REVIEW ARTICLE Sports Med 1999 May; 27 (5): 285-312

0112-1642/99/0005-0285/$14.00/0

© Adis International Limited. All rights reserved.

Exercise Stress Testing

An Overview of Current Guidelines
Scott A. Lear,1 Anka Brozic,1 Jonathan N. Myers2 and Andrew Ignaszewski1
1 Healthy Heart Program, St Paul’s Hospital, University of British Columbia, Vancouver, British
Columbia, Canada
2 Palo Alto Veterans Affairs Medical Center and Stanford University, Palo Alto, California, USA

Contents
Abstract . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 286
1. Haemodynamic Responses to Exercise . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 287
2. Metabolic Equivalents . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
3. Physiological Basis of Exercise Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 289
4. Pathology of Coronary Artery Disease . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
5. Potential Hazards of Exercise Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 290
6. Personnel . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
7. Emergency Preparation and Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 291
8. Contra-Indications for Exercise Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
9. Indications for Maximal Exercise Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
10. Indications for Terminating an Exercise Test . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
11. Types of Exercise Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
11.1 Submaximal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 292
11.2 Maximal . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
11.3 Gas Exchange . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
12. Testing Environment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 293
13. Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
13.1 Electrocardiograph . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
13.2 Electrodes . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
13.3 Blood Pressure Monitoring Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
13.4 Additional Equipment . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 294
14. Exercise Modalities . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
14.1 Treadmill . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 295
14.2 Bicycle Ergometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
14.3 Arm Ergometer . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 296
15. Exercise Protocols . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 297
16. Patient Preparation . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
16.1 Medications . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
16.2 Electrode Placement . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 299
17. Testing Procedures . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
18. Diagnostic Interpretation of Exercise Testing . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
18.1 Sensitivity and Specificity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 300
18.2 False-Positive and False-Negative Responses . . . . . . . . . . . . . . . . . . . . . . . . . . 301
18.3 Predictive Value . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
18.4 Interpretation of the Electrocardiogram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 301
18.5 Angina . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
19. Supplementary Diagnostic Tests . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 302
20. Exercise Prescription . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303
286 Lear et al.

20.1 Benefits and Risks . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 303

20.2 Principles . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304
21. Exercise Test Responses and Reporting . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
21.1 Pre-Test Information . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
21.2 Resting Electrocardiogram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
21.3 Haemodynamics . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
21.4 Exercise Capacity . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 307
21.5 Changes in the Electrocardiogram . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
21.6 Signs and Symptoms . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 308
22. Conclusion . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 309

Abstract Exercise stress testing (ET) is an inexpensive noninvasive tool that provides
valuable cardiopulmonary information in healthy and diseased populations. It is
most commonly used for diagnosing coronary artery disease (CAD) and devel-
oping appropriate exercise prescriptions (EP). With its widespread use and appli-
cation, it is imperative that safe and appropriate guidelines and procedures are
used, as there are a number of risks associated with testing in a population with or
suspected of having CAD. The focus should be on the patient’s safety: personnel
must be properly trained and aware of all emergency procedures, contra-indications
for ET and indications for test termination must be strictly adhered to.
Three main types of testing are prevalent: submaximal, maximal and maximal
utilising gas exchange. The maximal test is most commonly used, and the sub-
maximal is appropriate for hospitalised patients. Gas exchange data is essential
when assessing congestive heart failure and timing for heart transplantation.
ET is commonly performed using a treadmill or a bicycle ergometer. The
treadmill provides a more familiar exercise modality and has been shown to have
greater diagnostic sensitivity than the bicycle ergometer; it is, however, more
expensive and requires more space in the testing room. The bicycle ergometer is
more appropriate for those individuals who are severely obese or have problems
with extended periods of walking.
Regardless of the modality used, an appropriate exercise protocol should be
used. In North America, the Bruce protocol is the most common. However, the
Bruce protocol, and others that estimate exercise capacity based on equations,
tend to overestimate exercise capacity. They may be too demanding for those with
limited exercise capacity, and too long for those with high exercise capacity. For
these people, an exercise protocol that reaches maximal capacity in 8 to 12 min-
utes using smaller increments in workload should be considered.
Once completed, the results of ET needs to be correctly interpreted. This
includes reviewing the test results while considering the patient’s history, medi-
cations and indication for the test. ET can also be used to develop an EP for the
participant. An EP should take into account the intensity, modality of exercise,
frequency and duration, as well as being realistic for the individual and the goals
to be achieved. All the information from the test results and the pre-test exami-
nation should be presented in a report that also includes the advised EP.

Exercise stress testing (ET) is an inexpensive populations. Its widespread applications have re-
and noninvasive tool that provides valuable cardio- sulted in its increased utilisation in physicians’ of-
pulmonary information in healthy and diseased fices, hospitals and sports medicine facilities. It is

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 287

widely used in the diagnosis, prognosis and assess- for treadmill, cycle, step or arm cranking exer-
ment of numerous cardiopulmonary and neuro- cise.[1,6,10]
muscular disorders, along with assessing progres- Oxygen consumption is the product of cardiac
sion of disease and rehabilitation, medication ⋅
output (Q) and the arteriovenous oxygen differ-
efficacy and to screen for more resource-intensive ⋅
ence (a-vO2 difference). Q is directly related to
clinical procedures. heart rate (HR) and stroke volume (SV). At rest
ET has been an important procedure in cardio- ⋅
Q is usually in the range of 4.5 to 5.5 L/min. The
vascular medicine for decades; there is probably no normal cardiovascular response to progressively
other procedure that can rival ET in terms of its increasing dynamic exercise is a linear increase in
yield of clinically useful information. Over the ⋅
Q (fig. 1). This response is needed to meet the in-
years, guidelines have been developed by several creased demand of the working muscles for oxy-
organisations concerning the conduct and interpre- ⋅
gen. At maximal exercise, Q may increase to as
tation of ET.[1-7] It is important that these guide- much as 20 to 30 L/min.
lines be followed in order for ET to be safe and ⋅
As exercise intensity increases, Q increases pri-
accurate. This article draws upon current literature marily as a result of increasing HR. HR is influ-
and personal experience to provide an overview of enced by age (maximal HR decreases by 5 to 7
current practices in ET within the clinical setting. beats/minute per decade), state of health, medica-
To provide a more comprehensive review and un- tions (e.g. β-blockers), body position, type of ex-
derstanding of ET, background descriptions of the ercise, environmental conditions, the autonomic
physiological responses and pathological pro- nervous system and hormonal influences. The HR
cesses that underlie the basis of ET are provided as
response to graded exercise provides valuable
well as information on exercise prescription (EP)
information for diagnosis, prognosis and EP. A
and ET reporting. This article provides the depth
commonly used end-point during ET is the attain-
and scope of ET and should be of value to the prac-
ment of 80 to 90% of age-predicted maximal HR
tising clinician, technician and student alike.
(220 – age). However, because of individual vari-
ability in HR and the use of cardioactive medica-
1. Haemodynamic Responses
tions, this is not a very reliable end-point. Chrono-
to Exercise
tropic incompetence, or an abnormal increase in
To perform physical work, or exercise, the body HR with increasing workload, is associated with
must increase its delivery and use of oxygen to fuel the presence of heart disease and poor progno-
the working muscles. The amount of oxygen sis.[11] It is more important to consider exercise
.
utilised [oxygen uptake (VO2)] depends on the capacity than HR response during ET.
amount of physical work performed, which can be SV is the amount of blood pumped out of the
expressed by the Fick Principle. The Fick Principle left ventricle in a single heart beat, and is equal to
incorporates the physiological changes that occur the difference between left ventricular end-diastolic
.
at rest and during exercise (fig. 1) to quantify VO2. and end-systolic volumes. Resting values are in the
. range of 70 to 80 ml/beat. SV increases with exer-
VO2 = cardiac output × arteriovenous oxygen difference .
cise until about 50% of VO2max, at which point it
The amount of oxygen consumed by an individ- reaches a plateau. During maximum exercise, SV
ual at maximal exercise is termed the maximal may increase 1.5-fold from its resting value.
.
oxygen uptake (VO2max), and is influenced by age, The a-vO2 difference reflects the difference in
gender, exercise habits, heredity and health.[9] It the amount of oxygen in the arterial blood com-
can be measured directly when gas exchange data pared with venous blood. This difference is caused
are collected during cardiopulmonary ET, or pre- by the extraction of oxygen by tissues. At rest, a-vO2
dicted during standard ET by employing equations difference is approximately 4 to 5ml of O2/100ml

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
288 Lear et al.

Fig. 1 is not available for electronic viewing

Fig. 1. Response of basic haemodynamic and metabolic variables from rest to a moderately high level of exercise. a-vO2 difference =
arteriovenous oxygen difference (ml of O2/100ml of blood); TPR = total peripheral resistance (reproduced from Myers,[8] with
permission).

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 289

of blood. At maximum exercise, this can be as high ble I outlines some key MET values for maximal
as 15 to 17ml of O2/100ml of blood. The a-vO2 dif- exercise.
ference is affected by arterial haemoglobin con-
tent, pulmonary diffusion capacity, partial pressure 3. Physiological Basis of Exercise Testing
of alveolar O2 and the extraction of available O2.
⋅ When exercising, working muscles require a
Another important factor governing Q during
exercise is a decrease in total peripheral resistance. constant supply of oxygen. If the supply is insuffi-
Metabolic and neurological changes occur that cient, the muscles will become dependent on the
produce vasodilation in the active muscles, and va- anaerobic energy system, and soon fatigue. Under
soconstriction in the inactive muscles and visceral steady-state conditions the amount of oxygen re-
organs (e.g. kidneys, liver and gastrointestinal quired to maintain a given workload is sufficiently
⋅ met by the cardiovascular system. As workload and
tract), resulting in a major redistribution of Q with
acute exercise. oxygen demand increase, HR and SV increase. ET
Arterial blood pressure (BP) increases with ex- utilises this relationship to stress the cardiovascu-
ercise. Systolic BP increases linearly with increas- lar system and observe the response. The point at
ing workloads, whereas diastolic BP remains equal which the individual cannot continue is the partic-
to resting values or decreases slightly with increas- ipant’s maximal capacity, usually reported as
.
ing workloads in healthy individuals. It is not un- VO2max.
common for the fifth Karotkoff sound to be heard The ability of the heart to work is dependent
all the way to 0mm Hg in healthy individuals dur- upon the amount of oxygen supplied and extracted
ing exercise. Maximal systolic BP is related to age, from blood flowing through the coronary arteries.
exercise intensity, left ventricular function, medi- At rest, blood flows through the coronary arteries
cations and environment. during both diastole and systole, with most of the
Oxygen utilised by the heart is termed the myo- flow occurring during the relaxation phase, dia-
. .
cardial oxygen uptake (MVO2). During exercise stole. As HR and SV increase with exercise, MVO2
the metabolic requirement of the heart is 4 to 5 increases, primarily because of increases in blood
times that at rest, because of increases in HR, after- flow through arterial vasodilation. In healthy peo-
load and contractility. The main determinants of ple, the supply of oxygen to the heart during exer-
.
MVO2 are HR, ventricular wall tension and con- cise is sufficient and does not usually present a
tractility. An accurate assessment of these variables problem. In the presence of coronary artery disease
.
requires cardiac catheterisation. However, MV O2 (CAD), however, blood flow to the heart can be
can be reasonably estimated by the product of HR compromised even at rest, and the added stress of
and systolic BP (rate pressure product).
Table I. Clinically significant metabolic equivalents (MET) values
2. Metabolic Equivalents (from Fletcher et al.,[6] with permission)
. MET value Equivalent to
VO2 is often expressed as multiples of the aver- 1 Resting
age amount of oxygen required for sitting at rest 2 Level walking, 3.2 km/h (2 mile/h)
(3.5 ml/kg of bodyweight/min), termed metabolic 4 Level walking, 6.4 km/h (4 mile/h)
equivalents (METs). For any ET protocol, each <5 Poor prognosis, usual limit after myocardial
completed stage can be expressed in multiples of infarction, peak cost of basic daily living
10 Prognosis with medical therapy, as good as
METs. In general, 1 MET represents an increment
coronary artery bypass
on the treadmill of roughly 1.6 km/h (1 mile/h) or 13 Excellent prognosis regardless of other exercise
2.5% gradient. This allows for ease of comparison responses
between different test modalities and protocols, 18 Elite endurance athletes
and is superior to expressions of test duration. Ta- 20 World class athletes

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
290 Lear et al.

