Exercise Stress Testing for Coronary Artery Disease
February 8, 1999
Exercise Stress Testing for Coronary Artery Disease
Author: Kevin Berman, MD, Cardiologist, Cardiovascular Diseases & Cholesterol Control Center of Dayton, Inc., Dayton, OH.
Peer Reviewer: Beverly Yamour, MD, FACC, Staff Physician of Ohio Heart Care, Canton, OH.
Editor’s Note—Patients with coronary artery disease (CAD) frequently present with intermittent chest discomfort. Unless the description of these events is classic, in terms of precipitating factors, time course, quality, location, and alleviating factors for angina, further evaluation should be undertaken. First-hand observation of the event by the physician is most helpful in diagnosing angina. Determining the threshold for provocation of symptoms and correlating it with an electrocardiogram is useful. This rationale provided the basis for the creation of exercise stress testing to determine the presence or absence of coronary disease by Goldhammer and Scherf in 1933.1 Their protocols were patient specific. Master, who devised a "two step fitness test," generalized this technique. Unfortunately, the latter was not vigorous, and more vigorous testing to detect disease at an earlier stage was devised.2 Treadmills and bicycles subsequently became the standard techniques for inducing strenuous exercise and have led to the rise of "maximal stress testing."
Indications for Noninvasive Stress Testing
There are five basic indications for exercise stress testing. The first and most common is to aid in diagnosing the etiology of chest pain in adults, where the story is consistent for, but not necessarily diagnostic of, coronary disease. There should be a reasonable risk of finding disease in the patient; however, the story may not clearly suggest angina. As a subset of this indication, in patients with known disease, stress testing may help to delineate the severity of the disease and may help guide therapy. Those at low risk may continue with observation and medical therapy. Those at high risk should undergo definite revascularization (if possible).
The next common use for stress tests is evaluation of the efficacy of therapy, be it medical or postrevascularization (percutaneous transluminel coronary angioplasty [PTCA], stent, coronary artery bypass graft [CABG], rotoblator, transluminal extraction catheter, directional coronary atherectomy). Drug therapy includes vasodilators, beta blockers, antiarrhythmics, nitrates, calcium channel blockers, aspirin, or a combination of therapies. Again, as a subgroup, some myocardial infarction stress testing can help stratify which patients can proceed to vigorous rehabilitation and which must move more slowly due to angina, arrhythmias, or impaired myocardial reserve. Additionally, patients who wish to undertake an exercise program who clinically demonstrate poor cardiac conditioning may be stratified regarding risk using stress testing.
Stress testing also helps delineate the efficacy of noncoronary cardiac surgery (repair of congenital malformations, valve repair/replacement) and peripheral revascularization. Controversy still remains as to the use of routine testing in patients with high-risk jobs (i.e., jobs in which a sudden cardiac event could lead to major loss of life and injury to others—airline pilots, truck drivers, bus drivers). Unfortunately, there is a significant false-positive rate in asymptomatic individuals. Advocates of testing say that it is better to overtest to disprove a false-positive than to risk the general public welfare. The even larger controversy is whether there is any use in "blindly" screening the general public, in whom most are asymptomatic, the evidence of disease is low, and many false-positives will be seen. Applying Bayes’ Theorem, which indicates that the prevalence of a disease is directly related to the incidence of disease in a population, seeking out occult disease in the asymptomatic public should not be undertaken. Again, if risk factors exist in an asymptomatic patient, such as strong family history for coronary disease, tobacco abuse, diabetes, hypertension, dyslipidemias, age, or gender, then stress testing in conjunction with these can be judiciously used to predict the risk of developing coronary disease. Naturally, an abnormal test in this setting offers a golden opportunity to modify risk factors to prevent or retard the development of disease.
A third and important indication for stress testing is unmasking exercise-induced arrhythmias. Finally, stress tests are useful in objectively assessing exercise tolerance. It is well known that diminished exercise tolerance may well be an early indicator of ischemia. Exercise testing also may help to objectively determine when a patient’s diseased heart valve needs to be replaced. (See Table 1.)
| Table 1. Indications for Noninvasive Exercise Stress Testing |
• Diagnosis of chest pain—pain that is suggestive but not diagnostic of ischemia |
| _____________________________________________________________________ |
Issues and Concerns
Unfortunately, the ability to correlate stress tests with coronary angiograms has been poorer than anticipated. There is agreement, however, regarding stress testing and clinical follow-up. There are good epidemiological data to explain why "blindly" screening asymptomatic patients is not as accurate as when the test is used to screen chest pain. Clearly, monitoring heart rate, blood pressure, and exercise tolerance is paramount in accurately using the results of stress testing.
Physiology
Stress testing provides an invaluable assessment of cardiovascular physiology and provides invaluable information regarding a host of cardiac parameters.
