Volume 5, Issue 10 , Pages 760-766, November 2009
Cardiac Stress Testing in Women
Article Outline
- Abstract
- Non-Invasive Testing and Pretest Probability
- Test Selection: Stress and Data
- New Modalities
- Discussion
- References
- Copyright
Abstract
More women die every year from cardiovascular disease (CVD) than from any other cause. Non-invasive cardiac stress testing is key to timely diagnosis and treatment of CVD. However, providers need to be aware of how testing accuracy differs between men and women. This article reviews the standard non-invasive tests available, the sensitivity and specificity of each test, and how to choose the most appropriate test for female patients. It also discusses several new testing modes on the horizon.
Keywords: coronary artery disease , diagnosis , electrocardiography , exercise test , nuclear , women
Cardiovascular disease (CVD) is the leading cause of death among women worldwide, with 38.2 million U.S. women carrying this diagnosis.1 Contrary to popular belief, more women than men die each year of CVD; female CVD mortality exceeds that of breast cancer by a factor of 11.2 Coronary artery disease (CAD) is the number one cause of CVD death in women. More women than men will die within one year after their initial myocardial infarction (25% vs. 8%).3 As life expectancy continues to increase, the health care burden of CVD in women will continue to grow. The United States spent an estimated $403 billion on health care/lost productivity to CVD, compared with $190 billion for cancer, in 2006.1 Therefore, the need to accurately diagnose cardiovascular symptoms in women, using a cost-efficient paradigm, will become an increasingly valuable skill for health care providers.
Non-invasive diagnosis of CAD (ie, stress testing) is a key tool in diagnosis, yet the value and interpretation of these tests vary between male and female patients. Knowledge of these differences allow for best test selection and most accurate interpretation of the results.
Non-Invasive Testing and Pretest Probability
Non-invasive cardiac studies are performed to determine the probability of flow-limiting epicardial coronary artery disease (and at times to demonstrate the consequences of coronary disease, such as regional wall motion abnormalities or systolic dysfunction) in patients with suggestive symptoms. Predictive values define the usefulness of a diagnostic test. Positive predictive value (PPV = number of true positives/[number of true negative + number of false positives]) quantifies the probability of a positive test result in a patient that truly has the specified disease, while negative predictive value (NPV = number of true negatives/[number of true negatives + number of false negatives]) describes the probability of obtaining a negative test result in a patient without that disease.4 A key factor in maximizing the value of a study is matching the patient and test to give the best balance of PPV and NPV, and this is in part driven by the pretest probability of CAD in the patients studied.5
As in men, stress testing is best used in women classified as an intermediate risk for CAD (20–80% pretest probability), based on clinical criteria and Framingham global risk score, rather than in patients with low risk (due to a high incidence of false-positive results) or high risk (because of frequent false-negative results).6 Clinical history, including traditional atherosclerotic risk factors and characterization of symptoms (chest pain), is the most important tool in establishing pretest probability. The contribution of atherosclerotic risk factors (hypertension, hyperlipidemia, family history, physical activity, smoking history) to pretest probability are similar for men and women, although diabetes elevates pretest probability more in woman than in men.5 Chest pain is traditionally classified as (typical) angina, atypical angina, and non-anginal chest pain; however, these definitions, developed from studies of male populations, may not be as consistent or useful in females.7 The predictive value of typical angina in women equals that in men only after age 60, being less suggestive of CAD at younger ages.2 Women are more likely to use descriptors such as numbness, muscle tension, or burning sensation in the chest to describe ischemic discomfort, compared to men who use terms such as heaviness or pressure.8 Women also report more atypical or associated symptoms of myocardial ischemia such as abdominal pain, nausea, and vomiting.5
The predictive accuracy of a test is also related to the overall prevalence of the disease in the population being studied. Coronary disease typically presents 10 to 15 years of age later in women than in men, and is uncommon in premenopausal women without diabetes, vascular disease, or chronic kidney disease. Commonly available stress testing modalities diagnose obstructive epicardial CAD, and are more likely positive with increasing severity and number of obstructed vessels—conditions more prevalent in men. Women more commonly have non-obstructive or single vessel CAD, with a corresponding decrease in overall accuracy and predictive value of stress testing.7 Ischemic chest pain in women may also be secondary to endothelial dysfunction, a dynamic pathophysiological condition of the microvasculature, which cannot be diagnosed by standard non-invasive testing modalities. Hence, use of traditional pretest models to diagnose CAD in women, applied to conventional non-invasive testing, yields reduced accuracy and higher rates of false-positive and false-negative results.
