Exercise stress testing
Editor(s)-In-Chief: C. Michael Gibson, M.S., M.D. ; Ernest Gervino, Ph.D.; Brenna Southern, M.D.; Kapil Kumar, M.D.; Bruce D. Nearing, Ph.D.; Richard L. Verrier, Ph.D.
To read more about Exercise Tolerance Test Time as an Endpoint for Clinical Trials Evaluating Therapies for Refractory Angina, click here.
An exercise stress test (EST) is an evaluation modality used in cardiology in which the ability of the heart to respond to stress, either actually induced by exercise or stimulated by pharmacologic maneuvers, is measured in a controlled clinical setting. The image created by its recording is known as an electrocardiogram or ECG.
The test is typically included in the initial evaluation of suspected ischemic heart disease, and as a prognostic indicator after myocardial infarction.
- Low cost
- Short duration
- Functional status evaluation
- High sensitivity in 3 VD or left main disease
- Useful prognostic information
- Sub-optimal sensitivity in the detection of single vessel disease (50%), 85% in the presence of three vessel disease
- In all patients, overall sensitivity 68%, specificity 77%
- Beta blocker use is associated with a higher rate of false negatives (fail to achieve rate pressure product)
- Non diagnostic in patients with abnormal baseline EKG (dig, LVH, WPW)
- Poor specificity in certain patient populations: premenopausal women, LVH, dig, IVCD, hypokalemia, hyperventilation, severe hypertension, resting ST abnormalities
- The negative predictive value in women of low to intermediate risk is high, the positive predictive value in men is high
- Need to achieve > 85% of maximum heart rate for maximizing accuracy
- Its main values lies in excluding CAD in patients with a low pre-test probability of CAD based on gender and age.
Stress Radionuclide Myocardial Perfusion Imaging
- Simultaneous evaluation of perfusion and function with gated SPECT
- Higher sensitivity and specificity than exercise EKG: For exercise or pharmacologic SPECT imaging with Tl or Tc, in patients with chest pain the sensitivity for the detection of CAD is 85% to 90%. Specificity for excluding CAD is in the 90% range. Good in patients with LVH, dig, IVCD etc. ST segment depression response higher rate pressure product than does a perfusion abnormality with tracers. Therefore they are more sensitive. Adding stress perfusion imaging to the exercise ECG stress test greatly assists in differentiating true positive from false positive ETT ST segment depression. For single vessel disease, the sensitivity is 25% higher with SPECT imaging compared with exercise testing. The sensitivity for detecting 3VD with exercise SPECT is 95% to 100%.
- High specificity with Tc labeled agents: Half life is shorter than Tl, therefore dose is higher, therefore image is brighter and better. Also allows gated assessment of LV thickening.
- Studies can be performed in almost all patients
- Significant additional prognostic information, can quantitate LV function
- Comparable accuracy with pharmacologic stress testing
- Viability and ischemia when assessed simultaneously
- Quantitative image analysis
- Suboptimal specificity with thallium imaging, with a high false positive rate in many labs, particularly among women and obese patients.
- Long procedure time with Tc agents, higher costs than ETT
- Radiation exposure
- Poor images in obese patients
- Pharmacologic stress testing: sensitivity and specificity are similar for persantine and adenosine. Dobutamine is used in those patients with a history of bronchospasm, or for those patients who have consumed coffee before the procedure. Pharmacologic testing is the preferred method in patients with LBBB.
- Women with chest pain who are referred for exercise or pharmacologic stress testing benefit the most from the enhanced accuracy of Tc imaging. Both Tl and Tc had a sensitivity of about 70%, but the specificity rose to 92% with Tc. Most labs now use Tc because of its improved specificity, the ability to gait the images and assess regional wall thickening. Mild non reversible defects that show preserved systolic thickening usually represent attenuation artifacts, however, if there is abnormal wall thickening, then this is most likely scar.
Exercise/Pharmacologic Stress Echocardiography
- Higher sensitivity and specificity than exercise EKG: Metanalysis showed sensitivity of 84%, specificity 86%. Marked variation across trials though, highly operator dependent. If the max heart rate is < 85% of age predicted, then sensitivity drops to 42%. Sensitivity is 10% lower in women than in men, specificity is the same across genders. In women with single vessel disease the sensitivity was only 40%, if there was 2 or 3 vessel disease, this number increased to 60%.