Coronary Arterial Reserve
Every organ in the human body has built-in reserve potential. This allows shifting of function from a resting state to a much higher level of function at any given time or to allow continued normal function even in the face of various disease states. In other words, even if the disease state has significantly impaired a large volume of the organ, continued normal function, even at moderately increased demand levels, is maintained.
The cardiovascular system, including the heart, arterial tree, and venous tree, is no exception. Resting states are a poor predictor of functional capacity with vigorous exercise. With exercise, there may be reversal of shunts, changes in pulmonary pressures, and changes in transvalve gradients that are missed at rest but become critical with exercise. Induction of wall motion abnormalities and left ventricular dysfunction with exercise may lead to increased shearing forces and aneurysm formation. Also important is the fact that even with diffuse, advanced coronary disease, there may be no ischemia due to redistribution of flow as well as the intrinsic properties of the blood created by coronary vasodilatation and constriction. Exercise stimulates the myocardium to demand maximal or close to maximal flow safely. To date, it is the only known way of increasing O2 demand, allowing the unmasking of even moderate impaired coronary flow.
Cardiac Output
With exercise, adult cardiac output may rise to five times as high as its basal state (5 L/min to 25 L/min). This is accomplished through skeletal muscle vasculature vasodilatation. Mean arterial pressure also rises. A 50% increase in pressure is common, resulting in increased myofibril contractility (a prime determinant of O2 consumption). Mild increases in stroke volume are seen as well. There is limited adaptation of stroke volume in the face of increased cardiac output. Thus, the burden falls to the increased heart rate. Increasing heart rate shortens the systolic ejection phase. With the shortened ejection time, increasing the rate of myofibril tension preserves normal emptying. This adaptation, unfortunately, comes at the expense of increased O2 consumption.
Myocardial O2 Consumption
Myocardial O2 consumption is directly related to myofibril tension, borderline contractile state, and rate of change of the tension. Exercise increases the rapidity of the development of the tension and the degrees of shortening in the myofibrils. Thus, O2 consumption per contraction increases with exercise. O2 consumption is dependent on heart rate, which is dependent on the degree of exercise. In non-hypertensive individuals, heart rate is the largest determinant of coronary flow with exercise. Direct measures of wall contractility cannot be practically measured, but the O2 consumption correlates well with the heart rate/pressure product (heart rate times systolic blood pressure). This index correlates with coronary perfusion in both normal and diseased states. To wit, patients with stable angina develop chest pain at a reproducible rate pressure product. Coronary flow increase is impeded with moderate-to-severe obstruction of the principal coronary arteries and, thus, exercise-induced ischemia is seen. The cornerstone of stress testing is the premise that objective and subjective evidence of exercise-induced ischemia allows the diagnostic presumption of coronary disease.
The most common etiology of exercised-induced ischemia is coronary atherosclerosis. Another cause includes pulmonary disease (failure to obtain proper oxygenation). Because of the remarkable vasodilatory capacity of the coronaries, this etiology is rare. Coronary capillaries dilate and myocardial O2 extraction, which is extremely efficient, compensates for the pulmonary desaturation. Anemia and carbon monoxide poisoning, which affect O2 transport, can also cause ischemia. Left ventricular strain and hypertrophy due to aortic stenosis may cause a consumption perfusion mismatch (even if coronaries are normal). Naturally, any of these in the face of coronary disease acts synergistically to produce a mismatch.
It is important to realize that coronary vascular supply is not the limiting factor in assessing cardiac performance. Thus, by using high-performance exercise testing and correlating it with angiography, it was proven that maximal exercise in the face of normal coronaries does not result in ischemia. Again, if the heart, the coronary arteries, and the oxygen transport are normal, standard exercise testing does not result in ischemia. It is important to understand that these "standard" exercise stress tests are based on successively higher levels of exercise. Physiologically, each level serves as a "warm-up" for the next level. If the protocol is changed and sprint-type exercising is performed (i.e., warm-up levels are eliminated) even those with normal coronaries may manifest ischemia. The explanation for this lies in progressive "recruitment" of coronary reserve. This recruitment occurs with normal coronaries. It is vital in patients with mild-to-moderate atherosclerosis to prevent ischemia at levels of less than peak exercise. Angiography done in those with exertional ischemia invariably reveals moderate-to-severe disease.4-7
Myocardial Ischemia Secondary to Diminished Coronary Flow
Just how much coronary flow must be reduced to produce ischemia is not known. Minor changes would be virtually impossible to detect in "standardized" stress tests. Often the time exercised does not unmask ischemia. With longer exercise, other factors mitigating maximal exercise (i.e., pulmonary, orthopedic, peripheral, and vascular may prevent unmasking ischemia).
Rough thresholds for ischemia can be determined. In the standard treadmill test, the change from nonischemic to ischemic occurs from one stage to the next. Each stage correlates roughly with a 40% increase in the rate-pressure product.