Table II. Physician supervision recommended during exercise testing (adapted from American College of Sports Medicine,[1] with permission)
Apparently healthy Increased riska Known disease
youngerb older no symptoms symptoms
Submaximal testing No No No Yes Yes
Maximal testing No Yes Yes Yes Yes
a Persons with 2 or more risk factors for coronary artery disease.
b Persons less than 41 years for men and less than 51 years for women.

exercise may result in a greater demand for oxygen • smoking results in increased circulating free
than can be supplied. This may result in myocardial radicals that oxidise particles such as LDL-C,
ischaemia, manifesting as either angina and/or which are toxic to the endothelial cells.[18]
indicative electrocardiographic (ECG) changes. As a result of endothelial dysfunction/injury,
Ischaemia can be detected on the ECG because of there is an increase in monocyte adherence and
altered transmembrane potentials causing a region endothelial permeability.[13] This is accompanied
of abnormal diastolic current, which is usually by an increased LDL-C flux through the endothe-
manifested as depression of the ST segment of the lial layer, and, once in the endothelial wall, the
ECG. LDL-C particles undergo modifications such as
oxidation. Once in the intima, monocytes trans-
4. Pathology of Coronary Artery Disease form into macrophages and phagocytose the mod-
ified LDL-C particles, becoming foam cells.[13]
The most common indication for ET is to diag- Smooth muscle cells and fibroblasts migrate from
nose the presence of CAD. It is therefore important the media into the intima and proliferate because
for ET administrators to understand the disease of the increased production of growth factors by
process. This is only a brief review, and readers are the endothelium and macrophages.[19] These cells
directed to the selected references for further infor- may also phagocytose the LDL-C particles through
mation. their scavenger receptors and form foam cells. The
CAD usually results from the formation of an accumulation of foam cells results in the formation
atherosclerotic plaque within the arterial wall. This of a fatty streak, and over time their volume increases,
causes a decrease in artery diameter that restricts causing a decrease in vessel lumen diameter.[20]
blood flow and may, over time, result in plaque The smooth muscle cells and fibroblasts produce
rupture, leading to thrombosis formation and com- connective tissue, which forms a firm fibromuscu-
plete obstruction of blood flow. The most com- lar outer layer, consisting of cholesterol and other
monly held theory of atherosclerotic plaque forma- cellular debris, on the inner wall of the vessel.[21]
tion is the ‘response to injury’ of the arterial Progression of lesions occurs at variable rates, and
wall.[12,13] The endothelial wall may be injured by as the lumen diameter becomes smaller, circulating
the following: blood flow becomes turbulent. This may lead to
• chronically increased BP, which can cause in- rupture of the plaque, exposing its contents to
creased arterial wall stress and turbulence[14] blood, resulting in a thrombus and myocardial
• diabetes mellitus, which increases the perme- ischaemia.[22]
ability of the endothelial wall, allowing mono-
cytes and lipids to infiltrate more readily[15]
5. Potential Hazards of Exercise Testing
• increased circulating low density lipoprotein-
cholesterol (LDL-C) can lead to dysfunction of Although ET is relatively safe, potential hazards
the endothelial wall[16] exist during every testing situation. In men and
• increased circulating homocysteine, believed to women without diagnosed CAD the complication
injure the endothelial wall[17] rate is 0.8 per 10 000 tests.[23] However, others

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 291

have reported 1 mortality and 2 complications re- Medicine (ACSM) offers certification programmes
quiring acute intervention per 10 000 tests.[24] Al- for ET technicians, nurses and exercise physiolo-
though these complication rates are extremely gists. It is strongly recommended that healthcare
small, it is nevertheless crucial that standard guide- professionals performing ET be certified.
lines be followed by qualified personnel, not only
to reduce the occurrence of complications but also
7. Emergency Preparation
to safely handle any complications that do occur.
and Procedures

6. Personnel Before commencing testing in any facility, emer-

gency procedures and equipment must be organised.
Traditionally, direct physician supervision has These procedures should be documented, available
been a requirement for ET in patient populations. in an accessible location and practised at least every
Under certain circumstances (outlined in table II) 4 months. The emergency plan should include safe
properly trained staff, a trained nurse or exercise evacuation of unstable patients to emergency facil-
physiologist, are capable of testing patients safely ities or out of the building should the need arise. A
without direct physician supervision.[25] If a phys- telephone with emergency numbers clearly posted
ician is not required to be in the immediate ET area, should be readily accessible in the testing room. If
one should be readily available when needed. A list testing occurs within a hospital, emergency plans
of core competencies for physicians and other staff should include calling and response of the cardiac
performing ET was issued jointly by the American arrest team. All personnel involved in ET must be
College of Chest Physicians, American College of trained in cardiopulmonary resuscitation and pref-
Cardiology and the American Heart Association erably advanced life support. In the event of an
(AHA).[26] This document listed the necessary cog- emergency, the supervising physician should im-
nitive skills for testing, which include: plement the necessary emergency procedures.
• knowledge of basic exercise physiology To provide a safe ET environment a minimum
• indications and contra-indications for ET of emergency equipment and drugs to be used in
• principles of interpretation life-threatening situations are necessary. This should
• emergency procedures. include a defibrillator that is tested daily and re-
Experience of at least 50 procedures was sugges- placed and repaired if there is any question of its
ted to be necessary to acquire the necessary skills function. Table III outlines the emergency equip-
for performing ET. The American College of Sports ment and drugs that are recommended.

Table III. Standard emergency equipment and drugs that should be available during exercise testing
Equipment Drugs
Resuscitation bags Adenosine Morphine sulfate
Adhesive tape Atropine Nitroglycerin
Airway equipment Bretylium Norepinephrine (noradrenaline)
Blood drawing equipment Diazepam Procainamide
Defibrillator and monitor Digoxin Propranolol/esmolol
Intravenous fluids Dobutamine Sodium bicarbonate
Intravenous sets and stands Dopamine Sodium nitroprusside
Intubation equipment Epinephrine (adrenaline) Theophyllinea
Oxygen tanks Furosemide (frusemide) Verapamil
Suction apparatus Isoprenaline (isoproterenol)
Syringes and needles Lidocaine
a For laboratories performing dipyridamole testing.

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
292 Lear et al.

8. Contra-Indications for 9. Indications for Maximal

Exercise Testing Exercise Testing
As our knowledge and experience has increased,
Although ET is a widely applicable test, there
so have the reasons for, and benefits of, using max-
are a number of contra-indications for specific pop- imal ET. The most common use of ET in the hos-
ulations. Table IV outlines the absolute and relative pital setting is for diagnosis in patients suspected
contra-indications to ET. Relative contra-indica- or symptomatic of CAD. ET is also used in assess-
tions may be superseded if good clinical judgment ing functional capacity in healthy and diseased
determines that the results from the test will be of populations or assessing the effect of an interven-
sufficient benefit to outweigh the risks. It is impor- tion such as coronary artery bypass surgery, angio-
tant to recognise contra-indications before testing plasty, an exercise programme or medication. A
so that complications can be avoided and safe test- complete list of testing indications is shown in
ing can prevail. table V.

10. Indications for Terminating an

Table IV. Contra-indications to exercise testing (from American Exercise Test
College of Sports Medicine,[1] with permission)
ET is usually performed to volitional fatigue,
Absolute
Recent change in the resting electrocardiogram suggesting
but other reasons exist for which ET should be ter-
infarction or other acute cardiac event minated for safety purposes (table VI). As with the
Recent complicated myocardial infarction other indications, these should be weighed against
Unstable angina the benefit of the testing information, and purpose
Uncontrolled ventricular arrhythmia of the test. As always, good clinical judgement
Uncontrolled atrial arrhythmia that compromises cardiac function should dictate actions.
Third degree atrioventricular heart block without pacemaker
Acute congestive heart failure
11. Types of Exercise Testing
Severe aortic stenosis
Suspected or known dissecting aneurysm
Active or suspected myocarditis or pericarditis
11.1 Submaximal
Thrombophlebitis or intracardiac thrombi
Submaximal ET is most appropriate clinically for
Recent systemic or pulmonary embolus
predischarge evaluations post-myocardial infarc-
Acute infections
tion, and for patients with orthopaedic or neurolog-
Significant emotional distress (psychosis)
ical disorders whose cardiovascular status is uncer-
Relative
tain. For the former, submaximal testing has been
Resting diastolic blood pressure >115mm Hg or resting systolic
blood pressure >200mm Hg shown to be important in risk stratification. [27,28]
Moderate valvular heart disease Submaximal testing can be useful for making ap-
Known electrolyte abnormalities (hypokalaemia, propriate activity recommendations, for recogni-
hypomagnesaemia) tion of the need for modification of the medical
Fixed-rate pacemaker regimen, or the need for further interventions. Sub-
Frequent or complex ectopy maximal testing is also appropriate for patients
Ventricular aneurysm
with a high probability of serious arrhythmias.
Uncontrolled metabolic disease (e.g. diabetes mellitus,
thyrotoxicosis or myxoedema)
Testing end-points for submaximal testing have
Chronic infectious disease (e.g. mononucleosis, hepatitis, AIDS) traditionally been arbitrary, but should always be
Neuromuscular, musculoskeletal or rheumatoid disorders that based on clinical judgment. A HR limit of 140
are exacerbated by exercise beats/minute and a MET level of 7 are often used
Advanced or complicated pregnancy for patients younger than 40 years of age, and limits

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 293

of 130 beats/minute and a MET level of 5 are often Table V. Indications for maximal exercise testing
used for patients over 40 years of age. For patients Predischarge after MI
receiving β-blockers, a Borg perceived exertion While submaximal exercise testing is more common, maximal
level (table VII) in the range of 7 to 8 (1 to 10 scale) exercise testing can be safely administered using careful patient
selection and strict termination criteria. This can be useful in
or 15 to 16 (6 to 20 scale) is a conservative end- determining activities that are safe and that the patient is
point. properly medicated
Maximal ET is more appropriate for patients >1 Post-discharge after MI, angioplasty or coronary artery
month after myocardial infarction, but submaxi- bypass graft surgery
mal testing has been used effectively and safely for Performed when the participant is ready to resume full activity,
this test can be used to develop an exercise prescription, adjust
risk stratification among patients less than 1 month medications and determine if further intervention is required
after an infarction. However, because of budgetary
Diagnostic testing
constraints, many centres now forgo submaximal Best used for diagnosis in patients with an intermediate
ET and perform a maximal test when the patient is probability of CAD as determined by history of symptoms (chest
adequately stable. pain, dizziness, exercise-related palpitations, etc.). In individuals
with a high probability, the test is best used for prognostic
purposes. In spite of populations that may have low test
11.2 Maximal specificity, the ET remains the best initial diagnostic test