Test Selection: Stress and Data
Each stress modality (exercise, vasodilators, or dobutamine) is combined with a form of data collection (scaler or ECG, nuclear perfusion imaging, or echocardiography) to formulate a diagnosis and/or prognosis. Each has different requirements to maximize accuracy and reliability in different clinical settings. Sensitivity (number of true positives/[number of true positives + number of false negatives]) and specificity (number of true negatives/[number of true negatives + number of false positives]) of each test vary accordingly.4
Selection of Stress Modality
Exercise stress tests, regardless of data type, require achievement of minimum heart rate and adequate exercise time to optimize NPV (true negative rate). Traditional exercise studies involve walking on a medical treadmill. The standard Bruce protocol is most commonly used and progressively increases the treadmill speed and incline at 3-minute intervals. Others, such as the modified Bruce and Naughton protocols, use lower and more gradual increases in speed, accommodating patients with certain physical limitations in ability to walk.9 Usually, patients must be able to walk briskly up hill in a comfortable and coordinated manner to successfully complete a maximal exercise test.10 If a patient is unable to reach target heart rate (THR, defined as > 85% of [220 − age]), the test is considered submaximal and non-diagnostic (unless the ECG is clearly positive at the heart rate achieved) and an alternative stress modality is usually required for definitive testing. Overall, patients with orthopedic conditions; sinus node dysfunction or heart block; taking bradycardia-inducing medications that cannot be discontinued for testing; and with peripheral vascular or neuropathic disease, which limits walking ability, cannot usually exercise to the required goal. Generally, women have lower exercise capacity (an average of 2 minutes less) and are often unable to obtain THR with treadmill exercise.11 This limitation is due in part to women's older age, more frequent deconditioning, and higher rates of functional impairment from co-existing medical conditions (ie, arthritis, chronic obstructive pulmonary disease) at the time diagnostic studies are needed.5 Patients who are deconditioned achieve THR in a shorter period of time: although THR identifies an adequate workload, total exercise duration less than 6 minutes limits exposure time to ischemic supply-demand mismatch and decreases the likelihood that diagnostic ischemic ECG changes will develop, even in the presence of obstructive disease.12
For patients unable to complete a physical exercise protocol, pharmacological stress options include vasodilators (adenosine and dipyridamole, the latter working through adenosine receptors) and inoptropes (dobutamine).6 Vasodilators produce differences in isotope uptake by dilating non-obstructed coronaries more than atherosclerotic ones, a phenomenon termed coronary steal. The steal occurs because the arterioles of significantly obstructed arteries are already maximally dilated (via autoregulation to maximize flow) and therefore incapable of further vasodilation in response to medications. Vasodilator studies, which are THR independent, have overall lower diagnostic sensitivity in single-vessel disease which is more common in women.7 Adenosine has an ultra-short half-life and is completely cleared from the circulation within 10 seconds of infusion discontinuation. Dipyridimole can be given as a rapid IV infusion (slow bolus) but has a longer half-life and sometimes requires reversal with aminophylline. Studies using these agents are generally considered contraindicated in the presence of hypotension, bradycardia, second-degree (or higher) heart block in the absence of an artificial pacemaker, and in patients with severe reactive airway disease. Potent vasodilators are also contra-indicated in severe aortic or mitral stenosis.
Similar to exercise, dobutamine elevates heart rate and contractility to increase myocardial oxygen demand; perfusion defects represent true ischemia (rather than steal) in areas receiving submaximal blood flow.5 Patients who fail to reach THR with dobutamine infusion alone are administered atropine concurrently. Dobutamine stress eliminates the need for physical exercise while still generating true myocardial ischemia in patients with obstructive CAD. However, predictive value is not as good as with physical exercise, as other systemic stress components (eg, skeletal muscle oxygen consumption, pulmonary reserve) are not tested with dobutamine stress. As a catecholamine agonist administered at high doses, dobutamine stress is generally considered contra-indicated in patients with poorly controlled arrhythmias (including atrial fibrillation), uncontrolled hypertension, rest ischemia or recent myocardial infarction, outflow tract obstruction (hypertrophic obstructive cardiomyopathy or valvular aortic stenosis), and with an unstable thoracic aorta (aneurysm or chronic dissection). Patients with sinus node dysfunction, pacemaker dependency, or who cannot tolerate temporary discontinuation of bradycardic medications such as beta blockers, are unlikely to reach THR and therefore less likely to obtain a high quality study.