Coronary Atherosclerosis/Restriction to Flow
As angiography continues to improve, the ability to discern what level of stenosis correlates with clinical ischemia becomes vital as an offshoot. The degree of stenosis, below which normal health and the ability to remain symptom free, must be defined.
Gould and colleagues studied the degree of occlusion necessary to diminish coronary flow and induce "reactive hyperemia." In the resting state, at least 80% occlusion was needed to decrease coronary flow. A definable drop in inducible hyperemia occurred with 50-70% stenosis. It took at least 70% stenosis to have inducible hyperemia. Logan verified these results with post-mortem study of atherosclerotic arteries.9
The geometry of the stenosis obviously affects the degree of disruption to flow. Normal coronaries generally have an amazing degree of reserve. Even though a 50% stenosis implies disease processes at work, a potential for failure of these processes occurs, particularly with progression of decrease. These lesions rarely cause symptoms with "normal" levels of exertion.
The tests available are relatively insensitive to moderate atherosclerosis. Therefore, to enhance accuracy, maximal or near maximal exertion should be the goal of stress testing. Multiple modalities using high-quality leads are vital to detect ischemia.
Despite its insensitivity in predicting pressure coronary anatomy, stress testing is useful for unmasking ischemia. Ellestad and associates found a low level of events in "normal responders" as compared to a significantly higher level in those with a positive stress test (7% of normals had a future event in 4 years vs 46% of positives).20 Events were defined as angina, MI, and death.
Using ST segments, exercise tolerance, and maximum heart rate, McNeer verified these results.
Clinical Correlates
Chest pain, long held to be a marker of ischemia, unfortunately is not as sensitive as we would like. Chest pain is only 50% as common as ST segment depression in the face of ischemia. The presence of classic angina is presumptive evidence of coronary disease although other conditions, such as critical aortic stenosis, may also result in "non-coronary" chest pain.
One of the problems is in nomenclature. Many patients deny chest pain but will report pressure, burning, and palpitations among other symptoms. Avoidance of the term "pain" may obviate this pitfall.
Discerning the location and radiation pattern of the pain often helps to exclude or include the etiology as coronary disease. Again, there are many variations of the classic anginal pattern and if a high index of suspicion for coronary disease exists, the diagnosis of coronary disease should not be excluded.
Time of onset and time of disappearance of the pain should be noted and clearly delineated.
ST Segment Changes. The classic positive test occurs when horizontal ST depression in two anatomically contiguous leads is seen. Unfortunately, ST segment changes occur with hypertension, drugs, body habitus, and bundle branch blocks, to name a few.
Arrhythmias. Ventricular and atrial ectopy is not uncommon. Their presence in the absence of structural disease is of no consequence. Generally, an increase in ectopy with exercise, particularly if the history is compatible, is suggestive but not diagnostic for ischemia. Arrhythmias that extinguish with exercise are virtually always benign. In addition, ectopy with ST-T wave changes carry a more significant effect when trying to unmask occult ischemia.
Reduction in Exercise Tolerance. In patients with no valvular disease, no evidence of cardiomyopathy, and no evidence of myocarditis, loss of exercise tolerance is often an early sign of significant coronary disease. A drop in blood pressure (i.e., pump failure in the face of exercise), is an ominous finding strongly suggestive of severe disease. Thus, global ischemia can be unmasked. Unfortunately, regional ischemia can sometimes be missed if the normal myocardium overcompensates for the decreased myocardium and prevents an objective change in exercise tolerance. Thus, exercise tolerance must be used in conjunction with other aspects of stress testing.
Types of Exercise Testing
Exercise testing may be done in the noninvasive lab or in the cardiac catherization lab. Testing in the noninvasive lab hinges on monitoring EKG data. High-intensity exercise as described above is paramount in unmasking coronary disease.
Testing in the cath lab can be achieved by leg ergometry or hand ergometry. The former is often preclusive in the cath lab.
Isometric exercise has gained some interest. The premise is that isometric exercise increases heart rate and systemic vascular resistance and, thus, unmasking ischemia. (See Table 2.)
| Table 2. Use of Exercise Testing |
| Noninvasive Lab |
• assess for ischemia based on EKG |
| Invasive Lab |
• assess valvular gradients |
| ___________________________________________________________ |
Exercise Mode
There are two broad categories defining the method by which exercise testing occurs: physiologic/physical exercise and "chemical" exercise.