Functional testing
Maximal ET is the most commonly performed In either healthy or unhealthy populations, maximal testing can
type of ET in the absence of any of the indications be used to establish levels of functional capacity and evaluate
interventions such as medications and exercise programmes
outlined in table V. It requires that the individual
continues exercising as workload is increased until Assessment of disease severity and prognosis
Severity of disease can be determined, as ischaemia is
volitional fatigue, or until the desired testing cri-
proportional to the amount of ST depression, number of leads
teria are reached. In order for the test to be termed and duration into recovery
maximal, the participant should reach greater than CAD = coronary artery disease; ET = exercise stress test; MI =
85% of their age-predicted maximal HR or an HR myocardial infarction.

equivalent to 85% of 220 – age. Unless otherwise

stated, all guidelines in this article are applicable
to maximal ET. • the presence of handrail support on the tread-
mill.
11.3 Gas Exchange The assumptions necessary for predicting MET
levels from treadmill or bicycle ergometer work
Although most ET does not incorporate gas ex- rates raise uncertainties as to the accuracy of esti-
change, its addition can provide valuable informa-
. mating work performed for an individual patient.
tion. VO2max is a significant predictor of prognosis More comprehensive reviews of ET utilising gas
in patients with congestive heart failure[29] and can exchange can be found elsewhere.[11,34]
be used to help determine the timing of heart trans-
plantation.[30] ET with gas exchange is the most
12. Testing Environment
accurate method with which to determine maximal
.
capacity, expressed as VO2max, as compared with The ET laboratory should be big enough to
estimations of work derived from equations of accommodate all required equipment, including
treadmill gradient and speed.[31-33] This is because emergency equipment, a bed and a defibrillator.
.
of the limited accuracy of estimating VO2max, as The temperature and humidity as well as ventilation
the estimations are dependent on: [9]
should be controlled for maximum safety and com-
• the testing protocol fort. Temperature should be between 18 and 22°C
• the mechanical efficiency of the patient and humidity less than 60%.[4,35] To reduce patient
• the diagnosis anxiety and increase comfort, posters of relaxing

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
294 Lear et al.

Table VI. Indications for stopping an exercise test (from American during exercise and recovery. The ECG recording
College of Sports Medicine,[1] with permission) system should be equipped with a monitor that can
Absolute display a minimum of 3 leads continuously along
Acute MI or suspicion of an MI with a printer that can provide printouts of 12-lead
Onset of moderate to severe angina ECGs at desired intervals. With the incorporation
Drop in systolic BP with increasing workload accompanied by
of computers for signal processing, caution must
signs or symptoms or drop below standing resting pressure
Serious arrhythmias (e.g. second or third degree atrioventricular
be taken as computer processing can hide a poten-
block, sustained ventricular tachycardia, increasing premature tially important ECG change or lead to false posi-
ventricular contractions, atrial fibrillation with fast ventricular tives.[36,37] Therefore, the computer-enhanced ECG
response)
should always be compared with the raw signal for
Signs of poor perfusion, including pallor, cyanosis or cold and
clammy skin accurate interpretation.
Unusual or severe shortness of breath A number of added features can accompany the
Central nervous systems, including ataxia, vertigo, visual or gait ECG recording system, such as arrhythmia sensing,
problems, or confusion
automated ST segment monitor and various types
Technical inability to monitor the ECG
of summary reporting. While these features can be
Patient’s request
helpful when used in coordination with good clin-
Relative
ical judgement, they cannot replace the need for
Pronounced ECG changes from baseline (>2mm of horizontal or
down-sloping ST depression or >2mm ST elevation) experienced personnel and safe practices. The AHA
Any chest pain that is increasing has developed detailed specifications for ECG re-
Physical or verbal manifestations of severe fatigue or shortness cording systems.[38]
of breath
Wheezing
13.2 Electrodes
Leg cramps or intermittent claudication
Hypertensive response (systolic BP >260mm Hg, diastolic BP The use of disposable silver/silver chloride elec-
>115mm Hg)
Less serious arrhythmias such as supraventricular tachycardia
trodes with shielded cables is advisable to reduce
Exercise-induced bundle branch block that cannot be motion artifact. Having the participant wearing a
distinguished from ventricular tachycardia central box around the waist from which the cables
BP = blood pressure; ECG = electrocardiogram; MI = myocardial originate is also desirable to further stabilise the ECG
infarction.
signal. A flexible knit top can be worn over the
electrodes and cables to prevent additional move-
ment.
scenes are often used to decorate the room. For
patient preparation an examining table, towels, 13.3 Blood Pressure Monitoring Equipment
tape, silver/silver chloride electrodes, abrasive
tape, razors for shaving excess hair and alcohol BP monitoring is best performed manually using
swabs should be available. Patient preparation can a wall-mounted, or adjustable stand, mercury sphyg-
either be completed in the testing room or a nearby momanometer instead of the less accurate aneroid
examination room. The AHA has published guide- models. Automated BP monitors are available at a
lines for clinical ET laboratories in which these greater expense but tend to lose accuracy as exer-
issues are discussed in detail.[35] cise motion increases.[39,40] BP cuffs of various sizes
should also be available and the monitoring equip-
13. Equipment ment routinely calibrated, cleaned and inspected.

13.1 Electrocardiograph 13.4 Additional Equipment

All ET facilities should be equipped with an ECG All testing rooms should be equipped with a bed
machine that can continuously monitor the ECG to provide added monitoring during recovery and

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 295

for emergencies. Visual aids are also helpful when 14.1 Treadmill
testing. Large, easy to read, hand held placards of
Studies comparing treadmill and cycle ergo-
the angina scale should be available that the patient .
can either refer to or point to when necessary (fig. metry have reported VO2max to be generally 10 to
2). In addition, the Borg rating of perceived exer- 20% higher and maximum HR to be 5 to 20%
tion scales (table VII) should be mounted in clear higher on the treadmill, because of the relatively
view of the participant while exercising. It may be greater muscle mass utilised.[42,43] Treadmill ET
helpful to have all scales available in a variety of also provides greater diagnostic sensitivity, elicit-
languages. To assure proper use of these scales, the ing ST depression and angina more frequently be-
participant needs to be carefully instructed on their cause of the increased HR.[44] Work rate on a tread-
use before commencing ET. mill is usually expressed in km/h or miles/h and
gradient in percent, and separate equations exist
14. Exercise Modalities for determining work rate during walking and run-
ning.[1]
There are 3 types of exercise: isometric (static), The treadmill should be electrically driven with
isotonic (dynamic) and resistive. Dynamic exer- side platforms (in order for the participant to famil-
cise is preferred for ET because it provides a vol- iarise themselves before initiation) and have a wide
ume stress rather than a pressure load on the heart range of speed and elevation settings. An emer-
that can be graduated.[8] Dynamic exercises such gency stop button should be clearly visible and
as walking or cycling increase myocardial oxygen readily accessible by the patient and the testing
demand to a higher level than isometric or resistive personnel. Treadmills usually provide handrails at
exercise. the front and the sides to allow the patient to steady
The 2 most common modes of ET are the tread- themselves; however, these should not be used to
mill and the bicycle ergometer. In North America, support the individual during the test. Once the pa-
the treadmill is the most commonly used modal- tient becomes comfortable with the procedure,
ity.[25] In situations when the participant is re- gripping of the handrails should be discouraged as
stricted to upper body movement, arm ergometer it causes an overestimation of maximal capacity,
testing may be appropriate. The advantages and increases artifact on the ECG and prolongs the
disadvantages of each are listed in table VIII. test.[45,46]

Table VII. Borg rating of perceived exertion scales

Original scale Revised scale
6 0 Nothing at all
7 Very, very light 0.5 Very, very weak
8 1 Very weak
9 Very light 2 Weak
10 3 Moderate
11 Fairly light 4 Somewhat strong
12 5 Strong
13 Somewhat hard 6
14 7 Very strong
15 Hard 8
16 9
17 Very hard 10 Very, very strong
18 • Maximal
19 Very, very hard
20

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
296 Lear et al.

remain relatively still, facilitating manual BP meas-

urements and minimising motion artifact. How-
ever, isometric gripping of the handlebars should
be discouraged. There are 2 types of bicycle ergo-
1 2 meters: manually or automatically braked. Manually
braked ergometers are cheaper but restrict the patient
to cycling at a constant pedal speed, whereas auto-
matically braked models maintain a constant work-
load over a range of pedal speeds.
The work rate of a bicycle ergometer is usually
expressed in watts or kilopond-metres/min (1 watt
is equal to approximately 6 kpm/min) and oxygen
consumption can be estimated by the following: [1]
.
VO2 (ml O2/kg bodyweight/min) =
3 4 (kpm/min × 2) + (kg bodyweight × 3.5)

Note that absolute work on a bicycle ergometer

.
is independent of bodyweight and the absolute VO2
is closely related to workload. However, relative
.
VO2 will be different for people of differing body-
weight at a given workload.

14.3 Arm Ergometer

For those individuals who have limitations in

their lower extremities (common in individuals
with vascular, orthopaedic or neurological condi-
tions) that would preclude them from use of either
a treadmill or bicycle ergometer, the use of an arm
ergometer is appropriate. These, like bicycle ergo-
Fig. 2. Hand signals for expressing exertional chest pain non- meters, are either manually or automatically braked.
verbally. Three is the appropriate end-point (reproduced from The person should be seated upright during the test
Myers,[41] with permission).
and the arms should be slightly bent when fully
extended, with the fulcrum of the cranking handle
adjusted at shoulder height. The use of an arm er-
14.2 Bicycle Ergometer gometer does present problems for BP measure-
ment and may increase signal noise. It also results
The bicycle ergometer is appropriate for patients
in a reduced maximal capacity compared with the
who are unable to perform extended walking (for bicycle or treadmill, as fewer muscles are utilised.[47]
example, those with claudication or orthopaedic When using the arm ergometer for testing,
problems), as it is a non–weight-bearing exercise. personnel should be aware of the physiological
It is essential that the bicycle be equipped with ad- responses to arm exercise.[47] Physiologically, a
justable handlebars and seat. The seat should be dynamic arm exercise involves a smaller muscle
adjusted so that there is a slight bend at the knee mass for a given workload than do exercises in-
when the leg is fully extended at the bottom of the volving the lower body. Other muscles of the back,
pedal stroke. The bicycle allows the upper body to chest, buttocks and legs are used for stabilisation