Table 1 details stress modality selection criteria.
Table 1. Selection of Stress Modality
| Modality | Requirements | Contra-indications | Comments |
|---|---|---|---|
| Treadmill exercise | NPO at least 4 hrs prior to test | Poor coordination or balance | Provides assessment of functional capacity |
| Comfortable clothing | Joint or back conditions or peripheral vascular disease that limits walking | LBBB and LVH can result in false positives | |
| Safe walking shoes | Deconditioning | Provides information regarding exercise HR, and BP recovery—useful in prognosis | |
| Vasodilators (adenosine, dipyridamole) | NPO after midnight | Severe bronchospastic lung disease | Chest pain during study not indicative of CAD |
| Off theophylline and dipyridamole for 48 hrs | Second- or third-degree AV block | ||
| Off caffeine for 24 hrs | Severe bradycardia (HR < 40) | ||
| Hypotension (SBP < 90) | |||
| Theophylline dependence | |||
| Caffeine dependence | |||
| Severe/critical aortic or mitral stenosis | |||
| Dobutamine (with atropine if needed) | NPO after midnight | Recent MI (< 1 wk) | LBBB or post open-heart surgery may yield false positive |
| Off beta blockers for 48 hrs | Unstable angina | ||
| Severe aortic stenosis | |||
| Hypertrophic obstructive cardiomyopathy | |||
| Supraventricular arrhythmias (esp. with uncontrolled rate) | |||
| Ventricular tachycardia, current or prior history | |||
| Uncontrolled hypertension | |||
| Thoracic aortic dissection/large aneurysm | |||
| Severe pulmonary hypertension | |||
| Beta-blocker dependence | |||
| BBB or post open-heart surgery may yield false positive |
Selection of Data Method
Electrocardiography (ECG) is the simplest form of non-invasive data. A 12-lead ECG is recorded at rest, with hyperventilation, and at regular intervals during stress and recovery. Because interpretation of the results is based on ST and T wave changes, patients should have a normal resting ECG to maximize the accuracy of the study. Left bundle branch block (LBBB), left ventricular hypertrophy with repolarization changes, and other conditions associated with significant resting ST segment deviations decrease test specificity (and generally require imaging in both men and women to achieve an adequate study). Certain medications can limit the accuracy of the ECG in diagnosis of CAD. For example, digoxin and hormone replacement therapy can produce ST segment depression in patients without CAD. Women in general have a greater likelihood of false-positive ECG stress tests. Several reasons have been proposed for these differences, including the digoxin-like effects of estrogen on the ST segment response,13 the higher incidence of single-vessel disease in women, the higher incidence of mitral valve prolapse in women,5 and a possible “inappropriate” catecholamine response to exercise.9 Monitoring of ECG is also performed during perfusion imaging studies (see below).
Nuclear imaging involves the administration of a radioactive isotope (Thallium-201 or Technecium-99) intravenously and subsequent imaging using a gamma camera to determine coronary perfusion. Myocardial cells take up the isotope in proportion to blood flow, which then emits photons detectable by a gamma camera. By comparing images at rest and stress, myocardial perfusion and regional blood flow can be assessed.5 Ischemia is diagnosed when the myocardium demonstrates normal isotope uptake at rest, but localized (relative) decreased uptake with stress. This occurs as areas supplied by partially obstructed arteries cannot increase flow as much as normal coronaries, therefore creating conditional variance in isotope uptake in viable myocardium supplied by those partially obstructed arteries. Myocardial infarction or fixed scar is identified as decreased (or absent) uptake on both rest and stress images in the same location of the ventricle.14 Predictive accuracy may be reduced in women because of breast attenuation artifact (photon reduction due to soft tissue absorption independent of myocardial uptake) and the average smaller heart size in women. Gated single photon studies (gated SPECT), which correlates regional contractility to uptake, reduces false positives by separating soft tissue attenuation (decreased counts but preserved contractility) from true ischemia (reduced contractility in areas of decreased uptake). Nuclear imaging is the most expensive of all traditional non-invasive data modalities, but does significantly improve test accuracy (compared to ECG alone) in women.5
Echocardiography before and during stress assesses for changes in global and regional ventricular wall motion. Previous myocardial infarction (scar) is identified by hypokinetic, akinetic, or dyskinetic wall motion both at rest and during stress. Reversible ischemia is diagnosed when wall motion abnormalities develop during stress in areas showing normal function at rest.5 False-positive studies are far less common with echocardiographic imaging than with ECG monitoring alone.12 When using a Markov model (amount of resources expended to achieve a given improvement in patient outcomes), stress echocardiography was the most cost-effective coronary diagnostic test in women with a moderate probability of CAD.5
Table 2 summarizes comparative data selection criteria.
