Coronary artery disease (CAD) has remained the number one killer of American adults since 1900.1 One coronary event occurs every 29 seconds, and one coronary fatality every minute.
About 6.4 million Americans have angina pectoris. A larger number present for evaluation of chest pain. More than 12 million Americans have a history of myocardial infarction (MI), angina pectoris, or both.
The magnitude of the problem is evident from Medicare data. In 1998, more than 2 million cardiac stress tests were performed, in a predominantly elderly population, at a cost of more than $1 billion2; this likely represents just a fraction of all cardiac stress tests performed annually. In the same year, 1.3 million cardiac catheterizations, 540,000 angioplasty procedures, and 553,000 coronary artery bypass graft (CABG) operations were performed. Total direct and indirect costs of CAD in 2001 are estimated at more than $100 billion.
Cardiac stress testing is an important tool in the evaluation of patients with known or suspected CAD. It is frequently used to diagnose CAD in patients with symptoms that suggest ischemia, and may also be used to assess prognosis or to evaluate the response to antianginal therapy in those with established CAD. Test results should not be viewed dichotomously (ie, as positive or negative), but are best interpreted within the context of the patient's clinical status.
Cardiac stress imaging has become increasingly sophisticated: modalities such as stress echocardiography and nuclear scintigraphy can be used to assess ischemic burden and myocardial viability and to guide revascularization strategies. Nevertheless, standard noninvasive cardiac stress testing by means of the exercise treadmill test (ETT) still provides a wealth of clinical information. Although newer imaging modalities are generally more sensitive and specific than exercise electrocardiography, the cost is significantly higher. Select the test that best suits the patient's clinical profile, and also take into account the strengths and shortcomings of particular types of imaging at the testing facility.
Here we address the questions most frequently asked about the principal types of cardiac stress tests used to evaluate ischemia and highlight their advantages and limitations.
What are the clinical indications for cardiac stress testing?
Stress testing is performed pri- marily to detect CAD in patients with chest discomfort, but it may be useful in other settings (Table 1). It is mandatory in the evaluation of patients entering cardiac rehabilitation and for post-MI risk stratification in patients who have not undergone coronary angiography. Other indications include evaluation of the hemodynamic significance of valvular heart disease and assessment of the need for cardiac transplantation.
An important but incompletely defined application of cardiac stress testing is in the evaluation of preoperative risk. The lack of randomized controlled trials in this area makes it difficult to choose an optimal strategy. Identification of the appropriate candidates for a preoperative stress test is beyond the scope of this review, but guide- lines have recently been published in an American College of Cardiology (ACC)/American Heart Association (AHA) consensus statement.3 A history of CAD should not by itself mandate preoperative stress testing. However, patients with CAD symptoms or impaired exercise tolerance who undergo major surgery may benefit from preoperative stress imaging. The decision to screen patients before surgery generally requires the consideration of clinical markers (eg, angina or heart failure), functional capacity, and surgery-specific risks.
What is the role of noninvasive cardiac stress testing?
2In appropriately selected patients, ETT is safe, effective, and inexpensive. MI or death associated with ETT occurs at a rate of less than 1 per 2500 tests.2Table 2 lists relative and absolute contraindications to standard exercise electrocardiography; these are generally similar to those of other methods of cardiac stress testing.
Findings that suggest ischemia. The most widespread criterion used to diagnose ischemia with standard ETT is exercise-induced horizontal or downsloping ST-segment depression of greater than 1 mm. The risk of significant CAD increases markedly with the degree of ST-segment depression. The presence of anginal symptoms in conjunction with ST-segment depression further enhances the accuracy of the study.
Other findings on ETT associated with a high risk of an adverse outcome include:
ST elevation in leads without pathologic Q waves.
Onset of ST depression within the first 5 minutes of exercise.
ST depression greater than 2 mm in multiple leads.
ST depression that persists longer than 5 minutes into the recovery (postexercise) phase.
Clinical features that increase risk and affect prognosis include early onset of angina, exercise-induced hypotension, and sustained ventricular tachycardia. These features often predict left main or multivessel obstructive CAD. Finally, absolute exercise tolerance-expressed as metabolic equivalents (METS)-is strongly correlated with prognosis. A recent analysis of more than 6000 men referred for ETT found that maximal exercise capacity was the single best clinical predictor of mortality.4
Limitations of ETT. Standard ETT is often suboptimal in the evaluation of some patients with known or suspected CAD. Certain abnormalities on the baseline ECG (eg, left bundle- branch block [LBBB], left ventricular hypertrophy with strain pattern, or significant ST depression) limit the specificity of standard ETT. In these instances, stress imaging with either nuclear perfusion scintigraphy or echocardiography is required to establish the diagnosis. The clinical and ECG settings in which stress imaging is preferable to standard ETT are outlined in Table 3.