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 297

during arm exercise and oxygen consumption for progressive continuous exercise in which the de-
any equivalent workload is greater in arm com- mand is elevated to a patient’s maximal level
pared with leg exercise. Increases in HR and BP within a total duration of 8 to 12 minutes.[6,48,56]
are also more rapid. For patients with cardiovascular disease, modifica-
tions of the Balke-Ware protocol are recommended
15. Exercise Protocols because of its constant treadmill speed between 3.2
and 5.3 km/h (2 and 3.3 mile/h), equal increments
As ET can be performed for various reasons, the in gradient (2.5 or 5%), and equal (1 or 2) increases
protocol should be chosen to fit the individual be- in METs. In the absence of gas exchange tech-
ing tested and the purpose of the test. The many niques, it is important to report exercise capacity
different exercise protocols, listed in figure 3, have in METs rather then treadmill time, so that exercise
led to some confusion regarding how physicians capacity can be compared uniformly between pro-
compare tests between patients. Surveys have tocols.
shown that about two-thirds of testing is performed METs can be estimated from any protocol using
using the Bruce protocol.[24] In recent years, a standardised equations (see fig. 3 for MET val-
growing body of literature has supported more ues).[6] On a bicycle ergometer, 1 MET represents
gradual, individualised approaches to the test. an increment of roughly 20W (120 kg/min) for a
Large and unequal work increments have been 70kg person. The estimation of METs from equa-
shown to result in less accurate estimates of exer- tions requires several assumptions and can lead to
cise capacity, particularly for patients with cardiac inaccuracies. The steady-state requirement is rarely
disease. Recent investigations have demonstrated met for the majority of patients on most exercise
that work rate increments that are too large or rapid protocols, most clinical testing is performed among
result in a tendency to overestimate exercise capac- patients with varying degrees of cardiovascular or
ity, less reliability for studying the effects of ther- pulmonary disease, and individuals vary widely in
apy, and lowered sensitivity for detecting coronary their walking efficiency.[11] It has therefore been
disease.[6,33,48-55] Individualising the protocol, con- recommended that a patient be ascribed a MET
sidering the individual and purpose of the test, level only for stages in which all or most of a given
rather than employing the same protocol for every stage duration has been completed.
person, appears to offer several advantages for ET. An approach to ET that has gained interest in
Protocols suitable for clinical testing should in- recent years is the ramp protocol, in which work
clude a low-intensity warm-up phase followed by increases constantly and continuously. In 1981,

Table VIII. Comparisons between different exercise modalities

Mode Advantages Disadvantages
Treadmill Motorised Expensive
Familiar activity Takes up space
.
Greater VO2max and HR achieved Noisy
Best mode for eliciting angina Intimidating for some patients
Bicycle Inexpensive Exercise capacity depends on quadriceps strength
Small space requirement Uncomfortable for some
.
Relatively quiet Lower VO2max and HR achieved compared with treadmill
Easy to obtain BP and record ECG
Arm ergometer Inexpensive Smaller muscle group used
Use if orthopaedic problems Decreased exercise capacity
Lower maximum HR
.
BP = blood pressure; ECG = electrocardiogram; HR = heart rate; VO2max = maximum oxygen uptake.

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
© Adis International Limited. All rights reserved.

298
Functional Clinical O2 cost METs Bicycle Treadmill protocols METs
class status (ml/kg/min) ergometer

Bruce Balke-Ware USAFSAM 'Slow' McHenry Stanford ACIP CHF

USAFSAM

3 min %GR %GR %GR

1W = 6.1 stages at at at
kpm/min
Healthy, dependent on age, activity

mph %GR 3.3 mph 3 mph 2 mph

5.5 20 1 min
For 70kg stages
Normal bodyweight 5.0 18
56.0 16 kpm/min mph %GR mph %GR 16
and 26
52.5 15 25 3.3 25 3.4 24.0
I 24 mph %GR 15
49.0 14 23
1500 22 3.3 21 14
4.2 16 21 3.1 24.0
45.5 13 20 13
1350 19 3.3 20 22.5
42.0 12 18 3.3 18 3.0 21.0 12
17 20.0 mph %GR
38.5 11 1200 16 11
35.0 10
3.4 14 15 3.3 15 mph %GR 3.3 15 17.5 3.0 17.5 3.4 14.0
10
1050 14
Sedentary healthy

13 2 25 15.0 3.0 15.0

31.5 9 12 3.3 12 3.0 14.0 9
900 11 12.5 3.0 12.5
28.0 8 10 3.3 10 2 20 8
750 9 3.3 9 10.0 17.5 3.0 10.5 3.0 10.0
24.5 7 2.5 12 8 7
600 7 2 15 3.3 6 7.5 14.0 3.0 7.0 3.0 7.5
21.0 6 6 3.3 5 6
II 5.0 10.5
Limited

5 2.0 10.5
5 2 10
Symptomatic

17.5 450 1.7 10 4 5

3
2.5 7.0 3.0 3.0 2.0 7.0
14.0 4 300 2 2 5 4
1.7 5 1 3.3 0 2.0 3 0 3.5 2.5 2.0 2.0 3.5
III 10.5 3 3
150 2.0 0 2 0 2.0 0.0 1.5 0.0
7.0 2 1.7 0 2
1.0 0.0
IV 3.5 1 1
Sports Med 1999 May; 27 (5)

Fig. 3. Oxygen cost per stage of some commonly used protocols. %GR = percentage gradient; ACIP = Asymptomatic Cardiac Ischaemia Plot; CHF = Congestive Heart
Failure (modified Naughton); kpm/min = kilopond metres per minute; METs = metabolic equivalents; mph = miles/h; USAFSAM = United States Air Force School of
Aerospace Medicine.

Lear et al.
Guidelines for Exercise Stress Testing 299

Whipp et al.[43] first described cardiopulmonary re- cardiac disease or known CAD. If CAD has been
sponses to a ramp test on a cycle ergometer, and diagnosed then the discontinuation of cardiac med-
many of the gas exchange manufacturers now in- ications is not called for. The results from ET may
clude ramp software. Recently, treadmills have be used to titrate medical therapy up to a more
also been adapted to conduct ramp tests.[48,56] The effective level. For example, antihypertensive
recent call for ‘optimising’ ET would appear to be therapy can be evaluated to identify inappropriate
facilitated by the ramp approach, because large exercise-induced hypertension. If hypertension oc-
work increments are avoided and increases in work curs with minimal exercise then an adjustment to
are individualised, permitting test duration to be medications may be required. Similarly, antianginal
targeted.[6,48,50,53,54-56] As there are no ‘stages’, the therapy may be optimised by serial symptom-limited
errors associated with predicting exercise capacity ET to determine the most efficacious therapeutic
alluded to above are lessened.[6,48,56] agent, including β-blockers, calcium antagonists
and/or nitrates. In addition, ET while taking cardiac
16. Patient Preparation medications as prescribed is useful for the appro-
priate prescription of exercise and vocational coun-
Proper patient preparation is essential for carry-
selling.
ing out safe and accurate ET. When the ET is sched-
Conversely, when the purpose of ET includes
uled the participant should be given the following
the diagnosis of CAD, tests performed by patients
instructions: a light meal at least 2 hours before the
taking β-blockers, calcium antagonists and/or nit-
test; abstinence from cigarettes, alcohol and caf-
rates may have a reduced diagnostic value because
feine products for 3 hours before the test; bring a
of an altered haemodynamic response. In this situ-
list of current medications; and arrive in light com-
ation it may be appropriate to withdraw these med-
fortable clothing for exercise.
ications.[57] The absence of ST segment depression
Informed consent from the patient should also
despite the presence of CAD, so-called false neg-
be obtained and all questions and concerns of the
ative results, may occur if these medications are
patient addressed. This serves to make the patient
not withdrawn before testing. If tapering of medi-
aware of what to expect and any potential prob-
cations is considered necessary, it should be done
lems, greatly facilitating the testing process. A
under physician supervision.
thorough history of the patient, consisting of cur-
rent medical problems, physical activity practices, Table IX summarises some of the most com-
cardiac events and procedures, present medica- monly used medications in the cardiac population
tions and the indication for the test is essential for and their effects. Further information on their ac-
proper test interpretation, expectations of physio- tions can be found in the ACSM Resource Manual
logical responses and determination of the appro- for Guidelines for Exercise Testing and Prescrip-
priate testing protocol. The history should also as- tion.[58]
sess contra-indications to ET.
A clear demonstration on how to perform the 16.2 Electrode Placement
test, how to terminate it if necessary and any hand
Proper preparation of the areas of electrode
signals to be used should precede ET. It is impera-
placement is important to minimise resistance and
tive that the test does not begin until it is clear that
improve the signal-to-noise ratio. An ECG signal
the participant understands what is expected from
that cannot be properly interpreted is uninforma-
them.
tive and potentially dangerous. Proper skin prepa-
16.1 Medications
ration includes shaving the area of electrode place-
ment, cleaning with alcohol, marking electrode
Symptom-limited ET is a useful method of eval- placement with a felt pen and removing the super-
uating the medical management of patients with ficial layer of skin with an abrasive pad. The most

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
300 Lear et al.

Table IX. Common drugs and their impact on exercise testing (from American College of Sports Medicine,[1] with permission)
Drug Heart rate Blood pressure Electrocardiogram Exercise capacity
β-Blockers Rest: ↓ Rest: ↓, no change ↓ signs of ischaemia ↑ in those with angina
Exercise: ↓ Exercise: ↓ ↓ in those without angina
Calcium antagonists
felodipine Rest: ↑, no change Rest: ↓ ↓ signs of ischaemia ↑ in those with angina
isradipine Exercise: ↑, no change Exercise: ↓
nicardipine
nifedipine
bepridil Rest: ↓
diltiazem Exercise: ↓
verapamil
Digoxin No change in patients with No change ST depression (25-40%) ↑ in those with CHF and
sinus rhythm atrial fibrillation
↓ in patients with atrial
fibrillation and CHF
Nitrates Rest: ↑ Rest: ↓ Delayed signs of ↑ in those with angina
Exercise: ↑, no change Exercise: ↓, no change ischaemia (and CHF)
CHF = congestive heart failure; ↓ = decrease; ↑ = increase.

commonly used configuration of electrode place- upon the population tested, the definition of dis-
ment is that of Mason and Likar which places the ease and the criteria used for an abnormal test.
limb leads at the base of the limbs to reduce artifact
created by constant movement during exercise.[59]
18.1 Sensitivity and Specificity
17. Testing Procedures
The most common terms used to describe test
Table X outlines monitoring procedures to be accuracy are sensitivity and specificity. Sensitivity
followed before, during and after ET. It is impor- is the percentage of patients with CAD correctly
tant to follow these guidelines to minimise patient identified by ET. Specificity is the percentage of
complications. This will also ensure accuracy of patients without CAD correctly identified by ET.
the test and reduce errors in interpretation as well Sensitivity and specificity are inversely related,
as allow comparisons between tests to be made.
and are affected both by the population tested and
After the test is completed the patient should im-
by the choice of discriminant value for abnormal.
mediately be placed in the recovery position lying
supine while still monitored. This is generally rec- For example, if the population of individuals tested
ommended to further provoke indicative signs of has a greater prevalence or severity of disease (such
ischemia even if there have been no indications as triple vessel or left main coronary disease), the
during the test.[60] test will have a higher sensitivity. Alternatively, the
test will have a higher specificity when performed
18. Diagnostic Interpretation of in a group of younger healthier people. Table XI
Exercise Testing presents data from a compilation of numerous stud-
ET is mostly prescribed for the evaluation of ies assessing the diagnostic value of ET through
chest pain. In this role, the test serves the very coronary angiography. The table displays a wide
important purpose of screening individuals who range of sensitivity and specificity values that can
should or should not undergo additional proce- be explained by the differences in protocol among
dures. How accurately ET distinguishes individuals the individual studies, such as population selection
with disease from those without disease depends and different testing protocols.