Table 2. Selection of Data Method
| Method | Requirements | Contra-indications | Comments |
|---|---|---|---|
| Scalar (ECG) | Normal resting ECG | Resting ECG abnormalities (LBBB, LVH) can cause false positives | Positive test includes typical angina, ≥ 1 mm horizontal/downsloping ST segment depression |
| Nuclear perfusion | None | Equipment limitations (weight > 350 lbs) | Comparison of rest and stress images yield size, location, viability, and number of ischemic segments |
| Claustrophobia | Provides good prognostic information, quantifying degree of abnormality | ||
| Echocardiography | Sonographer proficiency | Inadequate echo windows or inability to identify endocardial border for any reason | Positive when new regional wall motion abnormalities develop with stress |
| Use of echo contrast |
New Modalities
Coronary computed tomography (CT) angiography may produce the most useful noninvasive images of the coronary circulation to date to detect CAD. It allows for more precise evaluation of coronary stenosis, arterial wall thickness, and plaque morphology. In patients with an intermediate pre-test probability, coronary CT angiography has an overall sensitivity of 99%, specificity of 75%, and NPV of 99%.15 Two recent studies have compared CT angiography in men versus women, with one study showing no significant differences in accuracy and the other reporting significantly lower diagnostic accuracy in women due to smaller coronary artery size. Limitations of coronary CT angiography include: more radiation exposure to the breast area as compared to invasive angiography, decreased visualization of luminal narrowing in areas of calcified atheromatous plaques, and artifact from metallic implants such as coronary stents, surgical wires, clips, pacemakers, and prosthetic valves. CT angiography may be an excellent option to rule out CAD safely and reliably in patients whose symptoms are atypical or when other test results are equivocal.16
Cardiac magnetic resonance imaging (MRI) may provide information on cardiac function, viability, metabolism, and coronary morphology. It is useful in determining composition of plaques and identifying vulnerable plaques in coronary arteries—providing critical information on plaque stability and helping target coronary interventions for maximal benefit. Myocardial perfusion can be calculated by using MRI imaging before and after pharmacological stress, similar to nuclear imaging. Sensitivity and specificity for CAD detection based on a meta-analysis of 25 studies was 73% and 86%, respectively. No data are available specific to women; however, determination of the severity of CAD is more difficult in patients with small coronary arteries.16
Currently, no method exists to directly visualize coronary microvasculature and diagnose endothelial dysfunction. Positron emission tomography (PET) may provide this information in the future.16
Discussion
The majority of studies examining non-invasive testing for CAD did not include, or enrolled a small percentage of women, leaving gaps in our understanding of the best application of these studies in female patients. In general, both specificity and sensitivity are lower in females than in males.7 Conventional stress tests detect flow-limiting lesions within the epicardial coronary arteries, which are more common in men. Chest pain or other ischemia-related symptoms in women due to endothelial dysfunction is poorly detected by current stress testing modalities.17 Table 3 summarizes comparisons between men and women for common stress testing combinations.
