that increases or decreases in specific values beyond an arbitrary threshold. The changes with deep inspiration were much more prominent than

C H A P T E R 2 Exercise Testing Methodology 31 those with deep expiration. With inspiration, 35% the exercise electrode placement can be kept to of the patients had a posterior shift in the hori- a minimum by: (1) keeping the arm electrodes zontal plane, whereas 8% had an anterior shift; off the patient’s chest and putting them on the 45% had a rightward shift in the frontal plane, shoulder (anterior deltoid) instead and (2) obtain- whereas 3% had a leftward shift. R-wave ampli- ing the pretest, resting ECG with the patient tude in lead Z decreased a mean of 0.5 mV, Q-wave supine. However, the standing ECG using the amplitude decreased, and the QRS-T angle modified exercise limb lead placement of Mason increased. During sitting, R-wave amplitude and Likar can serve as the reference ECG prior to increased in Z, Q-wave amplitude decreased in Y, an exercise test. and the mean QRS maximal spatial magnitude increased. Sitting also caused 26% of patients to Relative Sensitivity of Leads have a posterior shift in the horizontal plane, whereas 10% had an anterior shift; 15% had a Numerous studies comparing the relative sensi- rightward shift in the frontal plane, whereas 10% tivity of different ECG leads were reported in the had a leftward shift, and 36% of patients had an 1970s. Robertson et al66 compared the perform- increase in the QRS-T angle. ance of different ECG leads among 39 patients with both an abnormal exercise test and an abnor- The results of the UCSD lead study clarify mal coronary angiogram. Eighteen percent had much of the confusion regarding the pre-exercise an abnormal response in leads other than V5. test ECGs.57 Misplacement of the Mason–Likar Patients with right coronary artery lesions usually arm leads is common; published articles have showed ST-segment depression in the inferior even described it as the correct exercise modi- leads, and patients with left coronary system fication. By placing the leads medially, near the lesions usually showed ST-segment depression in midclavicular line, we found that the frontal plane leads I and aVL and in the chest leads. Almost a axis shifts rightward an average of 26 degrees. third of the patients showed ST-segment depres- This shift caused decreased amplitudes in the sion in leads other than those anticipated from Q wave in III and of the R wave in I and aVL, and their angiographic anatomy. Tucker et al67 per- it caused increased amplitudes in the Q wave in formed 12-lead exercise tests in 100 consecutive aVL and the R wave in II, III, and aVF. Of clinical patients who were also studied with coronary importance is what these shifts did to the visual angiography. Forty-eight had abnormal tests, interpretation of the ECG. In five patients the with 30% of the abnormal responses occurring in ECG diagnosis of old inferior infarct was lost. In leads other than V5 (17% in aVF and 13% in other addition, seven patients lost significant Q waves leads). Two false positives occurred in V5 and two in III alone. There were also instances of Q waves in aVF, whereas 16 positives occurred in leads gained. Eight patients had no Q waves in aVL, one other than V5 or aVF. Of those abnormal in aVF had a new Q wave in III and II, one had a new alone, five had lesions in the right coronary or Q wave in V6, and one gained an inferior infarct left circumflex artery, and two had disease in the diagnosis. Although these “serial” changes are left anterior descending artery. merely artifacts produced by electrode mis- placement, they could be very misleading. The Chaitman et al68 evaluated the role of multiple- changes, reported to be caused by misplaced lead ECG systems and clinical subsets in inter- electrode arrangement in the UCSD study,57 were preting treadmill test results. Two hundred men very similar to those reported by Kleiner et al.58 with normal ECGs at rest had a maximal tread- mill test using 14 ECG leads and then underwent Standing can cause many changes in the visual coronary angiography. This study included interpretation of the ECG, including those that standard leads plus three bipolar leads. The preva- would be most alarming—that is, appearance of lence of significant coronary stenosis was 86% in new Q waves (particularly inferiorly). The mis- 87 men with typical angina, 65% in 64 men with placement of the arm electrodes in the midclavic- probable angina, and 28% in 49 men with non- ular area also causes many clinical changes, specific chest pain. The predictive value of including the appearance of Q waves in aVL and ST-segment deviation in any one of 14 leads was large axis shifts and amplitude changes. The cor- 45% in men with nonspecific chest pain versus rect Mason and Likar modification can likewise 70% in men with probable angina, and 55% in cause amplitude and duration changes, but men with typical angina. In the latter, recording they are less clinically important. The modified a single lead such as CM5 was adequate. In men exercise electrode placement should not be used for routine resting ECGs. The changes caused by