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 301

18.2 False-Positive and figure 4. Few of these studies, however, have fol-
False-Negative Responses lowed accepted rules for evaluating a diagnostic
test.[61] Most published guidelines continue to sug-
Factors that can be associated with a false-positive gest the application of a traditional criterion, ≥1.0mm
or false-negative response should be considered
ST segment depression at the J point, which is hor-
before the test. A false-positive response is defined
izontal or down-sloping over the following 80 msec.
as an abnormal ET response in an individual with-
ST segment changes >1.0mm which are down-
out significant heart disease, and causes the speci-
sloping or horizontal are generally indicative of
ficity to be decreased. A false-negative response
more severe CAD. Slowly up-sloping ST segment
occurs when the test is normal in an individual with
depression should be considered a borderline re-
disease, and causes the sensitivity of the test to be
sponse and assessed with other clinical data. ST
reduced. Factors that have been associated with
segment depression appearing only in recovery
false-positive and false-negative responses are
provides useful clinical data; combined with
listed in table XII. In individuals who have a high
probability of a false-positive or false-negative changes during exercise, it increases the sensitivity
test, an alternative procedure (exercise or pharma- of the test. The severity of CAD is also reflected in
cological echocardiogram or radionuclide test) the number of positive leads and the time of onset
may be appropriate. of ischaemic ST segment changes during ET. ST
segment elevation occurring over leads without Q
18.3 Predictive Value waves is indicative of severe transmural ischaemia.
ST segment elevation over leads with Q waves is
Another important term that helps define the di- more common and is related to the presence of dys-
agnostic value of a test is the predictive value. The
predictive value of an abnormal test (positive pre-
dictive value) is the percentage of individuals with Table X. Testing procedure guidelines (from American College of
Sports Medicine,[1] with permission)
an abnormal test result who have disease. Con-
versely, the predictive value of a normal test (neg- Pre-test
Resting 12-lead ECG in both supine and standing positions (an
ative predictive value) is the percentage of individ- additional ECG in the testing position may be needed if not using
uals with a normal test result who do not have a treadmill)
disease. The predictive value of a test cannot be Resting BP as outlined above
determined directly from the sensitivity and spec- During the test
ificity, but are strongly associated with the preva- Continuous on-screen ECG monitoring throughout the test
lence of disease in the population tested. The cal- 12-lead ECG printouts during the last minute of each stage or
every 2 minutes
culations used to determine sensitivity, specificity
BP recorded at the end of each stage
and predictive value are presented in table XIII.
Rating of perceived exertion at the end of each stage using
either the original or revised scales
18.4 Interpretation of the Electrocardiogram Symptomatic ratings used every minute after initial onset until
test cessation
Ever since ECG changes were first associated All of the above should also be recorded at the onset of any
with myocardial ischemia in the 1920s, the diag- chest pain, angina, ST depression >1mm or any other symptom

nostic ECG criteria and leads which exhibit abnor- Post-test

malities during exercise have been the source of 12-lead ECG recorded immediately post-exercise and every 2
minutes thereafter for at least 5 minutes and until any changes
significant debate. Numerous ECG criteria, includ- have returned to baseline
ing complex mathematical constructs, combined BP recorded immediately post-exercise and every 2 minutes
scores and ST areas during exercise and recovery thereafter for at least 5 minutes and until stabilised near baseline
have been proposed to optimally diagnose the pre- Continual use of symptomatic ratings until symptoms disappear
sence of CAD, a number of which are depicted in BP = blood pressure; ECG = electrocardiogram.

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
302 Lear et al.

Table XI. Diagnostic value of exercise testing from angiographic tions, Froelicher’s Exercise and the Heart[39] should
studies (from Philbrick et al.,[61] with permission)
be consulted.
Investigator Year n Sensitivity Specificity
(%) (%)
18.5 Angina
Hultgren 1967 55 66 100
Eliasch 1967 65 84 81 Diagnostically, the distinction between typical
Demany 1967 75 64 49
and atypical angina is an important one. Typical
Mason 1967 84 78 89
angina tends to be consistent in its presentation
Kassenbaum 1968 68 47 97
Roitman 1970 100 73 82
(chest discomfort or squeezing) and location, is
Newton 1970 52 57 81 brought on by a physical or emotional stress, and
Fitzgibbon 1971 160 48 80 is relieved by rest or nitroglycerin. Atypical angina
Cohn 1971 110 86 73 refers to pain that has an unusual location, prolonged
McConahay 1971 100 35 100 duration, or inconsistent precipitating factors that
Ascoop 1971 96 59 94 are unresponsive to nitroglycerin. Exercise-induced
Martin 1972 100 62 89
chest discomfort that has the characteristics of sta-
McHenry 1972 166 81 95
ble typical angina provides better confirmation of
Kellerman 1973 74 54 96
Bartel 1974 465 65 92
the presence of CAD than any other test response.
Piessens 1974 70 65 83 Table XIV outlines the pretest probability of CAD
Rios 1974 50 83 89 based upon patient age, gender and symptoms;
Sketch 1975 251 53 88 details may be found in Diamond & Forrester.[63]
Borer 1975 89 49 41 A patient exhibiting the combination of typical
Jelinek 1976 153 45 89 angina and an abnormal ST response has a 98%
Goldschlager 1976 153 45 89 probability of having significant CAD. An impor-
Santinga 1976 283 73 78
tant indication to stop ET is moderately severe
Detry 1977 98 55 85
Chaitman 1978 100 88 82
angina, or pain that would normally cause the
McNeer 1978 1222 53 91 patient to stop daily activities and/or take sublin-
Balnave 1978 70 81 100 gual nitroglycerin.[5,41]
Berman 1978 164 84 67
Weiner 1978 302 76 76 19. Supplementary Diagnostic Tests
Chaitman 1979 200 84 72
Weiner 1979 2045 79 69 Radionuclide imaging techniques are a valuable
Aldrich 1979 181 40 92 complement to exercise ECG for evaluation of pa-
Raffo 1979 100 91 96 tients with known or suspected CAD. They are par-
Borer 1979 75 63 95 ticularly helpful in patients with equivocal exercise
Averages 66 84 ECGs or those likely to exhibit false-negative re-
n = number of patients. sponses, and are frequently used to clarify abnor-
mal ST segment responses in asymptomatic indi-
viduals, or in whom the cause of chest discomfort
kinetic areas. The majority (probably >90%) of ST remains uncertain.[64] They have also been shown
changes will occur in the lateral precordial leads.[6] to be useful in stratifying risk; patients exhibiting
both a positive exercise ECG and a positive radio-
Although it has historically been thought that the
nuclide scan have been shown to have a 3.6-fold
diagnostic performance of the test was incomplete increased risk for subsequent coronary events. [65]
without all 12 leads, recent studies suggest that ST Perfusion imaging of the coronary anatomy has
segment changes isolated to the inferior leads are been shown to be more sensitive and specific than
frequently false-positive responses.[62] For a com- the exercise ECG for detecting CAD. It permits the
prehensive review and analysis of ECG interpreta- localisation of ischaemia, which is not possible

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 303

with ST segment depression on the ECG. In addi- cological stress testing. Like standard ET and
tion, agents used in perfusion imaging permit the radionuclide techniques, the diagnostic accuracy
differentiation between fixed defects, which reflect of echocardiography depends primarily on the spe-
an area of myocardial infarction, versus reversible cific methodology used and the pre-test probability
defects, which represent an area of ischaemia. An of coronary disease in the individual tested. In ad-
extensive review of the literature suggested the sen- dition, the accuracy of echocardiographic testing is
sitivity and specificity of exercise thallium scinti- dependent upon observer experience. A summary of
graphy for detecting coronary disease were in the studies published since the advent of exercise
order of 84 and 87%, respectively. [66] However, the echocardiography in the early 1980s suggests that
use of thallium presents a number of drawbacks. the sensitivity and specificity of this technique for
Technetium-99m is a more recently developed radio- detecting coronary disease are both approximately
85%.[70]
nuclide agent that has a longer half-life, allowing
larger doses to be applied, it is more easily gener-
20. Exercise Prescription
ated and provides superior imaging.[67] A number
of studies have shown that technetium-99m and
20.1 Benefits and Risks
thallium are similar in their ability to identify patients
with CAD,[67,68] but technetium has been shown to Exercise is a common mode of disease preven-
be better able to identify individual diseased ves- tion and treatment, the benefits of which are nu-
sels[68] and shows significantly better specificity in merous for both healthy and diseased populations.
women.[69] Reductions in morbidity and mortality,[71,72] im-
Echocardiographic imaging of the heart is being provements in functional capacity,[73] psychologi-
increasingly utilised during exercise and pharma- cal well-being,[74,75] and quality of life[76,77] have
shown to be among the benefits of exercise pro-
grammes. Improvements in clinical health status
Table XII. Causes of false-negative and false-positive tests during
exercise testing are due to the beneficial changes that regular exer-
cise has on modifiable CAD risk factors, including
False positives
dyslipidaemia,[78] glucose resistance and insulin
Resting repolarisation abnormalities (e.g. left bundle branch block)
Cardiac hypertrophy
sensitivity,[79,80] hypertension,[81] obesity[82] and
Accelerated conduction defects (e.g. Wolfe-Parkinson-White hypercoaguability.[83] Regular exercise also elimi-
syndrome) nates physical inactivity as a risk factor. A number
Digitalis of resources are available that outline EP guide-
Nonischaemic cardiomyopathy lines for healthy and special populations; therefore
Hypokalaemia
this section will focus on EP for the CAD popula-
Vasoregulatory abnormalities
Mitral valve prolapse
tion only.[1,6,84,85]
Pericardial disease Patients participating in exercise programmes
Coronary spasm in absence of coronary artery disease present a number of safety concerns. Therefore,
Anaemia this is a primary reason for implementing an EP.
Female gender The risk of sudden death from exercise in patients
False negatives with CAD is higher than in healthy individuals, and
Failure to reach ischaemic threshold secondary to medications is estimated to be 1 in 80 000 to 160 000 exercise-
(e.g. β-blockers)
hours.[86] The risk of sudden death during exercise
Monitoring an insufficient number of leads to detect
electrocardiographic changes
has been difficult to predict, but may be higher in
Angiographically significant disease compensated by collateral patients with left ventricular dysfunction and/or a
circulation history of malignant dysrhythmias. Telemetered
Musculoskeletal limitations preceding cardiac abnormalities ECG monitoring during exercise is no longer

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
304 Lear et al.

Table XIII. Terms used to demonstrate the diagnostic value of a test mulated physical activity of moderate intensity
Term Definition every day.[87] In the CAD population, the basis of
Sensitivity 100 × TP/(TP + FN) the EP should provide enough caloric expenditure
Specificity 100 × TP/(TN + FP)
Positive predictive value 100 × TN/(FP + TN)
for bodyweight management and risk factor reduc-
Negative predictive value 100 × TP/(FP + TN) tion and be of an intensity to increase functional
FN = false negatives, or those with normal test results with disease; capacity. The main components of the EP include
FP = false positives, or those with abnormal test results and no frequency, intensity, duration, mode and progres-
disease; TN = true negatives, or those with normal test results and
no disease; TP = true positives, or those with abnormal test results sion of training, as well as an adequate warm-up
and with disease. and cool-down period. Table XV outlines the gen-
eral principles of an EP and may be used as a guide.
However, the EP that looks the best on paper is
thought to be necessary in most patients in cardiac worthless if it is not followed. Other aspects of the
rehabilitation, although the American Association patient must be considered when developing an in-
of Cardiovascular and Pulmonary Rehabilitation dividualised EP. The EP must be compatible with
and AHA guidelines have outlined specific popu- the individual’s lifestyle and activity preferences.
lations in which monitoring may be necessary.[6,7] A lack of time is the most commonly cited reason
The degree to which a given patient should be mon- for not exercising, and behavioural and psycholog-
itored by telemetry, HR, perceived effort or other
ical, in addition to physiological, variables are re-
means should be individualised and should be con-
lated to physical activity adherence.[88] Other fac-
sidered when designing an EP.
tors, such as self-efficacy and perceived barriers,
must also be taken into consideration. Present
20.2 Principles
physical status is also important; many CAD pa-
The EP must take into account the promotion of tients beginning an exercise programme have been
regular activity. The Centers for Disease Control previously inactive and/or have a low functional
and Prevention and the ACSM recommend that all capacity. The EP should therefore be directed towards
people participate in 30 minutes or more of accu- increasing activity levels at a reasonable rate.