A number of medications may also limit the diagnostic performance of ETT. In one study, digoxin produced abnormal (false-positive) ST-segment depression in up to 40% of healthy patients free of CAD.5 β-Blockers and nondihydropyridine calcium channel blockers (eg, verapamil and diltiazem) significantly blunt heart rate response to exercise. Whenever possible, these agents should be withheld for 48 hours (about 4 half-lives) before the initial evaluation of patients with suspected ischemic symptoms to increase the probability that the exercise study will be diagnostic. Dihydropyridine calcium channel blockers (eg, amlodipine and felodipine) and nitrates may reduce the sensitivity of stress testing.
In patients with established CAD, however, it is often most prudent not to discontinue antianginal therapy (particularly β-blockers) in order to avoid a withdrawal phenomenon that may lead to adverse cardiac ischemic events.6 Moreover, a major goal of stress testing is to ascertain the efficacy of antianginal therapy in patients with established CAD.
Other forms of provocative cardiac stress testing may be used for patients incapable of exercising on a treadmill. Although not available in every cardiovascular laboratory, bicycle ergometer devices may be helpful for older persons who have difficulty maintaining balance on a treadmill. More often, pharmacologic stress agents are used as alternatives to exercise. These include vasodilators (eg, adenosine and dipyridamole) and positive inotropes (eg, dobutamine). Dobutamine is preferred in patients with reactive airways disease because of the small risk of bronchospasm associated with vasodilators. Dobutamine protocols are generally more labor-intensive and require the presence of a nurse who is proficient with automated infusion pumps. Each of these alternate stress techniques must be done in conjunction with either echocardiography or nuclear scintigraphy because ECG data obtained from pharmacologic stress testing provide less meaningful physiologic and functional information than does exercise testing.
Time to onset of angina or ST-segment depression, maximal heart rate and blood pressure response, and total exercise duration are a few of the key data elements that are derived only from ETT and provide important prognostic information.7 Overall, the sensitivity and specificity of standard ETT for the diagnosis of CAD is 68% and 77%, respectively.8
What other noninvasive cardiac stress tests are used routinely?
3Nuclear perfusion scintigraphy. Many nuclear medicine laboratories use single photon emission CT (SPECT) imaging techniques rather than older planar imaging modalities. This allows the radiologist or cardiologist to view the heart in multiple anatomic slices, thereby providing more accurate localization of ischemia or infarct. Ischemia is diagnosed when a myocardial segment demonstrates normal perfusion at rest but shows hypoperfusion on reimaging shortly after stress. In contrast, infarction is defined by hypoperfusion at both stress and rest.
Although the radioactive tracer thallium-201 has been most widely studied, newer agents, such as technetium-99 sestamibi and technetium-99 tetrofosmin, are becoming the agents of choice in the United States. Despite the higher cost of technetium-based agents, they have 2 major advantages compared with thallium-201. First, much larger doses of tracer can be safely administered, which yields higher radioactive count density (brighter images) with less scatter and attenuation.9 Second, technetium-based agents allow for gated acquisition, which permits the simultaneous evaluation of both perfusion and cardiac motion. Gating techniques facilitate assessment of global systolic function (left ventricular ejection fraction) and regional evaluation of cardiac wall motion. Gating and other features (attenuation correction and prone imaging) have improved the accuracy of perfusion scintigraphy.
The average sensitivity of SPECT imaging for detecting ischemia is appreciably higher than that of standard ETT-about 88%; specificity is similar to that of ETT.8 As with standard ETT, however, the sensitivity of nuclear scintigraphy is diminished in the setting of single-vessel disease (particularly the right or left circumflex coronary artery), inadequate heart rate response to stress, and antianginal therapy.9
Geographic localization of ischemia with nuclear imaging is helpful in symptomatic patients with known CAD, particularly those who have undergone CABG surgery. In addition, nuclear imaging (with thallium-201) may be used to determine myocardial viability in CAD patients with previous MI or significant left ventricular systolic dysfunction. The information gained about viability may guide revascularization strategies in these high-risk patients.