The most basic form of exercise stress testing is the treadmill test. The patients are prepared by removing hair, wax, and skin secretions from the chest wall using liquid sandpaper. The electrodes are then applied in a prescribed pattern (which will be discussed later). The patients are then able to stand on the treadmill, which is a mechanically powered machine that automatically changes the speed and angle of incline at predetermined intervals. The goal is to have the patients exercised to at least submaximal levels (85% of predicted heart rate for age), (i.e., 220-age ´ 0.85) or maximal predicted heart if possible. Serial EKGs are performed monitoring for ST-T wave changes and ectopy. The patients are frequently questioned regarding whether they feel any symptoms. Heart rate and blood pressure are serially monitored. Once the patients achieve the target heart rate (ideally at least the submaximal rate) or if they cannot go further, the test is terminated. Blood pressure, heart rate, and EKGs are monitored for six minutes into recovery—again looking for ectopy, new EKG changes, resolution of ST-T wave changes, and changes (abnormal) in heart rate and blood pressure. At any point during the test, if significant ischemic symptoms or changes in blood pressure or heart rate suggesting ischemia or incipient pump failure, or EKG changes consistent with ischemia occur, the test should be terminated. The caveat to the final condition is that the baseline EKG should have been normal and no other factors to explain EKG shifts present.
Various protocols (standardized) can be used for exercise testing. In addition, manual changes can be programmed, if needed, to allow the patients to continue and ultimately achieve their target heart rate. (See Table 3.)
| Table 3. Most Common Treadmill Exercise Protocols | |||||
|
|
|
|
|
|
|
| Bruce |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Naughton |
|
|
|
|
|
| Blake |
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| Steffield |
|
|
|
|
|
| Note: Stage 2-7 |
|
|
|
|
|
| See Bruce |
|
|
|
|
|
| __________________________________________________________________________________________________________________________________________ | |||||
To do an exercise stress test, the patients must be awake and alert. They must also be able to follow instructions. There should be a reasonable expectation that the patients can walk a suitable period to be able to achieve target heart rate. Those with unsteady gaits, neurological defects (making walking restricted), orthopedic problems (ruptured discs, osteoarthritis), or claudication (bad peripheral vascular disease) would not be suitable candidates for exercise stress testing. Encumbrances such as an immobile chest (bandages, rib fractions, recent thorax surgery) or travascular catheters (Swan-Ganz catheter, femoral arterial/venous lines, temporary transvenous pacemakers) would also not be suitable. Foley catheters, orthopedic appliances, and, in some cases, 02 may also preclude exercise testing. Patients with severe lung disease would also be poor candidates for exercise testing.
For pure EKG exercise stress testing, the baseline EKG should be normal or have no ST changes that would lead to a false-positive or false-negative test. Absolute contraindications to exercise stress tests are listed in Table 4.
| Table 4. Contraindications to Stress Testing |
• Myocardial Infarction (Incipient/acute/healing) |
| _______________________________________________________________________ |
Clinically, a positive test is defined as 1 mm horizontal ST depression in two anatomically contiguous leads or a fall of systolic or diastolic blood pressure more than 15 mmHg upsloping ST depression, symptoms, ST elevation, arrhythmias, and rate-related conduction changes suggestive but not diagnostic for ischemia.
Other forms of exercise testing (other than treadmills) include bicycle riding and variable step. Bicycle riding has the advantage of eliminating thorax problems and can be done with upper extremity indwelling catheters. Their disadvantage is the lack of familiarity that most have with this technique and the higher level of patient cooperation required. Variable step is a useful alternative for elderly or sedentary patients.
Time Course
The length and pattern of exercise is naturally determined by the purpose of the test. Different levels of intensity are required for participation in sports vs. industrial screening vs. risk stratification for high-risk professionals vs. screening for coronary disease.
The test may be fixed or varied both in terms of intensity and duration. If a homogenous population is being tested, then a single level of exercise is suitable. For heterogenous groups or stratification of a group, multiple-level testing is more appropriate.
Termination of the test occurs when either of two conditions is met: 1) further testing would be harmful to the subject, or 2) the goals desired have been obtained. If the patients, prior to achieving the target heart rate/target goals, terminate the test, this is deemed a submaximal or nondiagnostic test.
Open-ended testing refers to testing when the subjects or their reactions determine the end point. Closed-ended testing is milder and of less intense duration.
End points for open-ended testing are heart rate (fixed), heart rate variable (i.e., 90% of maximal heart rate), exercise to symptom leveled maximal endurance, or physiologically proven maximal aerobic capabilities.
Exercise to a Fixed Heart Rate
The underlying premise here is that 10 poorly conditioned individuals or those with cardiac disease (be it coronary, valvular, cardiomyopathy) will achieve the prescribed rate faster than healthy individuals. The disadvantage to the test is it treats all age groups equally. Clearly, older individuals will achieve the prescribed heart rate at a much higher percentage of their maximal capacity. This test optimally tests within the same age range. (See Table 5.)
| Table 5. Target Heart Rate for Graded Exercise Heart Rate | ||||||||
| Age (years) Pred. Max HR: |
30
|
35
|
40
|
45
|
50
|
55
|
60
|
65
|
| Men |
193
|
191
|
189
|
187
|
184
|
182
|
180
|
178
|
| Women |
190
|
185
|
181
|
177
|
172
|
168
|
163
|
159
|
| TargetHR: (90% Max) | ||||||||
| Men |
173
|
172
|
170
|
168
|
166
|
164
|
162
|
160
|
| Women |
171
|
167
|
163
|
159
|
155
|
151
|
147
|
143
|
| ____________________________________________________________________________________________________ | ||||||||
Variable Heart Rate Targets
The prime advantage here is that the individual does not exercise to all-out maximal capacity. The sensitivity of the test remains intact. With age, the highest maximal heart rate attainable decreases. Also, among younger subjects, the maximal heart rate attainable varies with athletic (cardiac) conditioning. There is also an intergender variation in the fall in heart rate with age. The test is adjusted for age, sex, and physical training. As an example, one may choose 90% of predicted heart rate as target.