Time in milliseconds
2.0
100
1.5

1.0
mV

A A A

140
ST2 A
ST6 T
100 ST4
0.5 80
Ab 60
N 60 60
0
−0.2
−0.4 W Ab −2mV ST End of QRS
slope

'Classic' Upsloping Amplitude at ST index ST integral Spatial ST-T ST 60

time-normalised magnitudes
ST midpoint

Fig. 4. Visual and computer criteria for identifying ischaemia from the electrocardiogram. As shown, numerous criteria are in use to
determine ischaemia, which can be calculated based upon the amplitude, duration and slope of the ST segment. The ST integral
used by most commercial systems initiates the area at a fixed point after the R wave and then ends 80 msec thereafter. Ab = abnormal;
N = normal; W = worse.

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 305

Table XIV. Pretest probability of coronary artery disease (CAD) by age, gender and symptomsa (from Gibbons et al.,[3] with permission)
Age (y) Gender Typical/definite angina pectoris Atypical/probable angina pectoris Nonanginal chest pain Asymptomatic
30-39 Men Intermediate Intermediate Low Very low
Women Intermediate Very low Very low Very low
40-49 Men High Intermediate Intermediate Low
Women Intermediate Low Very low Very low
50-59 Men High Intermediate Intermediate Low
Women Intermediate Intermediate Low Very low
60-69 Men High Intermediate Intermediate Very low
Women High Intermediate Intermediate Very low
a No data exist for patients <30 or >69y, but it can be assumed that prevalence of CAD increases with age; in a few cases, patients with
ages at the extreme of the decades listed may have probabilities slightly outside the high or low range.
High = >90%; intermediate = 10-90%; low = <10%; very low = <5%.; y = years

The exercise programme should consist of a with resting HR. Exercise at this intensity is con-
minimum of 3 sessions per week with a progres- sidered moderate and is usually sufficient to meet
sion to regular daily activity. Having a more fre- all of the goals and safety considerations of an EP.
quent exercise programme should take precedence For those people who are even more limited, exer-
over considerations of intensity or duration, which cise may need to be started at a lower intensity of
allows the patient to incorporate exercise into their 40 to 60% of HR reserve. As the individual pro-
daily routine. Once regular activity has been gresses through the exercise programme, intensity
achieved, the focus can then turn to adjustments of may be increased to 70 to 85% of HR reserve under
intensity and duration. the guidance of an appropriate healthcare profes-
A number of formulae are available with which sional. When increasing intensity, duration of the
to prescribe exercise intensity. Too high an inten- activity may be decreased to more easily accom-
sity can unnecessarily increase the patient’s risk modate the new EP. For those patients with isch-
and discourage compliance. Intensity is usually ex- aemia and/or arrhythmias, the target HR for the EP
pressed in terms of percentage of maximal capac- should be in a 10- to 15-beats/min range 10 beats/min
ity, maximal HR, work level (such as METs or W), below the HR of onset. Intensity prescribed as a
or a rating of perceived exertion. It is recom- percentage of maximal capacity should begin at
mended that all CAD patients undergo ET before 50% of what was achieved on ET; this is roughly
any exercise programme to prescribe exercise in a equivalent to 60% of maximal HR and is consid-
safe range, especially in those patients with isch- ered to be of a moderate intensity.
aemia and/or arrhythmias. This is also helpful in Intensity can be prescribed in terms of workload
patients using chronotropic medications, as the re- such as METs or W. This is helpful when using
lationship between HR and work is altered. If ET treadmills and bicycle ergometers with computer
has not been performed, exercise intensity should displays. Intensity may be prescribed within the 50
not exceed 20 beats/minute above resting HR, and to 80% range with the above considerations. Using
these people should be carefully monitored. Pre- the rating of perceived exertion scale is ideal for
scribing exercise intensity based upon age-predicted those patients unable to increase their HR and who
maximal HR is not recommended, since maximal may have trouble with HR monitoring. Initial ex-
HR is quite variable and may result in an inappro- ercise intensity in the range of 3 or 4 on the 10-point
priate intensity.[89] scale, or 12 to 15 on the 20-point scale, is advisable.
If HR is used to guide intensity, exercise should As individuals improve in fitness, a given exercise
begin at 60% of HR reserve [(maximal HR (as de- workload will become easier and will have to be
termined by ET) – resting HR) × 0.6] combined increased to maintain the same perception of effort.

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
306 Lear et al.

Table XV. Exercise prescription guidelines based on results from an exercise stress test (ET)
Parameter ET interpretation
negative for ischaemia positive for ischaemiaa
Frequency × 3-7/wk × 3-7/wkb
.
Intensity RPE 11-15 on Borg scale, 40-85% HRR or VO2max or 10-15 beats/min below ischaemic thresholdc
60-90% HRmax
Duration 20-60 min;d warm-up and cool-down 5-10 min 20-60 min;d warm-up and cool-down >10 min
Mode Aerobic exercises including walking, bicycling, rowing, Aerobic exercises including walking, bicycling, rowing,
swimming etc. swimming etc.
a Angina or associated symptoms (severe dyspnoea, nausea etc.) or ≥1mm ST depression during ET.
b If exercise capacity is markedly low [<4 metabolic equivalents (METs)] then the frequency of exercise may be increased to × 2-3/day with
each session having a shorter duration.
c The heart rate at which angina or ≥1mm ST depression occurs.
d The duration of exercise may be accumulated over several exercise sessions per day.
HRmax = maximum heart rate; HRR = heart rate reserve; min = minutes RPE = relative perceived exertion; wk = weeks.

While useful, the rating of perceived exertion healthy individuals, CAD patients progress more
scales require greater patient education to be con- slowly and care should be taken not to speed the
sidered reliable. patients through the programme. Progression should
A major consideration of the EP is often total first focus on increasing activity duration rather
caloric expenditure; adjusting duration is the most than frequency and intensity. As the functional ca-
effective means of increasing caloric expenditure pacity of the patient improves, the amount of work
and will be tolerated more easily than adjustments
needed to elicit the same HR response will also
to intensity. Exercise sessions should have a goal
need to increase. Progression is a key part of the EP
between 20 and 60 minutes duration at the pre-
scribed intensity. If the patient cannot achieve this and one should always consider the goals and
initially, then intermittent sessions throughout the health of the individual.
day of shorter duration can be used. As functional An important component of every exercise ses-
capacity improves, these intermittent sessions can sion is a sufficient warm-up and cool-down. The
increase in duration and may be combined. Pro- warm-up should be approximately 10 to 15 min-
gression should be made to achieve a weekly cal- utes long and consist of aerobic activity with grad-
oric expenditure of 2000 kcal to optimise benefits ual increases to the target HR or workload. This is
to functional capacity and CAD regression.[73] especially important in those individuals with isch-
Appropriate activities are those that are aerobic aemia, as a warm-up allows time for the body to
and use large muscle groups, including walking, reach a steady state. The cool-down should also be
jogging/running, bicycling, swimming, rowing and a minimum of 10 minutes and consist of a gradual
stair climbing. For the EP to be effective, the mode decrease in workload. This avoids blood pooling
of exercise must be one the participant enjoys. For
that may occur with stopping exercise suddenly,
the CAD patient, it is useful to use one of the many
and limits any muscle discomfort that may occur.
indoor exercise machines available so that work-
load can be safely controlled. To alleviate boredom Gentle stretching exercises can also be incorpo-
that may occur when using indoor machines, rotat- rated into the warm-up and cool-down.
ing between different activities every 10 minutes Recent reports have shown that resistance train-
may be helpful. ing can be of benefit to the CAD patient and the
As the patient progresses, the EP should be re- reader is directed to resources that discuss this is-
evaluated based on performance. Compared with sue.[1,7,86]

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 307

21. Exercise Test Responses • in patients with right bundle branch block, only
and Reporting changes occurring in lead V5 and V6 are asso-
ciated with ischaemia
The test report should summarise the patient’s • in patients with ST segment depression on the
response and should contain relevant information resting ECG, an additional 2.0mm ST segment
about the patient’s history, pre-test risk factors, depression has been shown to have acceptable
resting HR, BP and ECG findings, exercise haemo- diagnostic characteristics.
dynamics, exercise ECG, symptomatic response to
the test and recovery ECG changes. Useful pro- 21.3 Haemodynamics
grammes have been developed which automatically
The HR and BP responses should be reviewed
summarise the test responses and apply regression
and recorded. The rate pressure product is a valid
equations that report pre- and post-test risk of dis-
estimate of myocardial oxygen demand, and has
ease, and some provide mortality estimates.[90]
been shown to be important prognostically (pa-
tients achieving greater than 25 000 have been
21.1 Pre-Test Information shown to have a superior prognosis). HR should
rise progressively in proportion to the work rate.
ET should be an extension of the history and Although maximal HR is related to age, there is a
physical, and the test results should only be inter- great deal of variability between the two. Blunted
preted in the context of the patient’s pre-test history HR responses can be caused by β-blockade, decon-
outlined in section 16. In fact, studies have shown ditioning and chronic disease states. Much of what
that the greatest diagnostic information is con- has been called ‘chronotropic incompetence’ is re-
tained in the pre-test data. Test accuracy is strongly lated to early test termination as a result of angina;
influenced by the patient’s chest pain character- however, others may be limited by myocardial dys-
istics, age, gender and other risk factors before the function caused by previous myocardial dam-
test. If resting pulmonary function data are avail- age.[90] Systolic BP should increase with increas-
able, they should also be reported in the pre-test ing workloads, generally in the range of 40 to
information. 80mm Hg. Diastolic pressure generally does not
change appreciably. Exercise-induced hypoten-
21.2 Resting Electrocardiogram
sion (defined as a failure of systolic pressure to
rise, a drop below standing rest, or a drop of 20mm
A record of the resting electrocardiogram is im- Hg or more) has been demonstrated in most studies
portant as it can greatly influence test interpreta- to predict either a poor prognosis or a high prob-
tion. Although the standard ECG interpretation ability of severe angiographic coronary disease.
should be made with the patient supine, ECG
21.4 Exercise Capacity
changes during exercise should be measured by
comparing only with the standing ECG. The rest- Exercise capacity is the most important prog-
ing ECG has an important impact on the interpre- nostic feature of the test, and suggests a great deal
tation of the exercise ECG and, occasionally, a rest- about the efficacy of current therapies and the ef-
ing ECG abnormality can contra-indicate the test. fects of interventions such as surgical procedures,
Several resting ECG abnormalities are notable in exercise training and percutaneous transluminal
terms of how they affect ET and they are sum- coronary angioplasty (PTCA). Exercise capacity
marised as follows: should be reported in METs rather than exercise
• left bundle branch block and Wolfe-Parkinson- time. A patient should only be ascribed a MET
White syndrome negate any association of the level for a given stage if they have completed all
ST segment response with ischaemia or most of the duration of that stage. It should also

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
308 Lear et al.