Table 3. Comparisons of Women to Men for Common Stress Tests
| ECG | Nuclear | Echocardiography | |
|---|---|---|---|
| Exercise | Insufficient evidence to remove as initial test in women with normal ECG, intermediate pretest probability of CAD, and able to exercise | When thallium-201 used, sensitivity 78%, specificity 64% in women. Specificity is reduced with thalium-201 because of increased breast attenuation | Interpretation is sex neutral |
| ETT in women who had intermediate probability of heart disease and were able to exercise for greater than 6 min was an equally valid test as for men | Technechium-99 results in increased specificity (cutting false-positive rate by as much as 50%) due to brighter signal intensity | Sensitivity 86%, specificity 79% (equally accurate in men and women) | |
| Women are less likely to attain target heart rate (exercise, on average, 2 minutes less than men; older at time of CAD presentation) | SPECT imaging has specificity of 58% compared with planar images (89%) in women | Observational data found more cost effective than exercise ECG | |
| Sensitivity 61%, specificity 70% in women (72% and 77% in men) | Overall, sensitivity about 10% lower in women than in men | ||
| Vasodilator | Obtain information from continuous ECG during infusion; however, traditionally combined with other (nuclear or echo) | More accurate than exercise nuclear imaging in both women and men with LBBB | Lower diagnostic sensitivity for single-vessel CAD, which is more common in women |
| Technecium-99 results in higher specificity as above | Few reports available for gender specific accuracy, but expect lower sensitivity and higher specificity than exercise or dobutamine echo | ||
| Dobutamine | Obtain information from continuous ECG during infusion; however, traditionally combined with other (nuclear or echo) | In men, sensitivity 87%, specificity 70%; likely to have slightly lower sensitivity and higher specificity as compared to vasodilator nuclear stress | Reliably detects multi-vessel CAD in women; however, single-vessel CAD more common in women |
| Technecium-99 results in higher specificity as above | Sensitivity 75%, specificity 86% (similar to men) |
Despite these limitations, standard stress testing remains of significant value in diagnosing atherosclerotic CAD in symptomatic women. New modalities offer promise of improved accuracy and possibly the ability to diagnose microvascular angina due to endothelial dysfunction.
References
- Evidence-based guidelines for cardiovascular disease prevention in women: 2007 update . Circulation . 2007;115:1481–1501
- ACC/AHA 2002 guideline update for exercise testing: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Committee on Exercise Testing) . Circulation . 2002;106:1883–1892
- . Guidelines update on the prevention of heart disease in women . J Am Acad Nurse Pract . 2008;20:191–193
- . The Cambridge dictionary of statistics . 1st ed.. United Kingdon: Cambridge University Press; 1998;
- . Cardiovascular health and disease in women . 2nd ed.. Philadelphia, PA: W.B. Saunders Company; 2002;
- ACC/AHA/ASNC guidelines for the clinical use of cardiac radionuclide imaging: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (ACC/AHA/ASNC committee to revise the 1995 guidelines for the clinical use of cardiac radionuclide imaging) . J Am Coll Cardiol . 2003;42:1318–1333
- Role of noninvasive testing in the clinical evaluation of women with suspected coronary artery disease. Consensus statement from the Cardiac Imaging Committee, Council on Clinical Cardiology, and the Cardiovascular Imaging and Intervention Committee, Council on Cardiovascular Radiology and Intervention, American Heart Association . Circulation . 2005;111:682–696
- . Detection of exercise-induced myocardial ischemia from symptomatology experienced during testing in men and women . Can J Cardiol . 2006;22:411–417
- . Effective exercise stress testing for CAD . Emerg Med . 2001;33(5):5–12
- . In: Manual of exercise testing . St Louis, MO: Mosby Year Book Press; 1989;p. 13–48
- Value of exercise treadmill testing in women . J Am Coll Cardiol . 1998;32:1657–1664
- . In search of the false-negative exercise treadmill testing evidence based use of exercise echocardiography . Clin Cardiol . 2008;31:35–40
- Beneficial effect of estrogen on exercise-induced myocardial ischemia in women with coronary artery disease . Lancet . 1993;342:133–136
- . Noninvasive cardiac stress testing: eight questions physicians often ask . Consultant . 2002;43(1):23–32
- Non invasive coronary computed tomography for patients with suspected coronary artery disease: the Coronary Angiography by Computed Tomography with the Use of a Submillimeter resolution (CACTUS) trial . 2007;28(24):3034–3041
- Current diagnostic concepts to detect coronary artery disease in women . Eur Heart J . 2008;29(6):707–717
- Women's ischemic syndrome evaluation. Current status and future research directions: report of the National Heart, Lung and Blood Institute Workshop . Circulation . 2004;109:47–49
In compliance with national ethical guidelines, the authors report no relationships with business or industry that would pose a conflict of interest.
PII: S1555-4155(08)00605-3
doi:10.1016/j.nurpra.2008.10.017
© 2009 American College of Nurse Practitioners. Published by Elsevier Inc. All rights reserved.
Volume 5, Issue 10 , Pages 760-766, November 2009