Subjective Maximal Exercise
Patients exercise to their maximum subjective capacity. End points include pain, fatigue, and dyspnea. The disadvantage here lies in what the subjects perceive as their maximal level of exercise.
Maximum Aerobic Activity
Oxygen uptake and use are monitored throughout the test. Maximal aerobic capacity determines the end of the test. This is a cumbersome test suited for the physiological lab but poorly suited for general use.
Determining the Protocol
All protocols share some common threads. An appropriate history and physical should precede all testing. A fasting state or semifasting state is required. At the very least, a light meal should be eaten no later than two hours prior to testing. This prevents the possibility of stomach upset and nausea. More important, blood flow to the heart is not factitiously lower by increasing flow to abdominal organs. Informed consent is obtained. A baseline EKG and relevant laboratory studies should be reviewed to rule out a contraindication to testing.
Age, body habitus, orthopedic and vascular factors, lung disease, and cardiac conditioning help determine the most suitable protocol.
Exercise Electrocardiography
Eithoven performed the first exercise stress test by recording an EKG on a patient after the patient ran up a flight of stairs.12 Since that time, significant refinement of the technique has occurred. Masters created a step-wise protocol for stress testing by standardizing the test. Improvisation and interpolation of the data were eliminated.
ST Depression/Myocardial Ischemia
It was Feel who first demonstrated the relationship between angina and ST depression. The ST depression was of new onset with the beginning of chest pain and lasted throughout the episode. Three basic forms of ST depression are of clinical significance. (See Table 6.)
| Table 6. Types of ST Segment Depression |
• Transient ST depression that resolves within one minute of termination of exercise |
| ______________________________________________________________________________________________________________________ |
Practical Aspects in Exercise Stress Testing
It is of vital importance that the electrode skin interface be maximized to eliminate motion artifact and allow maximal accuracy in recording the EKG to allow maximal accuracy in interpreting the changes seen.
There is some controversy as to the number of leads required. Some use a single bipolar lead using three electrodes. Mason and colleagues all feel a single lead may miss ischemic changes.15 At present, the best accepted and most used is the Mason-Likar 12-lead set using the torso and limbs.
Etiology of Falsely Abnormal EKG Tracing
It is beyond the scope of this article to discuss all the causes of false-positive stress testing but some of the more common and clinically relevant will be discussed here.
Elevated Left Ventricular Diastolic Pressure. Pressure overload of the ventricle most commonly occurs with systemic arterial hypertension or with outflow tract obstruction (idiopathic hypertrophic subaortic stenosis [IHSS], aortic stenosis, etc.). Even if atherosclerosis is absent, subendocardial ischemia may be seen. Lepeschkin and colleagues, while studying college students with hypertension, demonstrated exercise-induced ST depression.16-17
One of the other caveats in patients with hypertension is that they are frequently treated with diuretics. Hypokalemia is common and this in and of itself causes ST-T wave abnormalities.
Barlow’s Syndrome (murmur, chest pain, ventricular ectopy) shows chronic ST-T wave changes even in the absence of coronary disease.
Bundle branch blocks and paradoxical septal activation (Pacemaker, prior cardiac surgery) or any deviation from the normal activation sequence of the ventricles can lead to ST-T wave changes. Such deviation precludes accurate interpretation of the EKG changes seen with exercise. Left bundle branch block, right bundle branch block, interventricular conduction delay, and Wolff-Parkinson-White can all lead to false-positive stress tests.
Drug Effects. Digitalis, a common cardiac drug, frequently leads to false-positive testing even if the baseline EKG does not show "classic" digitalis effect. In addition, adequate levels of digitalis may preclude patients from obtaining their target heart rate. Lastly, digitalis may mask the forme fruste of certain arrhythmias.
Tricyclic antidepressants as well as many other CNS active drugs can lead to false-positive tests, particularly in women.18 Certain antihypertensives are known to cause false-positive exercise stress tests. Most notably are methyldopa and the potassium "wasting" diuretics. Nitrates, as well as other vasodilatory drugs, allow higher thresholds before ischemia occurs. Beta blockers, a cornerstone in treating ischemic heart disease, often obfuscate the results of stress testing. Beta blockers are more likely to produce false-negative test-blunting of heart rate and antihypertensive effects would have negative test results although significant CAD is present. By recoupling the oxygen supply ratio and by their very properties (antihypertensive and antiarrhythmic), they may lead to a false-positive test.