be noted that an estimated MET level is usually the resting level (regardless of the presence of ele-
overpredicted for patients with heart disease. An vation or depression at rest).
additional and useful way that exercise capacity (iv) The interpretation of ST segment changes
should be expressed is as a percentage of the age- and should include those occurring in the recovery
gender-predicted value. A variety of equations and period.
nomograms have been developed for this purpose The ST segment response should be reported as
(figs. 5 and 6). When exercise capacity is measured up-sloping, horizontal or down-sloping, and the
directly using gas exchange techniques, both the lead(s) in which the ECG changes occurred should
measured and estimated values should be reported. be specified. Other information which should be
noted includes exercise-induced dysrhythmias and
21.5 Changes in the Electrocardiogram the development of conduction defects. Premature
ventricular contractions (PVCs) are usually classi-
Clinically, the degree of ST segment depression fied as occasional (<6/min), or frequent (>6/min),
induced by exercise is the cornerstone of ET. When and it is important to note whether they are coupled
reporting ST changes, the following rules should or multifocal and whether there were any runs of
be applied. ventricular tachycardia (defined as 3 or more PVCs
(i) ST segment depression should be reported as in succession).
a change from the standing resting level. When ST
segment depression is present at rest, only the amount 21.6 Signs and Symptoms
of additional depression should be reported.
(ii) When there is ST elevation at rest, the degree Signs and symptoms that should be noted in the
of ST depression should be measured from the iso- test report include subjective responses, chest pain
electric line (P-R segment). characteristics, dyspnoea and claudication. Subjec-
(iii) The degree of exercise-induced ST eleva- tive responses involve careful observation of the
tion should be reported as an absolute change from patient’s appearance, identification of those who
are uncooperative or exaggerate their limitations or
symptoms and neurological manifestations such as
20 0 lightheadedness or vertigo. Lightheadedness, a
25
1 drop in skin temperature or pallor may be signs of
30 2 inadequate cardiac output.
20
35 3 ET is the best opportunity to evaluate a patient’s
40
30 4 chest pain. The specific characteristics of chest
40
45 5 pain sensations represent the most important diag-
50
nostic feature of the test, so it is vital that it be
Age (y)

6
METs

50 60
55 70 7 reported accurately. This should involve question-
80
60 90 8 ing the patient carefully to distinguish between dif-
100
65 110 9 ferent types of pain. The pain should be qualified
120
70
130
140 10 in the report as noncardiac, atypical or typical
150 angina; otherwise the interpretation is inadequate.
75 11
80 12 Also, it should be clear if it is the reason for stop-
85 13 ping the test.
90 14 Dyspnoea can be caused by heart disease or lung
15 disease or their combination. The cause of dys-
pnoea can be determined by ET in conjunction with
Fig. 5. Nomogram of percentage normal exercise capacity for
age in male referral patients (reproduced from Morris et al.,[91] pulmonary function tests. Dyspnoea can also be an
with permission). METs = metabolic equivalents; y = years. anginal equivalent, but this is only likely if the rest-

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 309

ing ECG is normal, pulmonary function is normal, nose CAD. Patient safety is of utmost concern
and there is abnormal ST depression. When it is when conducting ET. Adhering to the specified
caused by pulmonary disease, there is usually a guidelines of physician supervision, emergency
history, physical examination and pulmonary func- procedures, contra-indications to ET and indica-
tion evidence of pulmonary disease. Wheezing tions for terminating a test can avoid potential haz-
during the test can also occur in the presence of ards. After considering patient safety, it is impor-
pulmonary disease. Dyspnoea of cardiac origin tant that the information obtained from ET is
usually occurs with ECG evidence of myocardial reliable and accurate. Proper patient preparation is
damage and other signs and symptoms of chronic essential. Assessing the patient’s clinical history is
heart failure. imperative when interpreting the results from ET
A final but important distinction that should be and is helpful in determining the proper exercise
made is between leg fatigue and localised leg pain protocol and modality. Information from ET may
caused by peripheral vascular disease (claudica- also help in determining other diagnostic tests to
tion). Claudication is typically leg, hip or buttocks perform. Once completed, the results from ET are
pain brought on by walking, which is relieved by best presented in the form of a test report that in-
rest and is not positional (like the similar pain of cludes patient history, interpretation of the resting
spinal stenosis). and exercise ECGs, haemodynamic responses to
ET and any signs and symptoms. This will facili-
tate the interpretation of ET. Results from ET may
22. Conclusion
also be used to develop an EP for the patient. Care
ET is an important and frequently used tech- should be taken when developing the EP, as one
nique for assessing efficacy of interventions along needs to balance the clinical needs of the patient
with prognosis and diagnosis for a number of ail- with what is realistically attainable by the patient.
ments, but it is most commonly performed to diag- An EP that is not fulfilled is of no use to the clini-
cian or patient. It is hoped that by following the
outlined guidelines that ET can continue to be safe
0
and reliable.
20
1
25
30 20 2 References
1. American College of Sports Medicine. Guidelines for exercise
35 Sedentary 30 20 3 testing and prescription. 5th ed. Pennsylvania: Williams and
40 30 4 Wilkins, 1995
40
50 40 5 2. Gibbons RJ, Balady GJ, Beasley JW, et al. ACC/AHA Guide-
45
60 50
lines for exercise testing: a report of the American College of
Age (y)

70 6
METs

50 Cardiology and the American Heart Association task force on

80 60
55 90 70
Active 7 practice guidelines (committee on exercise testing). J Am
100 Coll Card 1995; 30: 260-315
110 80 8
60 120
130 90 3. Gibbons RJ, Balady GJ, Beasley JW, et al. ACC/AHA Guide-
140 100 9
65 150 110 lines for exercise testing: executive summary. A report of the
120 American College of Cardiology and the American Heart
70 130 10
140 Association task force on practice guidelines (committee on
75 150 11
exercise testing). ACC/AHA practice guidelines. Circulation
80 12 1997; 96: 345-54
85 13 4. ESC Working Group on Exercise Physiology, Physiopathology
and electrocardiography: guidelines for cardiac exercise test-
90 14
ing. Eur Heart J 1993; 14: 969-88
15 5. Cardiac Society of Australia and New Zealand. Clinical exer-
Fig. 6. Nomogram of percentage normal exercise capacity in cise stress testing: safety and performance guidelines. Med J
sedentary and active male referral patients (reproduced from Aust 1996; 164: 282-4
Morris et al.,[91] with permission). METs = metabolic equivalents; 6. Fletcher GF, Balady G, Froelicher VF, et al. Exercise standards:
y = years. a statement for healthcare professionals from the American

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
310 Lear et al.

Heart Association [AHA medical scientific statement: special 26. Schlant RC, Friesinger II GC, Leonard JJ, et al. Clinical com-
report]. Circulation 1995; 91: 580-615 petence in exercise testing: a statement for physicians from
7. American Association of Cardiovascular and Pulmonary Reha- the ACP/ACC: AHA Task Force on Clinical Privileges in Car-
bilitation. Guidelines for cardiac rehabilitation programs. diology. J Am Coll Cardiol 1990; 16: 1061-5
Champaign (IL): Human Kinetics, 1995 27. Froelicher VF, Purdue S, Pewen W, et al. Application of meta-
8. Myers J. The physiology behind exercise testing. Prim Care analysis using an electronic spreadsheet to exercise testing in
1994; 21: 415-37 patients after myocardial infarction. Am J Med 1987; 83:
9. Wasserman K, Hansen JE, Sue DY, et al. Principles of exercise 1045-54
testing and interpretation. Philadelphia (PA): Lea & Febiger, 28. Senaratne MPJ, Hsu L, Rossall RE, et al. Exercise testing after
1987 myocardial infarction: relative values of the low-level pre-
10. Milani J, Fernhall B, Manfredi T. Estimating oxygen consump- discharge and the post-discharge exercise test. J Am Coll Car-
tion during treadmill and arm ergometry activity in males with diol 1988; 12: 141-6
coronary artery disease. J Cardiopulm Rehabil 1996; 16: 29. Parameshwar J, Keegan J, Sparrow J, et al. Predictors of prog-
394-401 nosis in severe chronic heart failure. Am Heart J 1992; 123:
11. Myers J. Essentials of cardiopulmonary exercise testing. Cham- 421-6
paign (IL): Human Kinetics, 1996 30. Mancini DM, Eisen HM, Kussmaul W, et al. Value of peak
12. Fuster V. Atherosclerosis: A. Pathogenesis, pathology, and pre- exercise oxygen consumption for optimal timing of cardiac
sentation of atherosclerosis. In: Giuliani ER, Fuster V, Gersh transplantation in ambulatory patients with heart failure. Cir-
BJ, et al., editors. Cardiology: fundamentals and practice. 2nd culation 1991; 83: 778-86
ed. Chicago (IL): Mosby Yearbook, 1991: 1172-210 31. Roberts JM, Sullivan M, Froelicher VF, et al. Predicting oxygen
13. Ross R. The pathogenesis of atherosclerosis. In: Braunwals E, uptake from treadmill testing in normal subjects and coronary
editor. Heart disease: a textbook of cardiovascular medicine. artery disease patients. Am Heart J 1984; 108: 1454-60
3rd ed. Philadelphia (PA): WB Saunders Company, 1988: 32. Sullivan M, Genter F, Savvides M, et al. The reproducibility of
1135-52
hemodynamic, electrocardiographic, and gas exchange data
14. Alexander RW. Theodore Cooper Memorial Lecture: hyperten-
during treadmill exercise in patients with stable angina pec-
sion and the pathogenesis of atherosclerosis. Oxidative stress
toris. Chest 1984; 86: 375-82
and the mediation of arterial inflammatory response: a new
33. Sullivan M, McKirnan MD. Errors in predicting functional ca-
perspective. Hypertension 1995; 25: 155-61
pacity for postmyocardial infarction patients using a modified
15. Hempel A, Maasch C, Heintze U, et al. High glucose concen-
Bruce Protocol. Am Heart J 1984; 107: 486-91
trations increase endothelial cell permeability via activation
34. Wasserman K. Exercise gas exchange in heart disease. 3rd ed.
of protein kinase C alpha. Circulation Res 1997; 81: 363-71
Armenk (NY): Futura Publishing Company Inc., 1996
16. Guretzki HJ, Gerbitz KD, Olgemoller B, et al. Atherogenic lev-
35. Pina IL, Balady GJ, Hanson P, et al. Guidelines for clinical
els of low density lipoprotein alter the permeability and com-
exercise testing laboratories: a statement for healthcare pro-
position of the endothelial barrier. Atherosclerosis 1994; 107:
15-24 fessionals from the Committee on Exercise and Cardiac Re-
17. Tawakol A, Omland T, Gerhard M, et al. Hyperhomocyst(e)in- habilitation, American Heart Association. Circulation 1995;
emia is associated with impaired endothelium-dependent va- 91: 912-21
sodilation in humans. Circulation 1997; 95: 1119-21 36. Milliken JA, Abdollah H, Burggraf GW. False-positive tread-
18. Pech-Amsellem MA, Myara I, Storogenko M, et al. Enhanced mill exercise tests due to computer signal averaging. Am J
modifications of low-density lipoproteins (LDL) by endothe- Cardiol 1990; 65: 946-8
lial cells from smokers: a possible mechanism of smoking- 37. Ribisl PM, Liu J, Mousa I, et al. Comparison of computer ST
related atherosclerosis. Cardiovasc Res 1996; 31: 975-83 criteria for diagnosis of severe coronary artery disease. Am J
19. Ross R, Glomset J, Kariya B, et al, A platelet-dependent serum Cardiol 1993; 71: 546-51
factor stimulates the proliferation of arterial smooth muscle 38. Bailey JJ, Berson AS, Garson Jr A, et al. Recommendations for
cells in vitro. Proc Natl Acad Sci U S A 1974; 71: 1207-10 standardization and specifications in automated electrocardi-
20. Stary HC. Evolution and progression of atherosclerotic lesions ography: bandwidth and digital signal processing. A report for
in coronary arteries of children and young adults. Arterioscle- health professionals by an ad hoc writing group of the Com-
rosis 1989; 9: 119-32 mittee on Electrocardiography and Cardiac Electrophysiol-
21. Ross R. The pathogenesis of atherosclerosis – an update. N Engl ogy of the Council of Clinical Cardiology, American Heart
J Med 1986; 314: 488-500 Association. Circulation 1990; 81: 730-9
22. Cheseboro JH, Zoldelyi P, Fuster V. Plaques disruption and 39. Froelicher VF. Exercise and the heart. 3rd ed. St Louis (MO):
thrombosis in unstable angina pectoris. Am J Cardiol 1991; CV Mosby Company, 1993
68: 9c-15c 40. Perloff D, Grim C, Flack J, et al. Human blood pressure deter-
23. Gibbons L, Blair SN, Kohl HW, et al. The safety of maximal mination by sphygmomanometry. Circulation 1993; 88:
exercise testing. Circulation 1989; 80: 846-52 2460-70
24. Stuart RJ, Ellestad MH. National survey of exercise stress test- 41. Myers JN. Perception of chest pain during exercise testing in
ing facilities. Chest 1980; 77: 94-7 patients with coronary artery disease. Med Sci Sports Exerc
25. Knight JA, Laubach Jr CA, Butcher RJ, et al. Supervision of 1994; 26: 1082-6
clinical exercise testing by exercise physiologists. Am J Car- 42. Hermanson L, Saltin B. Oxygen uptake during maximal tread-
diol 1995; 75: 390-1 mill and bicycle exercise. J Appl Physiol 1969; 26: 31-37