32 E X E R C I S E A N D T H E H E A R T with typical or probable angina, a normal 15 normal volunteers during supine exercise. response in 14 leads associated with treadmill Analyses of the exercise isopotential maps exercise time longer than 9 minutes reduced the revealed a minimum amplitude during the early chance of three-vessel disease to less than 10%. portions of the ST segment located below the The likelihood of multivessel disease in a patient standard V3 and V4 chest positions, with negative with an abnormal ST response and a treadmill potentials involving most of the precordial time equal to or less than 3 minutes was approxi- region. Isopotential “difference” maps were con- mately 90%. In patients with angina, the use of structed by subtracting potentials at the begin- 14 leads increased sensitivity over that of V5 alone ning of the ST segment from potentials later in from 52% to 75%. This value was increased even the ST segment. These maps characterized the further, to 86%, by the additional consideration direction and magnitude of ST-segment slopes of bipolar leads. and revealed up-sloping ST segments over regions of negative ST potentials. Miranda et al69 studied 178 males who had undergone both exercise testing and coronary Miller et al71 obtained total thoracic surface angiography to determine the diagnostic value of exercise maps in 20 normal subjects, recording ST-segment depression occurring in the inferior from 24 electrode sites and deriving the remaining leads. Lead V5 had a better sensitivity (65%) and potentials at 150 locations using previously devel- specificity (84%) than that of lead II (sensitivity oped mathematical transformations. Isopotential and specificity being 71% and 44%, respectively) maps during the early ST segment were less nega- at a single cut point. Receiver operating charac- tive than those described by Mirvis et al70 primarily teristic curve analysis demonstrated that lead V5 because of Miller’s “zone” reference potential. The (area = 0.759) was markedly superior to lead II end of the PR segment was chosen, in contrast to (area = 0.582) over multiple cut points. Moreover, the use of the ST segment by Mirvis et al70 (which the area under the curve in lead II was not signif- shortens with heart rate increases). Exercise isopo- icantly greater than 0.50, suggesting that for tential maps in normal subjects were characterized the identification of CAD, ST-segment depression by a left anterior maximum during ST-T. isolated to lead II is unreliable. Studies in Patients with Coronary It remains to be demonstrated precisely what Artery Disease the specificity of leads other than V5 is, but one has the impression that inferior leads produce Fox et al72 from London used very simple exercise more false positives and may require different ECG mapping techniques to detect myocardial criteria. This apparent lack of specificity may be due ischemia. Using a 16-lead precordial map and to the effect of atrial repolarization in the inferior visual interpretation of the ECG data, they drew leads, which causes depression of the ST segment. contour maps for each patient illustrating regions With adequate experience, atrial repolarization on the precordial surface where significant can be recognized as causing ST-segment depres- ST-segment depression was observed. The first sion. The end of the PR segment can be seen to be of these studies involved 100 patients undergoing depressed in a curved fashion to the same level coronary angiography for evaluation of chest that the ST segment begins. Such findings also pain. The sensitivity of the precordial mapping support the concept of intercoronary artery steal technique (96%) for diagnosing coronary disease during exercise; that is, ischemic areas obtain was better than the modified 12-leads (80%), blood flow through collaterals. This phenomenon using 0.1 mV horizontal ST-segment depression makes it impossible for ST-segment depression as the criterion for abnormal. The higher sensi- with multiple-lead exercise testing to enable tivity using mapping was due to the improved prediction of the location of coronary artery recognition of single-vessel disease. Also of inter- occlusions. est was the regional localization of the ischemic ST contours in single-vessel disease. ST-segment BODY SURFACE MAPPING depression involving the uppermost horizontal row of electrodes was highly sensitive for Studies in Normal Subjects proximal left anterior descending or left main CAD. This result was obtained without a loss in The normal repolarization response to exercise specificity (90%). using large electrode arrays has been described by Mirvis et al70 and by Miller et al.71 Mirvis et al70 In a second study involving 200 patients used a 42-electrode left precordial lead system in undergoing coronary angiography, Fox et al73

C H A P T E R 2 Exercise Testing Methodology 33 again compared the 12-lead ECG to a 16-lead sites on the body surface. Since currents of injury map and found that the standard precordial leads primarily occur at the boundaries between nor- sampled only 41% of the ST-segment depression mal and abnormal tissue, cancellation of forces projected to the front of the chest. However, in will likely distort the relationships between body only 7% of patients was the ST-segment depres- surface ST-segment changes and the degree of sion not apparent in the standard precordial ischemia. The subendocardial and nontransmural leads. The rightmost column of electrodes never locations of most exercise-induced ischemia make recorded ST-segment depression that was not it unreasonable to expect that body surface ECG seen in one or more of the remaining 12 elec- recordings will reflect the extent, magnitude, and trodes on the precordial surface. The researchers location of the ischemic tissues. concluded that these 12 precordial leads, along with the standard limb leads, would optimize the Experimental confirmation of these concerns detection of ST-segment changes. has been provided by Mirvis and Ramanathan.76 Body surface maps were obtained in dogs with Yanowitz et al74 at the University of Utah have previously placed amaroid constrictors around reported using a 32-lead electrode array to derive one of the three major coronary arteries. Atrial torso potential distributions at 192 locations by pacing induced reversible myocardial ischemia. means of a mathematical transformation. These Although the location of the ischemic repolariza- investigators evaluated this system during exer- tion abnormalities on the maps varied with the cise testing in 25 patients with documented CAD. particular artery involved, significant spatial The distribution of 80-msec ST-segment isoarea overlap was observed so as to preclude any identi- contours (ST80 isoarea maps) was plotted and fication of discrete ischemic zones unique to a compared with the standard precordial leads. It given arterial lesion. These