Nuclear perfusion testing can be performed with either exercise or pharmacologic agents. Exercise is the preferred method of provocative stress, with one exception: in the presence of LBBB, the use of exercise nuclear perfusion imaging is associated with a high rate of false-positive findings of defects involving the interventricular septum.10 In this setting, perfusion imaging coupled with pharmacologic vasodilation is more specific in determining whether CAD is present,11 although the reasons for this are unclear.
It is worth noting that all cardiac stress testing modalities are limited to the detection of obstructive coronary artery stenoses. Newer modalities, such as CT and MRI, may be helpful in identifying patients with early atherosclerosis, although their role in current clinical practice is not yet fully defined.
The major advantage of nuclear stress imaging modalities compared with standard ETT is greater accuracy in detecting CAD, particularly when the resting ECG is abnormal. Nuclear perfusion imaging also provides higher sensitivity, better characterization of the extent of myo- cardial ischemia, and a direct measurement of left ventricular systolic function.
Stress echocardiography. Like nuclear scintigraphy, stress echocardiography detects ischemia by visualizing the heart at rest and immediately after stress. Ischemia is said to be present when ultrasound images of the heart after stress identify a wall-motion abnormality in a segment that contracted normally at rest. In this way, stress echocardiography is able to localize ischemia to a specific vascular territory, much like nuclear imaging. Stress echocardiography also provides the opportunity to screen for significant valvular disease during the baseline (rest) echocardiogram. Another advantage is that images can be viewed immediately by the interpreting cardiologist (in "real time"), whereas nuclear imaging requires longer acquisition and processing times. Like thallium-201 and positron emission tomography scanning, dobutamine echocardiography is useful in the assessment of myocardial viability.
As with nuclear imaging, the stress portion of the examination may be provoked pharmacologically or via exercise protocols. Dobutamine is generally preferred to vasodilators because a graded infusion protocol allows for attainment of target heart rate (85% of age-predicted maximal heart rate: 0.85 × ([220 − patient's age]). Recent data suggest that stress echocardiography is a useful predictor of future coronary events in women,12 and may be a cost-effective approach in this group compared with standard ETT as well.13 The overall sensitivity and specificity of stress echocardiography is approximately 76% and 88%, respectively.8 Compared with nuclear stress imaging, stress echocardiography is less sensitive but may be more specific for the diagnosis of ischemic heart disease.
Which factors guide the selection of cardiac stress tests?
4Two crucial factors are the accuracy of the test and the clinician's assessment of the patient's pretest probability of CAD. According to the Bayes theorem, the predictive value of any test depends on both the test's accuracy and the prevalence of the disease (ie, the pretest probability of disease). A sensitive test has few false-negative results, and a specific test has few false-positive results. In patients with a very high likelihood of CAD (more than 90%), noninvasive testing often provides useful prognostic information; however, negative or nondiagnostic results do not necessarily preclude coronary angiography.
In persons with a very low likelihood of CAD (less than 10%), it is often best to explore a noncardiac diagnosis rather than order a cardiac stress test. A negative result would merely confirm a low probability of disease, and a positive result might not increase the likelihood of disease enough to make a clinical difference.2 Furthermore, a false-positive result may cause anxiety in the patient and could lead to invasive testing, which carries a finite risk. Therefore, noninvasive stress testing is most helpful when the pretest probability of obstructive CAD is in the intermediate range-roughly between 10% and 90%. For patients with angina at low workload and a greater than 90% likelihood of CAD, consider proceeding directly to cardiac catheterization.
Which modalities are preferred for the evaluation of chest pain in women?
5Although CAD is the leading cause of death in both men and women,1 the overall prevalence of CAD in persons younger than 55 years is uniformly lower in women than in men. This sex difference disappears in the elderly; however, young women with unexplained chest pain present a diagnostic challenge. This group is less likely to have significant atherosclerotic CAD and more likely to have false-positive results on ETT. In fact, evidence suggests an inherently lower sensitivity and specificity of ETT in women.14,15 Hence, standard ETT provides less diagnostic accuracy in women than men.7 Some clinicians therefore believe that a stress-imaging study should be the test of choice for the evaluation of chest pain in women.