Mitral Valve Prolapse. Mitral valve prolapse (MVP) is a not uncommon cause of chest pain and arrhythmias, particularly in younger women. Whether due to exercise-induced changes in the mitral apparatus causing a change in electric vectors or basic hormonal changes or a combination, MVP frequently causes ST-T wave changes compatible with ischemia even in the absence of coronary disease.
Interpretation of Study
Clinically, a positive stress test occurs when 1 mm of horizontal ST depression in two anatomically contiguous leads is seen. The deeper the depression and the longer it lasts, the more likely it is that the subject has coronary disease. This assumes none of the aforementioned etiologies for false-positives is present.
EKG responses in men are more likely to correlate with true coronary disease than in women. The most likely explanation for this lies in the increased prevalence of coronary disease in men. With increases in smoking, women in the workforce being exposed to environmental toxins, and the aging of the population (loss of protective effect of female hormones), there has been some decline in the gender gap and stress testing in women often correlates with coronary disease.
Aside from assessing the electrocardiograph data, other parameters must be assessed and evaluated. Blood pressure response, heart rate response, development of arrhythmias, and development or worsening of murmurs are all important in assessing the patient as discussed above. Evaluation of exercise tolerance is also vitally important.
As an interesting note, ST-T wave change correlated with coronary disease in 64% of patients with coronary disease in Cole and Ellestad’s study. If exertional angina accompanied these changes, the true positivity of the test exceeded 85%.19
A common physiologic variant on EKGs is J point depression as long as the "depression" does not last more than 0.08 seconds after the initial deflection. This is of no clinical relevance. Similarly, in the absence of other clinical or electrocardiographic evidence of ischemia, minor (i.e., less than 1 mm) or upsloping ST depression do not imply ischemia. If any doubt remains after stress testing, imaging with a nuclear agent in conjunction with stress testing should be performed.
Mathematical Evaluation of Stress Testing
Bayes’ Theorem states that the predictive value of a test is influenced by its sensitivity and the prevalence of a disease in the population tested. Thus, testing asymptomatic individuals with a low risk factor profile often leads to false-positive tests. Patients with ST depression with mild exercise have a greater than 50% chance of having or developing coronary disease. Those with ischemia at moderate exercise levels have a less than 50% chance. Ischemia only with strenuous exercise confers a less than 20% chance of having or developing disease.20 (See Table 7.)
| Table 7. Common Statistical Definitions | ||
| 1. Sensitivity |
|
True positives
Total positives |
| 2. Specificity |
|
True normals
Total normals |
| 3. Accuracy |
|
True positives + true negatives
Total number of tests |
| 4. True-positive |
|
Patient actually has disease predicted by test |
| 5. True-negative |
|
Patient is normal when test predicted normalcy |
| 6. False-positive |
|
Patient disease predicted by test |
| 7. False-negative |
|
Test fails to document disease in patient who actually has the disease |
| 8. Predictive value |
|
True positive
True + false positives |
| 9. Relative risk (risk ratio) |
|
Disease rate in patients with positive test
Disease rate in patients with negative test |
| __________________________________________________________________________ | ||
Other Forms of Stress Testing
All of the preceding discussion regarding stress testing by exercise is relevant to the following discussion. Sometimes it is not possible to simply perform an exercise stress test and be able to rely on its results. In addition, sometimes the patient simply can’t exercise. In the latter case, other forms of inducing "myocardial stress" must be used.
Nuclear Imaging as an Adjunct to Stress Testing (Thallium, Sestambi, Cardiolyte, etc.). If the patients are otherwise able to exercise but have baseline EKG abnormalities that preclude accurate assessment of the ST-T wave changes, nuclear imaging will help to clarify the picture. Nuclear imaging can be done in one of two ways. In the first case, the patients do a "routine" stress test. At the peak of exercise, thallium or the chosen material is injected. One more minute of exercise is performed and the test is terminated. The patients are previously advised that they must give the observer one minute of warning before they must "terminate" the test. That final minute is important for circulating the tracer. Stress and rest images are obtained. Alternatively, the resting part may be done first and the stress images done later if the appropriate nuclear material is chosen. Nuclear tracers are inert and go where blood supply goes. A defect at stress levels that persist at rest usually connotes a scar. A defect that "fills" in at rest connotes ischemia.
With any test, some disadvantages are present. A fixed lesion may not reperfuse or balance ischemia (e.g., triple vessel disease may be missed). Motion artifact, breast artifact, patient cooperation, and inability to exactly reproduce the same "angle" of examination with rest and stress images may cause false-positive or false-negative tests. Nonetheless, thallium and other nuclear materials distinctively enhance the accuracy of stress tests.