findings, plus consid- was found that in 25% of patients with ischemic eration of the added cost of specialized recorders ECG changes the maximal ST change was located and more electrodes, leave mapping as a research at sites distant to the standard leads. In addition, tool without much clinical applicability. there was some evidence of localization in patients with single-vessel disease. NUMBER OF LEADS TO RECORD Simoons and Block75 recorded exercise body In patients with normal resting ECGs, V5 or a surface maps in 25 normal subjects and 25 patients similar bipolar lead along the long axis of the with coronary disease using a system of 120 tho- heart usually is adequate. In patients with ECG racic surface electrodes. Evaluation of normal evidence of myocardial damage or with a history subjects revealed a low-level (<90 μV) precordial suggestive of coronary spasm, additional leads are minimum during early ST, followed by the devel- needed. As a minimal approach for these patients, opment of a prominent maximum later in the it is advisable to record three leads: a V5-type lead, ST-T wave, similar to the observations of Miller an anterior V2-type lead, and an inferior lead such et al.71 In the coronary patients, exercise maps as aVF. Alternatively, the Frank X, Y, and Z leads frequently showed a prolonged negative area in may be used. The continuous monitoring of at the precordium, with varying locations of the least three leads is also helpful for the detection minimum. There was no relation between the and identification of dysrhythmias. It is advisable specific ST isopotential distributions and either to record a second three-lead grouping consisting the coronary anatomy or the location of thallium of V4, V5, and V6. Occasionally, abnormalities that scan defects. However, the maps were more sensi- are seen as borderline in V5 are clearly abnormal in tive (84%) in detecting abnormal repolarization V4 or V6. This issue is less of a concern today patterns in coronary patients than the 12-lead because the medical electronics industry has exercise ECG (60%), using an ST-segment mini- made 12 leads the standard available in all mum of 90 μV at 60 msec after the J-point as the machines. criterion for an abnormal map. None of the normal subjects had negative ST potentials of this POSTEXERCISE PERIOD magnitude (specificity of 100%). If maximal sensitivity is to be achieved with an It is simplistic to consider ST-segment map- exercise test, patients should be supine during ping data as having the ability to directly quanti- the postexercise period. It is advisable to record tate ischemic myocardium. The physiologic about 10 seconds of ECG data while the patient is mechanisms responsible for ST-segment (and TQ-segment) shifts in ischemic injury are com- plex and depend upon the shape and location of the ischemic region in relation to the electrode

34 E X E R C I S E A N D T H E H E A R T standing motionless but still experiencing near if good clinical judgment is used. Absolute indica- maximal heart rate, and then have the patient lie tions include a drop in systolic blood pressure down. Some patients must be allowed to lie down despite an increase in workload, anginal chest immediately to avoid hypotension. Having the pain becoming worse than usual, central nervous patient perform a cool-down walk after the test system symptoms, signs of poor perfusion (such can delay or eliminate the appearance of ST- as pallor, cyanosis, and cold skin), serious dysrhyth- segment depression.77 According to the law of mias, technical problems with monitoring the Laplace, the increase in venous return and thus patient, patient’s request to stop, and marked ECG ventricular volume in the supine position changes such as 0.2-mV or greater ST-segment increases myocardial oxygen demand. Data from elevation. Relative indications for termination our laboratory78 and others suggest that having include other worrisome ST or QRS changes such patients lie down as immediately as possible as marked ST depression (e.g., = 2.0 mm hori- enhances ST-segment abnormalities in recovery. zontal or downsloping), increasing chest pain, Therefore, the exercise testing guidelines have fatigue, shortness of breath, wheezing, leg cramps suggested that having the patient in the supine or intermittent claudication, worrisome appear- position in recovery is preferable to a cool-down ance, a hypertensive response (systolic pressure walk. However, a cool-down walk has been sug- greater than 250 mmHg, diastolic pressure gested by some researchers in order to minimize greater than 115 mmHg), and less serious the postexercise chances for arrhythmic events in dysrhythmias including supraventricular tachy- this high-risk time when catecholamines are ele- cardias. In some patients estimated to be at high vated. The supine position after exercise is less risk by their clinical history, it may be appropriate important when the test is not being performed to stop at a submaximal level, since the most for diagnostic purposes. When testing is not severe ST-segment depression or dysrhythmias performed for diagnostic purposes, it may be may occur only after exercise. If more informa- preferable to have the patient walk slowly (1.0 to tion is required, the test can be repeated later. 