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
Guidelines for Exercise Stress Testing 311

43. Whipp BJ, Davis JA, Tores F, et al. A test to determine param- 62. Miranda CP, Liu J, Kadar A, et al. Usefulness of exercise-in-
eters of aerobic function during exercise. J Appl Physiol duced ST-segment depression in the inferior leads during ex-
1981; 50: 217-21 ercise testing as a marker for coronary artery disease. Am J
44. Hambrecht R, Schuler GC, Muth T, et al. Greater diagnostic Cardiol 1992; 69: 303-7
sensitivity of treadmill versus cycle exercise testing of 63. Diamond GA, Forrester JS. Analysis of probability as an aid in
asymptomatic men with coronary artery disease. Am J Car- the clinical diagnosis of coronary-artery disease. N Engl J
diol 1992; 70: 141-6 Med 1979; 300: 1350-8
45. Von Duvillard SP, Pivirotto JM. The effect of front handrail and 64. Allison T, Bardsley W, Behrenbeck T, et al. Cardiovascular
nonhandrail support on treadmill exercise in healthy women. stress testing: a description of the various types of stress tests
J Cardiopulm Rehabil 1991; 11: 164-8 and indications for their use. Mayo Clin Proc 1996; 75: 390-1
46. Zeimetz GA, McNeill JF, Hall JR, et al. Quantifiable changes 65. Fleg JL, Gerstenblith G, Zonderman AB, et al. Prevalence and
in oxygen uptake, heart rate, and time to target heart rate when prognostic significance of exercise-induced silent myocardial
hand support is allowed during treadmill exercise. J Car- ischemia detected by thallium scintigraphy and electrocardi-
dpulm Rehabil 1985; 5: 525-30 ography in asymptomatic volunteers. Circulation 1990; 81:
47. Smith JJ, Kampine JP. Circulatory physiology - the essentials. 428-36
3rd ed. In: Satterfield TS, editor. Baltimore (MD): Williams 66. Kotler TS, Diamond GA. Exercise thallium 201 scintigraphy in
and Wilkins, 1990 the diagnosis and prognosis of coronary artery disease. Ann
48. Myers J, Buchanan N, Walsh D, et al. Comparison of the ramp Intern Med 1990; 113: 684-702
versus standard exercise protocols. J Am Coll Cardiol 1991; 67. Kahn JK, McGhie I, Akers MS, et al. Quantitative rotational
17: 1334-42 tomography with 209Ti and 99mTc 2-methoxy-isobutyl-isonitr-
49. Haskell W, Savin N, Oldrige R. Factors influencing factors ox- ile: a direct comparison in normal individuals and patients
ygen uptake during exercise testing soon after myocardial with coronary artery disease. Circulation 1989; 79: 1282-93
infarction. Am J Cardiol 1982; 50: 299-304 68. Kiat H, Maddahi J, Roy LT, et al. Comparison of technetium-
99m methoxy isobutyl isonitrile and thallium 201 for evalu-
50. Webster MWI, Sharpe DN. Exercise testing in angina pectoris:
ation of coronary artery disease by planar and tomographic
the importance of protocol design in clinical trials. Am Heart
methods. Am Heart J 1989; 117: 1-11
J 1989; 117: 505-8
69. Taillefer R, DePuey EG, Udelson JE, et al. Comparative diag-
51. Tamesis B, Stelken A, Byers S, et al. Comparison of the asymp-
nostic accuracy of Ti-201 and Tc-99m sestamibi SPECT im-
tomatic cardiac ischemia pilot versus Bruce and Cornell ex-
aging (perfusion and ECG-gated SPECT) in detecting
ercise protocols. Am J Cardiol 1993; 72: 715-20
coronary artery disease in women. J Am Coll Cardiol 1997;
52. Panza J, Quyyumi AA, Diodati JG, et al. Prediction of the fre-
29: 69-77
quency and duration of ambulatory myocardial ischemia in
70. Armstrong WF. Treadmill exercise echocardiography: method-
patients with stable coronary artery disease by determination
ology and clinical role. Eur Heart J 1997; 18 Suppl. D: D2-8
of the ischemia threshold from exercise testing: importance
71. Blair SN. 1993 CH McCloy Research Lecture: physical activ-
of the exercise protocol. J Am Coll Cardiol 1991; 17: 657-63
ity, physical fitness, and health. Res Q Exerc Sport 1993; 64:
53. Redwood DR, Rosing DR, Goldstein RE, et al. Importance of 365-76
the design of an exercise protocol in the evaluation of patients
72. Paffenbarger Jr RS, Hyde RT, Wing AL, et al. The association
with angina pectoris. Circulation 1971; 43: 618-28 of changes in physical activity level and other lifestyle char-
54. Myers J, Froelicher VF. Optimizing the exercise test for phar- acteristics with mortality among men. N Engl J Med 1993;
macological investigations. Circulation 1990; 82: 1839-46 328: 538-45
55. Buchfuhrer MJ, Hansen JE, Robinson TE, et al. Optimizing the 73. Hambrecht R, Niebauer J, Marburger C, et al. Various intensi-
exercise protocol for cardiopulmonary assessment. J Appl ties of leisure time physical activity in patients with coronary
Physiol 1983; 55: 1558-64 artery disease: effects on cardiorespiratory fitness and pro-
56. Myers J, Buchanan N, Smith D, et al. Individualized ramp gression of coronary athersclerotic lesions. J Am Coll Cardiol
treadmill: observations on a new protocol. Chest 1992; 101: 1993; 22: 468-77
2305-415 74. King AC, Taylor CB, Haskell WL, et al. Influence of regular
57. Ho SWC, McComish MJ, Taylor RR. Effect of beta-adrenergic aerobic exercise on psychological health: a randomized con-
blockade on the results of exercise testing related to the extent trolled trial of healthy middle-aged adults. Health Psychol
of coronary artery disease. Am J Cardiol 1985; 55: 258-62 1989; 8: 305-24
58. American College of Sports Medicine. Resource manual for 75. Stern MJ, Cleary P. The national exercise and heart disease
exercise testing and prescription. 2nd ed. Pennsylvania: Wil- project: psychosocial changes observed during a low-level
liams & Wilkins, 1993 exercise program. Arch Intern Med 1981; 141: 1463-7
59. Pina IL, Chahine RA. Lead systems: sensitivity and specificity. 76. Rejeski WJ, Brawley LR, Schumaker SA. Physical activity and
Cardiol Clin 1984; 2: 329-35 health related quality of life. Exerc Sport Sci Rev 1996; 24:
60. Lachterman B, Lehmann KG, Abrahamson D, et al. ‘Recovery 71-108
only’ ST segment depression and the predictive accuracy of 77. Jones NL. Clinical exercise testing. Philadelphia (PA): WB
the exercise test. Ann Intern Med 1990; 112: 11-6 Saunders Company, 1988
61. Philbrick JT, Horowitz RI, Feinstein AR. Methodological prob- 78. Tran ZV, Weltman A, Glass GV, et al. The effects of exercise
lems of exercise testing for coronary artery disease: groups, on blood lipids and lipoproteins: a meta-analysis of studies.
analysis and bias. Am J Cardiol 1980; 46: 807-12 Med Sci Sports Exerc 1983; 15: 393-402

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)
312 Lear et al.

79. Bjorntorp, P. Effects of exercise on plasma insulin. Int J Sports 87. Pate RR, Pratt M, Blair SN, et al. Physical activity and public
Med 1981; 2: 125-9 health: a recommendation from the Centers for Disease Con-
80. Exercise and NIDDM. Diabetes 1991; 40 Suppl.: 52-6 trol and Prevention and the American College of Sports Med-
81. Hagberg JM, Seals DR. Exercise training and hypertension. icine. JAMA 1995; 273: 402-7
Acta Med Scand 1986; 711: 131-6 88. Martin JE, Dubbert PM. Exercise applications and promotion
82. Ponjee GA, Janssen EM, Hermans J, et al. Effects of long-term in behavioral medicine. J Consult Clin Psychol 1982; 50:
exercise of moderate intensity on anthropometric values and 1004-17
serum lipids and lipoproteins. Eur J Clin Chem Clin Biochem 89. Hammond K, Froelicher VF. Normal and abnormal heart rate
responses to exercise. Prog Cardiovasc Dis 1985; 27: 271-96
1995; 33: 121-6
90. Froelicher VF. EXTRA: an exercise test interpretation system.
83. Elwood PC, Yarnell JW, Pickering J, et al. Exercise, fibrinogen,
St Louis (MO): CV Mosby, 1995
and other risk factors for ischaemic heart disease. Caerphilly
91. Morris CK, Myers J, Froelicher VF, et al. Nomogram based on
Prospective Heart Disease Study. Br Heart J 1993; 69: 183-7
metabolic equivalents and age for assessing aerobic exercise
84. Myers JN, Froelicher VF. Exercise testing and prescription. capacity in men. J Am Coll Cardiol 1993; 22: 175-82
Phys Med Rehabil Clin North Am 1995; 1: 117-51
85. Pollock ML, Gaesser GA, Butcher JD, et al. The recommended
quantity and quality of exercise for developing and maintain-
ing cardiorespiratory and muscular fitness, and flexibility in Correspondence and reprints: Dr Andrew Ignaszewski,
healthy adults. Med Sci Sports Exerc 1998; 30: 975-91 Healthy Heart Program, St. Paul’s Hospital, 180-1081
86. Cobb LA, Weaver WD. Exercise: a risk for sudden death in Burrard Street, Vancouver, British Columbia, V5Z 1Y6,
patients with coronary heart disease. J Am Coll Cardiol 1986; Canada.
7: 215-9 E-mail: [email protected]

© Adis International Limited. All rights reserved. Sports Med 1999 May; 27 (5)