However, despite the limitations of ETT in women, the ACC/AHA/American College of Physicians-American Society of Internal Medicine (ACP-ASIM) Committee for the Evaluation of Chronic Stable Angina does not recommend the routine initial use of stress imaging simply on the basis of female sex.2 In the case we describe in the Box, the patient had none of the ECG or clinical features that would warrant stress imaging. Furthermore, evidence suggests that a simple risk prediction instrument, the Duke Treadmill Score, can enhance the prognostic value of ETT without the use of imaging techniques.16-18 The Duke score is a weighted index that combines ST-segment deviation, treadmill time, and degree of angina. This instrument provides equivalent diagnostic and prognostic ability in both women and men19 and has been included in the ACC guidelines for exercise testing.7
What other clinical considerations are relevant in the assessment of patients with suspected myocardial ischemia?
6A key issue is the patient's ability to ambulate to an adequate workload (usually 85% of his or her age-predicted maximal heart rate). If this is unlikely, consider ordering a pharmacologic stress study. Instruct the patient to avoid methylxanthines (caffeine or theophylline) for 24 to 48 hours before the study. Methylxanthines antagonize the vasodilatory effects of adenosine and dipyridamole. Withholding caffeine affords you the option of immediately converting an aborted standard ETT into a more meaningful vasodilator study.
Advanced age. Elderly patients are increasingly referred for cardiac stress testing and present a particular challenge. Their functional capacity is often limited by deconditioning, poor balance, and muscle weakness. Special attention must therefore be given to the mechanical hazards of exercise. The possibility of using a less challenging exercise protocol may also be considered.20 As mentioned earlier, a bicycle exercise test may be an option in patients with gait and coordination difficulties, but this form of exercise is often unfamiliar to elderly patients.21 Therefore, in frail elderly patients or those with a physical handicap, a reasonable approach may be to start with a vasodilator stress test. Finally, note that advanced age is associated with a higher pretest probability of multivessel CAD. This phenomenon increases the sensitivity but may also decrease the specificity of exercise testing in the elderly because of the greater prevalence of left ventricular hypertrophy in these patients.2
Obesity. An increasingly common challenge in noninvasive cardiac stress testing is the evaluation of very obese patients. These persons often have musculoskeletal disabilities and significant deconditioning that prevent them from being able to exercise adequately. Although stress imaging may seem preferable, it can be technically difficult in obese patients. Stress echocardiography images tend to be of lower quality in these persons because of the difficulty in visualizing endocardial borders; this may prevent adequate assessment for wall-motion abnormalities. This limitation has been partially remedied by the advent of tissue harmonic imaging and intravenous contrast agents that improve endocardial border detection.22
Even nuclear scintigraphic images are often suboptimal in the very obese because of significant radioactive attenuation secondary to extensive soft tissue density. In these patients, technetium-based agents provide brighter images than thallium-201 and are therefore preferred. Finally, morbidly obese patients (heavier than 300 lb) often exceed the weight-bearing limits of most SPECT imaging tables, which makes noninvasive evaluation impossible.
What is the role of ETT in the evaluation of asymptomatic patients?
7In light of the enormous morbidity and mortality associated with CAD, attention has turned to early diagnosis and disease prevention. Unfortunately, the accuracy of stress testing in patients who are asymptomatic has not been well defined. Moreover, no evidence has shown definitively that screening asymptomatic patients will ultimately reduce the occurrence of cardiac events. Furthermore, screening large numbers of asymptomatic patients would result in a significant number of false-positive results that, in turn, would lead to unnecessary, expensive, and potentially hazardous interventions (such as cardiac catheterization).
In the recent ACC/AHA/ACP-ASIM guidelines, routine ETT screening of asymptomatic men and women is categorized as level of evidence class III (not useful/effective).3 However, it may be reasonable to screen asymptomatic men older than 40 years and women older than 50 years if they have multiple cardiac risk factors (class IIb [usefulness/efficacy is less well established by evidence/opinion]). It is also appropriate to screen sedentary middle-aged patients who plan to start a vigorous exercise program or those in positions that affect public safety (such as airline pilots, police officers, and firefighters).7
Before ordering a screening ETT in an asymptomatic patient, discuss with him the risks and benefits of testing. Explain the potential adverse psychological implications-as well as the impact on employment and insurance status-should test results be positive. The decision to proceed with further testing should be based not only on the ST-segment response, but also on additional treadmill data (exercise capacity, symptoms, blood pressure response), risk factor status, and patient preference.
What cost considerations affect the choice of a cardiac stress test?