1.5 mph) or continue cycling against zero or minimal resistance (0 to 25 W when testing with EXERCISE TEST ANCILLARY a cycle ergometer) for several minutes following TECHNIQUES the test. Several “add on” or ancillary imaging techniques Monitoring should continue for at least 6 to have been shown to provide a valuable comple- 8 minutes after exercise or until changes stabilize. ment to exercise electrocardiography for the An abnormal response occurring only in the evaluation of patients with known or suspected recovery period is not unusual, and these CAD. Some of these methods can localize responses are important to consider for optimiz- ischemia and thus help to guide interventions. ing the diagnostic performance of the test. All These techniques are particularly helpful among such responses are not false positives, as has been patients with equivocal exercise ECGs or those suggested. Experiments confirm mechanical likely to exhibit false-positive or false-negative dysfunction and electrophysiologic abnormalities responses. They are often added to confirm test in the ischemic ventricle following exercise. results in patients with more than 1.0-mm ST Although the supine position is preferred for depression at rest, LBBB, WPW, and paced diagnostic testing, a cool-down walk can be help- rhythms. They are also frequently used to clarify ful for patients with an established diagnosis who abnormal ST-segment responses in asymptomatic are undergoing testing other than for diagnostic individuals or those in whom the cause of chest reasons, as well as for athletes or patients with discomfort remains uncertain. These tests can clinically significant arrhythmias. be an important complement to the standard exercise test because they can either confirm the INDICATIONS FOR TREADMILL diagnosis of CAD or obviate the need for angio- TEST TERMINATION graphy or other interventions. They can also help direct a patient to an appropriate interven- The absolute and relative indications for termina- tion because ischemia can be localized. When an tion of an exercise test listed in Table 2-1 have exercise ECG and an imaging technique are been derived from many years of clinical experi- combined, the diagnostic and prognostic accu- ence. Absolute indications are clear-cut, whereas racy is enhanced. The major imaging procedures relative indications can sometimes be disregarded

C H A P T E R 2 Exercise Testing Methodology 35 are myocardial perfusion and ventricular function tomography) was possible and subtle differences studies using radionuclide techniques and could be plotted and scored. In recent years, exercise or pharmacologic echocardiography. ventriculograms based on the imaged wall as The nonexercise stress techniques (pharmaco- opposed to the blood in the chambers (as with logic stress testing using persantine or adenosine radionuclear ventriculography) could be con- with nuclear perfusion, dobutamine or arbuta- structed. Because of the technical limitations mine with echocardiography) permit diagnostic with thallium (i.e., source and lifespan), it has assessment of patients unable to walk on the largely been replaced by chemical compounds treadmill or pedal a cycle ergometer due to ortho- called isonitriles which could be tagged with tech- pedic or neurologic conditions. Although these netium, which has many practical advantages newer technologies are often suggested to have over thallium as an imaging agent. The isonitriles better diagnostic characteristics, this is not are trapped in the microcirculation, permitting always the case, particularly when multivariate imaging of the heart with a scintillation camera. scores are used with standard exercise testing. Rather than a single injection, as with thallium, these compounds require an injection at maximal Nuclear Techniques exercise, then later in recovery. The differences in technology over the years and the differences Nuclear Ventricular Function Assessment. One in expertise and software at different facilities of the first techniques added to exercise was can complicate the comparisons of the results radionuclear ventriculography. This involved the and actual application of this technology. The intravenous injection of technetium-tagged red ventriculograms obtained with gated perfusion blood cells. Using ECG gating of images obtained scans do not permit the assessment of valvular from a scintillation camera, images of the blood lesions or as accurate an assessment of wall circulating within the LV chamber could be motion abnormalities or ejection fraction as obtained. While regurgitant blood flow from echocardiography. valvular lesions could not be identified, ejection fraction and ventricular volumes could be esti- Vasodilators, such as dipyridamole and adeno- mated. The resting values could be compared to sine, are commonly used to assess coronary those obtained during supine exercise and criteria perfusion in conjunction with a nuclear imaging were established for abnormal. The most com- agent. Dipyridamole and adenosine cause maxi- mon criteria involved a drop in ejection fraction. mal coronary vasodilation in normal epicardial This procedure is now rarely performed because arteries, but not in stenotic segments. As a result, its test characteristics appear to be inferior to a coronary steal phenomenon occurs, with a newer technologies. While initially popular in relatively increased flow to normal arteries and a the 1970s and 1980s, the blood volume tech- relatively decreased flow to stenotic arteries. niques have come to be surpassed by perfusion Nuclear perfusion imaging under resting condi- techniques. tions is then compared with imaging obtained after coronary vasodilation. Interpretation is Nuclear Perfusion Imaging. The first agent used similar to that for exercise nuclear testing. for perfusion imaging was thallium, an isotopic analog of potassium that is taken up at variable Echocardiography. The impact of the echocar- rates by metabolically active tissue. When taken diogram on cardiology has been considerable. up at rest, images of metabolically active muscle, This imaging technique comes second only to such as the heart, are possible. With the nuclear contrast ventriculography via cardiac catheteriza- camera placed over the heart after intravenous tion for measuring ventricular volumes, wall injection of this isotope, images were initially motion, and ejection fraction. With doppler viewed using x-ray film. The normal complete added, regurgitant flows can be estimated as donut-shaped images gathered in multiple views well. Echocardiographers were quick to add this would be broken by “cold” spots where scar was imaging modality to exercise, with most studies present. Defects viewed after exercise could be showing that supine, postexercise assessments due to either scar or ischemia. Follow-up imaging were adequate and the more difficult imaging confirmed that the “cold” spots were due to ische- during exercise was not necessary. The patient mia if they filled in later. As computer imaging must be placed supine as soon as possible after techniques were developed, three-dimensional exercise and imaging begun. A problem can imaging (single photon emission computed occur when the imaging requires removal or displacement of the important V5 electrode,