Chemically Induced Stress. If the patients are unable to exercise, chemically induced myocardial stress may be used. Two basic forms are used: 1) cyclic AMP blockers; and 2) dobutamine.
Drugs, such as persantine and adenosine, block cyclic adenosine monophosphate (C-AMP). C-AMP is responsible for vasomotor tone. If this is blocked, vascular beds dilate (arterial bed). Normally, arteries dilate whereas abnormal ones dilate not as well or not at all. A "steal" phenomenon is created and "ischemia" is seen. The drug is infused over 4-6 minutes. At three minutes, thallium is injected. "Stress" and "rest" pictures are taken and interpreted as above. Serial EKG and BP recordings are made. Since heart rate and BP rarely increase, it is rare to see significant EKG changes.
This form of testing is not physiologic since the heart rate and BPs rarely increase to a significant degree. Exercise tolerance cannot be assessed. Many drugs such as caffeine, xanthines, etc., must be stopped 24-48 hours prior to the test. This may not be practical in patients admitted for 23-hour observation with atypical chest pain. The most significant contraindication is in patients who are dependent on xanthines or with bronchospastic disease. Blocking C-AMP blocks its smooth muscle relaxant effect in bronchioles and may precipitate an attack. C-AMP blockers induce a plethora of effects, not the least of which is chest pain. They correlate poorly with true disease.
The effects of C-AMP blockers can be reversed by administration of theophylline. Despite its drawbacks, C-AMP blockers to induce myocardial stress are important in the clinician’s armamentarium. It is particularly useful in patients whose significant increase in heart rate and blood pressure is not desired and in those with significant baseline arrhythmias.
Adenosine may induce high-degree AV block and should be used cautiously in those with baseline AV nodal disease.
The second important drug used for chemical stress testing is dobutamine, a beta agonist. Dobutamine is infused sequentially to a maximum of 40 mcg/kg/mn to achieve the target heart rate. One to two mg of atropine may also be administered. Sequential blood pressure and EKG recordings are performed. At the maximum infusion or when the target heart rate is achieved, nuclear imaging is performed and standard interpretation is undertaken. The EKG changes are just as valid with dobutamine as with exercise. Thus, ST-T wave changes can be construed to mean ischemia if they meet the heretofore described criteria. This is a much more physiologic test than C-AMP blockers since heart rate and blood pressure rise, as does O2 demand. Arrhythmias induced by dobutamine, however, lose a lot of significance since dobutamine itself induces arrhythmias.
Patients with significant borderline hypertension or baseline arrhythmias may not be suitable for dobutamine. This is, however, an excellent drug in patients who can’t exercise or where C-AMP blockers can’t be used (particularly COPD patients who have bronchospasm and are xanthine-dependent).
Stress Echocardiography. In patients where the baseline EKG is abnormal and where nuclear imaging is equivocal, stress echocardiography may be invaluable. A baseline echocardiogram is performed. The patient is then exercised on the treadmill. As above, if the patient cannot walk, a chemically induced myocardial stress using dobutamine can be used. Serial EKGs, BP, and HR recordings are made. An echocardiogram is obtained at each stage.
If the patients develop a "regional wall motion abnormality," there is excellent evidence of ischemia. In addition, exercised-induced exacerbation of baseline "minor" valvular abnormalities can be seen. Often, this helps to unmask a chemically induced valvular lesion (mitral stenosis) or an otherwise hard-to-diagnose dynamic valvular lesion that is causing the patient’s symptoms.
The downside to stress echocardiography is that it is labor intensive. Also, the interpreter has to be practiced in assessing wall motion changes—some of which are subtle. The test also assumes that adequate echo images are obtainable. Poor echo windows due to body habitus, among other factors, may preclude use of this modality.
Other Modalities. Stress multigated angiogram (MUGA) scans (exercise, chemical), stress PET scanning, and any combination of the above modalities may be used to unmask ischemic heart disease. Stress MUGA is performed with exercise or chemically induced stress. Baseline and stress MUGA scans are obtained looking for acquired wall motion abnormalities at the peak of "exercise." Stress PET scanning involves chemically induced stress looking for metabolic abnormalities. The goal is to discern scar from hibernating myocardium from ischemic myocardium.
Safety of Stress Testing
The goal of any test is to obtain accurate, clinically useful information in the safest setting possible. We have previously discussed the contraindications to stress testing. (See Table 4.)
The stress test should be terminated when either the patient subjectively wishes to terminate the test or the desired goals have been achieved. (See Table 8.)
| Table 8. Indications for Stopping Stress Test |
• Attainment and maintenance of target heart rate for 1 minute |
| _________________________________________________________________________________________________________________________________________ |
As with any test, awareness of, anticipation of, and the ability to treat complications is paramount. The appropriate equipment, including airway management materials, defibrillators, drugs, and a staff well versed in resuscitation must be available. Complications include:
• Arrhythmias (supraventricular, atrial fibrillation/flutter, and atrial tachycardia) and ventricular (PVCs, couplets, ventricular tachycardia)
• AV nodal block
• Vasovagal reactions
• Circulatory failure
• Syncope
• Refractory chest pain, ischemia, and cardiac arrest.