8In this era of managed care, cost-effectiveness considerations have come into sharper focus in medical decision making.2 A standard ETT is the least expensive option. Stress echocardiography is approximately twice as expensive as standard ETT, and SPECT imaging is approximately 5 times more expensive.2 However, the lower cost of standard ETT does not necessarily translate into a net cost savings. If additional testing and intervention will ultimately be required, the relative cost-effectiveness of standard ETT is dramatically offset. This highlights the need for the referring physician to be well versed in the selection of the most appropriate cardiac stress tests based on knowledge of the strengths and limitations of the individual modalities.
1. 2001 Heart and Stroke Statistical Update. Dallas: American Heart Association; 2000.
2. Gibbons RJ, Chatterjee K, Daley J, et al. ACC/ AHA/ACP-ASIM guidelines for the management of patients with chronic stable angina: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Management of Patients With Chronic Stable Angina). J Am Coll Cardiol. 1999;33:2092-2197.
3. Eagle KA, Berger PB, Calkins H, et al. ACC/AHA guideline update for perioperative cardiovascular evaluation for noncardiac surgery-executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 2002;39:542-553.
4. Myers J, Prakash M, Froelicher V, et al. Exercise capacity and mortality among men referred for exercise testing. N Engl J Med. 2002;346:793-801.
5. Sketch MH, Mooss AN, Butler ML, et al. Digoxin-induced positive exercise tests: their clinical and prognostic significance. Am J Cardiol. 1981;48:655-659.
6. Psaty BM, Koepsell TD, Wagner EH, et al. The relative risk of incident coronary heart disease associated with recently stopping the use of beta-blockers. JAMA. 1990;263:1653-1657.
7. Gibbons RJ, Balady GJ, Beasley JW, et al. ACC/ AHA Guidelines for Exercise Testing. A report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. J Am Coll Cardiol. 1997;30:260-311.
8. Lee TH, Boucher CA. Noninvasive tests in patients with stable coronary artery disease. N Engl J Med. 2001;344:1840-1845.
9. Beller GA, Zaret BL. Contributions of nuclear cardiology to diagnosis and prognosis of patients with coronary artery disease. Circulation. 2000;101:1465-1478.
10. Depuey EG, Guertler-Krawczynska E, Robbins WL. Thallium-201 SPECT in coronary artery disease patients with left bundle branch block. J Nucl Med. 1988;29:1479-1485.
11. Larcos G, Brown ML, Gibbons RJ. Role of dipyridamole thallium-201 imaging in left bundle branch block. Am J Cardiol. 1991;68:1097-1098.
12. Heupler S, Mehta R, Lobo A, et al. Prognostic implications of exercise echocardiography in women with known or suspected coronary artery disease. J Am Coll Cardiol. 1997;30:414-420.
13. Marwick TH, Anderson T, Williams MJ, et al. Exercise echocardiography is an accurate and cost-efficient technique for detection of coronary artery disease in women. J Am Coll Cardiol. 1995;26:335-341.
14. Hlatky MA, Pryor DB, Harrell FE, et al. Factors affecting sensitivity and specificity of exercise electrocardiography. Multivariate analysis. Am J Med. 1984;77: 64-71.
15. Barolsky SM, Gilbert CA, Faruqui A, et al. Differences in electrocardiographic response to exercise of women and men: a non-Bayesian factor. Circulation. 1979;60:1021-1027.
16. Mark DB, Hlatky MA, Harrell FE Jr, et al. Exercise treadmill score for predicting prognosis in coronary artery disease. Ann Intern Med. 1987;106:793-800.
17. Mark DB, Shaw L, Harrell FE Jr, et al. Prognostic value of a treadmill exercise score in outpatients with suspected coronary artery disease. N Engl J Med. 1991;325:849-853.
18. Fearon WF, Gauri AJ, Myers J, et al. A comparison of treadmill scores to diagnose coronary artery disease. Clin Cardiol. 2002;25:117-122.
19. Alexander KP, Shaw LJ, Delong ER, et al. Value of exercise treadmill testing in women. J Am Coll Cardiol. 1998;32:1657-1664.
20. Vasilomanolakis EC. Geriatric cardiology: when exercise stress testing is justified. Geriatrics. 1985;40: 47-50.
21. Martinez-Caro D, Alegria E, Lorente D, et al. Diagnostic value of stress testing in the elderly. Eur Heart J. 1984;5(suppl E):63-67.
22. Cohen JL, Cheirif J, Segar DS, et al. Improved left ventricular endocardial border delineation and opacification with OPTISON (FS069), a new echocardiographic contrast agent: results of a phase III multicenter trial. J Am Coll Cardiol. 1998;32:746-752.