36 E X E R C I S E A N D T H E H E A R T where most of the important ST changes are and appropriate emergency equipment are essen- observed. tial. A brief physical examination is always neces- sary to rule out significant aortic valve disease and Dobutamine is the most common pharmaco- other contraindications for the test. Pretest stan- logic agent used with echocardiography. dard 12-lead ECGs are needed in both the supine Dobutamine is a synthetic catecholamine that and standing positions. The changes caused by acts directly on β1 and β2 receptors. It increases exercise electrode placement can be kept to a mini- the force of myocardial contractility more than it mum by keeping the arm electrodes off the chest, increases heart rate. It is titrated intravenously placing them on the anterior deltoid instead, and by from low to higher doses while observing four recording the baseline ECG supine. When electrode views of the heart, similar to exercise echocardiog- preparation and placement are properly performed, raphy. Wall motion abnormalities are caused by the modified exercise limb lead placement of Mason the associated increased demand for myocardial and Likar can serve well as the reference resting oxygen and shortening of coronary artery filling ECG prior to an exercise test. time resulting from the increase in heart rate. If ischemia is induced, a wall motion abnormality Few studies have correctly evaluated the appears at a specific heart rate. relative yield or sensitivity and specificity of dif- ferent electrode placements for exercise-induced Biomarkers. The latest ancillary tool to supple- ST-segment shifts. Studies show that using other ment the exercise ECG in an attempt to improve leads in addition to V5 will increase the sensitivity; its diagnostic characteristics are biomarkers. however, the specificity is decreased. ST-segment The first and most logical biomarker evaluated changes isolated to the inferior leads can some- to detect ischemia brought out by exercise times, but not always, be false-positive responses. was troponin. However, it has been shown that VCG and body surface mapping lead systems do even in patients who develop ischemia during not offer any advantage over simpler approaches exercise testing, serum elevations in cardiac- for clinical purposes. specific troponin do not occur, demonstrating that myocardial damage does not occur.79,80 The exercise protocol should be progressive, B-type natriuretic peptide, which is released by with even increments in speed and grade when- myocardial stretching, also appears to be released ever possible. Small, even, and frequent work by myocardial hypoxia. Armed with this knowl- increments are preferable to large, uneven, and edge, investigators have reported several studies less frequent increases, because the former yield a suggesting an improvement in the diagnostic more accurate estimation of exercise capacity. characteristics of the exercise test with B-type Many investigators, rather than using the same natriuretic peptide and its isomers.81,82 The point protocol for every patient, have emphasized the of contact analysis techniques available for these value of individualizing the exercise protocol. assays involves a hand-held battery-powered unit The optimal test duration is 8 to 12 minutes, and that uses a replaceable cartridge. Finger stick the protocol workloads should be adjusted to blood samples are adequate for analyses and the permit this duration. Because ramp testing results are available immediately. If validated uses small increments, it permits a more accurate using an appropriate study design (similar to estimation of exercise capacity and can be QUEXTA), biomarker measurements could individualized for every patient to yield a targeted greatly improve the diagnostic characteristics of test duration. For equipment that does not have a the standard office/clinic exercise test. controller that performs such tests, the workload can be increased manually in equal increments SUMMARY using any treadmill. Proper methodology is critical to patient safety Target heart rates based on age should not and acquiring accurate results from the exercise be used, because the relationship between maximal test. Preparing the patient physically and emo- heart rate and age is poor and has a wide tionally for testing is necessary. Good skin pre- scatter around many different recommended paration must cause some discomfort but is regression lines. Such heart rate targets result necessary for good conductance and to avoid in a submaximal test for some individuals, a artifact. Specific criteria for exclusion and maximal test for some, and an unrealistic goal termination, physician interaction with the patient, for others. The Borg scales are an excellent means of quantifying an individual’s effort. Exercise capacity should not be reported in total time but rather as the VO2 or MET equivalent of the