With appropriate vigilance, the more ominous complications as well as the more minor ones can be avoided or, when they occur, can be appropriately treated.
The incidence of complications remains rare. In a large study, Rodimes and Blackburn21 documented 16 deaths and 40 patients requiring hospitalization. The calculated mortality was 0.01% and morbidity was 0.024%.
Stress Testing vs. Other Measures of Cardiac Ischemia
Previous noninvasive testing sought to reproduce the conditions that precipitated angina. Patients were subjected to stress, heavy meals, cold air, and cold water. Reduced ambient O2 and ergonovine were also used to induce symptoms. The latter two were used to assess coronary spasm. Unfortunately, these two methods can in and of themselves cause spasms and were not clinically useful. Hand grip testing has been used. All of the above have a low sensitivity and specificity when compared to standard stress testing.
Several invasive tests are used for assessing the patient for coronary disease. All have the drawback of being invasive tests.
Atrial pacing has been used in those who can’t perform treadmill tests. The advantage of pacing is that it obviates the need for drugs. The disadvantage is that blood pressure is not raised and even the maximal heart rate response is blunted (even if the absolute number is achieved) since exercise-induced catacholamine discharge does not occur. The aforementioned factors diminish the sensitivity of the test.
Coronary Angiography. These tests definitively document the extent and nature of the stenosis. Unfortunately, the assessment of the degree of stenosis is subjective. In addition, the geometry seen on angiograms may not accurately correlate with the true intravascular geometry (inadequate visualization of arteries distal to stenosis, eccentric lesions, etc.). In addition, what angiographically correlates with a clinically relevant lesion is not always well defined. Many patients with a noncritical lesion (< 50%) suffer an event, whereas those with "tight" lesions do not suffer an event.
Contrast Ventriculography. Injection of contrast in the ventricle during angiography may help define valvular lesions as well as segmental abnormalities. Again, this is an invasive test and interpretation is subjective.
Ambulatory (Holter) Monitoring. EKG monitoring looking for ST segment changes (particularly in patients with presumed silent ischemia) may unmask coronary disease. The advantage is that the patients are "stressed" in their typical environment. The disadvantages include the cost and the time involved in interpretation. This test does seem to be useful in unmasking Prinzmetal’s angina, which is often missed on routine stress testing.
Exercise scintigraphy with thallium appears to be a promising technique. One study quoted a 93% sensitivity compared with a 40% sensitivity for EKG stress testing. The relative specificity compared to angiography was 86%, which, if used in the appropriate subpopulation, will be a valuable addition to the clinician’s armory.
Conclusion
Stress testing in all its various forms, if applied and appropriately interpreted, is a safe, effective method for assessing the appropriate population for coronary disease. As the techniques continue to evolve, the sensitivity and specificity will improve, making this an invaluable clinical tool.
References
1. Goldhammer S, Scherf D. Ascher Kim Med 1933;122:134.
2. Rowell LB, et al. Am Heart J 1965;80:461.
3. Borer JS, et al. N Engl J Med 1975;292:367.
4. Mason RE, et al. Circulation 1967;36:517.
5. Roctman D, et al. Ann Intern Med 1970;72:641.
6. Ascoop CA, et al. Am Heart J 1971;82:609.
7. Bartel AG, et al. Circulation 1974;49:348.
8. Sheffield LT, Roctman D. Chest 1973;63:327.
9. Logan SE. IEEE Trans Biomed Eng 1975;22:327.
10. Eclestad MH. Stress testing: Principles and practice. Philadelphia: FA Davis Co., 1975, p. 175.
11. Wrener DA, et al. Circulation 1976;54/(Suppl II) II-10.
12. Einthoven W. Arch Ges Physical 1908;122:517.
13. Master AM, et al. Am Heart J 1942;24:777.
14. Fecl H, Siegel ML. Am J Med Sc 1928;175:255.
15. Mason RE, Likar I. Trans Am Clin Clematol Assoc 1964;76:40.
16. Roitman D, et al. Ann Intern Med 1970;72:641.
17. Lepeschkin E, Surawizc B. N Engl J Med 1958;258:511.
18. Linhart JW, et al. Circulation 1974;50:1173.
19. Cole JP, Ellestad MH. Am J Cardiol 1978;41:227.
20. Ellestad MH, Wan MK. Circulation 1975;51:363.
21. Rodimes P, Blackburn H. JAMA 1971;217:1061.
You have reached your article limit for the month. Subscribe now to access this article plus other member-only content.
- Award-winning Medical Content
- Latest Advances & Development in Medicine
- Unbiased Content