C H A P T E R 2 Exercise Testing Methodology 37 workload achieved. This practice permits the com- circumstances in the exercise testing lab should parison of the results of many different exercise be considered: testing protocols. Hyperventilation should be avoided prior to testing. Subjects both with and ■ Testing patients with aortic valvular disease without disease may or may not exhibit ST-segment should be done with great care. Consequently, changes with hyperventilation; the value of this a physical examination, including assessment procedure in lessening the number of false-positive of systolic murmurs, should be done before all responses is no longer considered useful by most exercise tests. If a significant murmur is heard, researchers. The postexercise period is a critical an echocardiogram should be considered. period diagnostically, and whenever possible the patient should be placed in the supine position ■ ST-segment elevation without diagnostic immediately after testing for this reason. Q waves (which is due to transmural ischemia) can be associated with dangerous arrhythmias Key points worth emphasizing include the and infarction; it occurs in about 1 out of following: 1000 clinical tests. ■ The treadmill protocol should be adjusted to ■ When a patient with an ischemic cardiomyop- the patient; one protocol is not appropriate for athy exhibits severe chest pain owing to all patients. ischemia (angina pectoris), a cool-down walk is advisable, since the ischemia can worsen in ■ Report exercise capacity in METs, not minutes recovery. of exercise. ■ A potentially dangerous situation exists when ■ Hyperventilation prior to testing is not a patient develops exertional hypotension indicated. accompanied by ischemia (angina or ST depression) or when it occurs in a patient with ■ ST measurements should be made at ST0 a history of chronic heart failure, cardiomyop- (J-junction), and ST depression should be athy, or recent myocardial infarction. considered abnormal only if horizontal or downsloping. ■ When a patient with a history of sudden collapse develops frequent premature ventric- ■ Patients should be placed supine as soon as ular contractions, a cool-down walk is advis- possible after exercise. For diagnostic testing, a able (premature ventricular contractions can cool-down walk should be avoided in order for increase during recovery, particularly after an the test to have its greatest value. abrupt cessation of exercise). ■ It is critical to include the 2- to 4-minute Appreciation of these circumstances can help recovery period in analysis of the ST response. avoid any complications in the exercise lab. ■ Measurement of systolic blood pressure during REFERENCES exercise is extremely important, and exertional hypotension is ominous; at this point, only 1. Rochmis P, Blackburn H: Exercise tests: A survey of procedures, manual blood pressure measurement tech- safety, and litigation experience in approximately 170,000 tests. niques are valid. JAMA 1971;217:1061-1066. ■ Age-predicted heart rate targets are largely 2. Gibbons L, Blair SN, Kohl HW, Cooper K: The safety of maximal useless because of the wide scatter for any age; exercise testing. Circulation 1989;80:846-852. a relatively low heart rate can be maximal for a patient of a given age and submaximal for 3. Myers J, Voodi L, Umanu T, Froelicher VF: A survey of exercise test- another. Thus, a test should not be considered ing: Methods, utilization, and safety in the VAHCS. J Cardiopulm nondiagnostic if a percentage of age-predicted Rehab 2000;20:251-258. maximal heart rate (i.e., 85%) is not reached. 4. Yang JC, Wesley RC, Froelicher VF: Ventricular tachycardia during ■ The Duke treadmill score should be calculated routine treadmill testing. Arch Intern Med 1991;151:349-353. automatically on every test. 5. Irving JB, Bruce RA: Exertional hypotension and post exertional ■ Other predictive equations should be consid- ventricular fibrillation in stress testing. Am J Cardiol 1977;39: ered as part of the treadmill report. 849-851. To ensure the safety of exercise testing and 6. Shepard RJ: Do risks of exercise justify costly caution? Phys Sports reassure the noncardiologist performing the test, Med 1977;5:52-58. the following list of the most dangerous 7. Cobb LA, Weaver WD: Exercise: A risk for sudden death in patients with coronary heart disease. J Am Coll Cardiol 1986;7:215-219. 8. Rodgers GP, Ayanian JZ, Balady GJ, Beasley JW, Brown KA, Gervino EV, Parison S, Quinnones M, Schlant RC: American College of Cardiology/American Heart Association clinical competance statement on stress testing. Circulation 2000;102:1726-1738. 9. American College of Sports Medicine: Guidelines for Exercise Testing and Prescription, 6th ed. Baltimore, Lippincott, Williams & Wilkins, 2000.

38 E X E R C I S E A N D T H E H E A R T 10. Astrand P, Ekblom B, Messin R, et al: Intraarterial blood pressure 36. Bruce RA: Exercise testing of patients with coronary heart disease. during exercise with different muscle groups. J Appl Physiol Ann Clin Res 1971;3:323-330. 1965;20:253-256. 37. Ellestad MH, Allen W, Wan MCK, Kemp G: Maximal treadmill 11. Bevegard S, Freyschuss U, Strandell T: Circulatory adaptation to stress testing for cardiovascular evaluation. Circulation arm and leg exercise in supine and sitting positions. J Appl Physiol 1969;39:517-522. 1966;21:37-46. 38. Stuart RJ, Ellestad MH: National survey of exercise stress testing 12. Bobbert AC: Physiological comparison of three types of ergometry. facilities. Chest 1980;77:94-97. J Appl Physiol 1960;15:1007-1014. 39. Sullivan M, McKirnan MD: Errors in predicting functional capacity 13. Davis JA, Vodak P, Wilmore JH, et al: Anaerobic threshold and for postmyocardial infarction patients using a modified Bruce maximal aerobic power for three modes of exercise. J Appl Physiol protocol. Am Heart J 1984;107:486-491. 1976;41:544-550. 40. Webster MWI, Sharpe DN: Exercise testing in angina pectoris: The 14. Asmussen E, Nielsen M: Regulation of body temperature during importance of protocol design in clinical trials. Am Heart J work performed with arms and legs. Acta Physiol Scand 1989;117:505-508. 1947;14:373-382. 41. Panza JA, Quyyumi AA, Diodati JG, et al: Prediction of the 15. Klefbeck B, Mattsson E, Weinberg J: The effect of trunk support on frequency and duration of ambulatory myocardial ischemia in performance during arm ergometry in patients with cervical cord patients with stable coronary artery disease by determination of injuries. Paraplegia 1996;34:167-172. the ischemic threshold from exercise testing: Importance of the exercise protocol. J Am Coll Cardiol 1991;17:657-663. 16. Astrand I, Guharay A, Wahren J: Circulatory responses to arm exercise with different arm positions. J Appl Physiol 1968;25: 42. Redwood DR, Rosing DR, Goldstein RE, et al: Importance of the 528-532. design of an exercise protocol in the evaluation of patients with angina pectoris. Circulation 1971;43:618-628. 17. Tuttle WW, Horvath SM: Comparison of effects of static and dynamic work on blood pressure and heart rate: J Appl Physiol 43. Smokler PE, MacAlpin RN, Alvaro A, Kattus AA: Reproducibility of 1957;10:294-296. a multi-stage near maximal treadmill test for exercise tolerance in angina pectoris. Circulation 1973;48:346-351. 18. Wahren J, Bygdeman S: Onset of angina pectoris in relation to cir- culatory adaptation during arm and leg exercise. Circulation 44. Lipkin DP, Canepa-Anson R, Stephens MR, Poole-Wilson PA: 1971;44:432-441. Factors determining symptoms in heart failure: Comparison of fast and slow exercise tests. Br Heart J 1986;55:439-445. 19. Shaw DJ, Crawford MH, Karliner JS, et al: Arm crank ergometry: A new method for the evaluation of coronary artery disease. 45. Guyatt GH, Sullivan MJ, Thompson PJ, et al: The sixminute walk. Am J Cardiol 1974;33:801-805. A new measure of exercise capacity in patients with chronic heart failure. Can Med Assoc J 1985;132:919-923. 20. DeBusk RF, Valdez R, Houston N, Haskell W: Cardiovascular responses to dynamic and static effort soon after myocardial infarc- 46. Bittner V, Weiner DH, Yusuf S, et al: Prediction of mortality tion: Application to occupational work assessment. Circulation and morbidity with a 6-minute walk test patients with left 1978;58:368-375. ventricular dysfunction. SOLVD Investigators. JAMA 1994;27: 661-662. 21. Vander LB, Franklin BA, Wrisley D, Rubenfire M: Cardiorespiratory responses to arm and leg ergometry in women. Phys Sports Med 47. Curtis JP, Rathore SS, Wang Y, Krumholz KM: The association of 1984;12:101-106. 6-minute walk performance and outcomes in stable outpatients with heart failure. J Card Fail 2004;10:9-14. 22. Fardy PS, Webb D, Hellerstein HK: Benefits of arm exercise in cardiac rehabilitation. Phys Sports Med 1977;5:30-41. 48. Rostagno C, Olivo G, Comeglio M, et al: Prognostic value of 6-minute walk corridor test in patients with mild to moderate 23. Asmussen E, Hemmingsen I: Determination of maximum working heart failure: comparison with other methods functional capacity at different ages in work with the legs or with the arms. evaluation. Eur J Heart Fail 2003;5:247-252. Scand J Clin Lab Invest 1958;10:67-71. 49. Golding LA, Myers CR, Sinning WE (eds). The Y’s Way to Physical 24. Franklin BA, Vander L, Wrisley D, Rubenfire M: Aerobic require- Fitness, 3rd ed. Champaign, Ill, Human Kinetics, 1989. ments of arm ergometry: Implications for exercise testing and training. Phys Sports Med 1983;11:81-90. 50. Chang JA, Frolicher VF: Clinical and exercise test markers of prognosis in patients with stable coronary artery disease. Curr 25. Schwade J, Blomqvist CG, Shapiro WA: A comparison of the Probl Cardiol 1994;19:533-538. response to arm and leg work in patients with ischemic heart disease. Am Heart J 1977;94:203-208. 51. Olona M, Candell-Riera J, Permanyer-Miralda G, et al: Strategies for prognostic assessment of uncomplicated first myocardial infarction: 26. Balady GJ, Schick EC, Weiner DA, et al: Comparison of determi- 5-year follow up study. J Am Coll Cardiol 1995;25:815-822. nants of myocardial oxygen consumption during arm and leg exercise in normal persons. Am J Cardiol 1986;57:1385-1387. 52. Borg G: Perceived exertion as an indicator of somatic stress. Scand J Rehabil Med 1970;23:92-93. 27. Balady GJ, Weiner DA, McCabe CH, et al: Value of arm exercise testing in detecting coronary artery disease. Am J Cardiol 1985; 53. Borg G: Borg’s Perceived Exertion Scales. Champaign, Ill, Human 55:37-39. Kinetics, 1998. 28. Myers J, Froelicher VF: Optimizing the exercise test for pharmaco- 54. Mason RE, Likar I: A new system of multiple-lead exercise logical investigations. Circulation 1990;82:1839-1846. electrocardiography. Am Heart J 1966;71:196-205. 29. Hermansen L, Saltin B: Oxygen uptake during maximal treadmill 55. Froelicher VF, Wolthuis R, Keiser N, et al: A comparison of and bicycle exercise. J Appl Physiol 1969;26:31-37. two bipolar electrocardiographic leads to lead V5. Chest 1976;70: 611-616. 30. Buchfuhrer MJ, Hansen JE, Robinson TE, et al: Optimizing the exercise protocol for cardiopulmonary assessment. J Appl Physiol 56. Rautaharju PM, Wolf HK, Eifler WJ, Blackburn H: A simple proce- 1983;55:1558-1564. dure of positioning precordial ECG and VCG electrodes using an electrode locator. J Electrocardiol 1976;9:35-40. 31. Myers J, Buchanan N, Walsh D, et al: Comparison of the ramp versus standard exercise protocols. J Am Coll Cardiol 1991; 57. Gamble P, McManus H, Jensen D, Froelicher VF: A comparison of 17:1334-1342. the standard 12-lead electrocardiogram to exercise placements. Chest 1984;85:616-622. 32. Niederberger M, Bruce RA, Kusumi F, Whitkanack S: Disparities in ventilatory and circulatory responses to bicycle and treadmill 58. Kleiner JP, Nelson WP, Boland MJ: The 12-lead electrocardiogram exercise. Br Heart J 1974;36:377-382. in exercise testing. Arch Intern Med 1978;138:1572-1573. 33. Wicks JR, Sutton JR, Oldridge NB, Jones NL: Comparison of the 59. Rautaharju PM, Prineas RJ, Crow RS, et al: The effect of modified electrocardiographic changes induced by maximum exercise limb positions on electrocardiographic wave amplitudes. J Electro- testing with treadmill and cycle ergometer. Circulation 1978; cardiol 1980;13:109-114. 57:1066-1069. 60. Shapiro W, Berson AS, Pipberger HV: Differences between supine 34. Balke B, Ware R: An experimental study of physical fitness and sitting Frank-lead electrocardiograms. J Electrocardiol of Air Force personnel. US Armed Forces Med J 1959;10: 1976;9:303-308. 675-688. 61. Sigler LH: Electrocardiographic changes occurring with alter- 35. Astrand PO, Rodahl K: Textbook of Work Physiology. New York, ations of posture from recumbent to standing positions. Am Heart McGraw-Hill, 1986, pp 331-365. J 1938;15:146-152.

C H A P T E R 2 Exercise Testing Methodology 39 62. Dougherty JD: Change in the frontal QRS axis with changes in 73. Fox KM, England D, Jonathan A, et al: Precordial surface mapping the anatomic positions of the heart. J Electrocardiol 1970;3:299- of the exercise ECG. Br J Hosp Med 1982;27:291-299. 311. 74. Yanowitz FG, Vincent GM, Lux RL, et al: Application of body 63. Bruce RA, Detry JM, Early K, et al: Polarcardiographic responses to surface mapping to exercise testing: ST80 isoarea maps in maximal exercise in healthy young adults. Am Heart J 1972;83: patients with coronary artery disease. Am J Cardiol 1982;50: 206-212. 1109-1113. 64. Mimbs JW, deMello V, Roberts R: The effect of respiration on 75. Simoons M, Block P, Ascoop C, et al: Computer processing of normal and abnormal Q-waves. Am Heart J 1977;94:579-584. exercise ECGs-A cooperative study. In VanBemmel KJ, Williams JL (eds): Trends in Computer-Processed Electrocardiograms. 65. Reikkinen H, Rautaharju P: Body position, electrode level, and res- Amsterdam, North-Holland Publishing, 1977, p. 383. piration effects on the Frank lead electrocardiogram. Circulation 1976;53:40-45. 76. Mirvis DM, Ramanathan KB: Alterations in transmural blood flow and body surface ST segment abnormalities produced by ischemia 66. Robertson D, Kostuk WJ, Ahuja SP: The localization of in the circumflex and left anterior descending coronary arterial coronary artery stenoses by 12-lead ECG response to graded beds of the dog. Circulation 1987;76:697-704. exercise test: Support for intercoronary steal. Am Heart J 1976; 91:437-444. 77. Gutman RA, Alexander ER, Li YB, et al: Delay of ST depression after maximal exercise by walking for two minutes. Circulation 67. Tucker SC, Kemp VE, Holland WE, et al: Multiple-lead ECG sub- 1970;42:229-233. maximal treadmill exercise tests in angiographically documented coronary heart disease. Angiology 1976;27:149-156. 78. Lachterman B, Lehmann KG, Abrahamson D, Froelicher VF: “Recovery only” ST segment depression and the predictive accu- 68. Chaitman BR, Bourassa MG, Wagniart P, et al: Improved efficiency racy of the exercise test. Ann Intern Med 1990;112:11-16. of treadmill exercise testing using a multiple lead ECG system and basic hemodynamic exercise response. Circulation 1978;57: 79. Akdemir I, Aksoy N, Aksoy M, et al: Does exercise-induced severe 71-78. ischaemia result in elevation of plasma troponin-T level in patients with chronic coronary artery disease? Acta Cardiol 2002; 69. Miranda CP, Liu J, Kadar A, et al: Usefulness of exercise-induced 57:13-18. ST-segment depression in the inferior leads during exercise testing as a marker for coronary artery disease. Am J Cardiol 1992;69: 80. Ashmaig ME, Starkey BJ, Ziada AM, et al: Changes in serum 303-307. concentrations of markers of myocardial injury following treadmill exercise testing in patients with suspected ischaemic heart disease. 70. Mirvis DM, Keller FW, Cox JW, et al: Left precordial Med Sci Monit 2001;7:54-57. isopotential mapping during supine exercise. Circulation 1977; 56:245-252. 81. Foote RS, Pearlman JD, Siegel AH, Yeo KT: Detection of exercise- induced ischemia by changes in B-type natriuretic peptides. J Am 71. Miller WT, Spach MS, Warren RB: Total body surface Coll Cardiol 2004;44:1980-1987. potential mapping during supine exercise. Circulation 1980; 62:632-645. 82. Sabatine MS, Morrow DA, de Lemos JA, et al: TIMI Study Group. Acute changes in circulating natriuretic peptide levels 72. Fox KM, Selwyn AP, Shillingford JP: A method for precordial in relation to myocardial ischemia. J Am Coll Cardiol 2004; surface mapping of the exercise electrocardiogram. Br Heart J 44:1988-1995. 1978;40:1339-1343.