- Additional prognostic value over exercise EKG
- Dobutamine stress has higher sensitivity than does pharmacologic stress
- Time to complete examination is short
- Identification of co-existent structural cardiac abnormalities (valvular disease)
- Simultaneous evaluation of perfusion with contrast agents
- Relatively lower costs than with other techniques
- No radiation
- Decreased sensitivity for the detection of single vessel disease or mild stenosis with post exercise imaging
- Inability to image the entire ventricle in some patients
- Highly operator dependent in the analysis of images
- No quantitative image analysis
- Poor windows in patients with COPD
- Infarct zone ischemia less well detected
Comparison of exercise SPECT imaging and Exercise Echocardiography
- Both have a higher sensitivity and specificity than regular exercise EKG testing
- Both provide functional information that EKG testing does not
- Both provide information about myocardial viability, which the angiogram does not
Strengths of Stress ECHO over SPECT
Noninvasive, safe and repeatable, no radiation exposure, quick, little sophisticated equipment and portable, low costs, can identify co-existing valvular heart disease
Limitations of Stress ECHO over SPECT
Images are difficult to obtain at peak exercise, an ischemic response is required to observe wall motion abnormalities, wall motion can recover quickly in the presence of mild ischemia, detection of residual ischemia is difficult in an akinetic wall zone, the technique is highly operator dependent, good quality images were only acquired in 70% of cases.
Strengths of SPECT over stress ECHO
Does not require an ischemic response to be abnormal, just requires an abnormality in flow reserve, sensitivity is slightly higher by about 8-10 percentage points (mostly because the ability to detect single vessel disease or mild stenoses of 50-70% is not as good with stress Echo), can see defects in areas that contain scar and viable myocardium, acquisition of images is not operator dependent, in virtually 100% of patients diagnostic images are obtained, with Tc simultaneous assessment of perfusion and function is obtained, resting LV ejection fraction can be obtained, vasodilator SPECT has significantly higher sensitivity than vasodilator stress ECHO, dobutamine ECHO is associated with higher sensitivity and specificity than vasodilator ECHO.
Limitations of SPECT imaging in relation to stress ECHO
Longer imaging protocols, greater expense of equipment, must inject and store radiopharmaceuticals, inability to visualize the heart in real time, lower spatial resolution than ECHO, higher costs to patients.
In general, the sensitivity is lower for stress ECHO while the specificity is higher.
Exercise Tolerance Testing
- 1 mm or more of horizontal or downsloping ST depression is associated with a poor prognosis
- Failure to achieve 6 METS is associated with an elevated mortality rate over the next 2.5 years.
- Failure of heart rate to rise is associated with higher mortality, even after adjusting for perfusion defects.
- Failure to reach 85% of age adjusted max HR is associated with a RR of 1.85 in mortality.
- Limitation of ETT is the fact that the magnitude of ST depression is not strongly associated with the extent of CAD
- Exercise testing alone has excellent prognostic ability among patients with atypical chest pain or non anginal pain who have a normal EKG at baseline. If these patients have a normal ETT, the prognosis is excellent.
Stress Myocardial Perfusion
- The following are associated with a poor prognosis:
- 20% of the LV is a perfusion defect
- Defects in more than one distribution suggestive of multivessel CAD
- A large number of non reversible defects
- Transient LV cavitary dilation
- Increased lung uptake
- Resting LVEF of < 40%
- Normal thallium: Mortality 1% per year
- Normal Tc: annual mortality 0.6%, 12 fold higher if there is a Tc defect
- The positive predictive value of stress myocardial perfusion imaging and stress ECHO is low: That is the percentage of people who die or sustain an MI is low among patients with abnormal findings. On the other hand the negative predictive value is high and exceeds 95%.
Non Invasive Risk Stratification According to the ACC Appropriate Use Criteria
High-Risk (greater than 3% annual mortality rate)
1. Severe resting left ventricular dysfunction (LVEF less than 35%)
2. High-risk treadmill score (score less than or equal to 11)
3. Severe exercise left ventricular dysfunction (exercise LVEF less than 35%)
4. Stress-induced large perfusion defect (particularly if anterior)
5. Stress-induced multiple perfusion defects of moderate size
6. Large, fixed perfusion defect with LV dilation or increased lung uptake (thallium-201)
7. Stress-induced moderate perfusion defect with LV dilation or increased lung uptake (thallium-201)
8. Echocardiographic wall motion abnormality (involving greater than two segments) developing at low dose of dobutamine (less than or equal to 10 mg/kg/min) or at a low heart rate (less than 120 beats/min)
9. Stress echocardiographic evidence of extensive ischemia
Intermediate-Risk (1% to 3% annual mortality rate)
1. Mild/moderate resting left ventricular dysfunction (LVEF equal to 35% to 49%)
2. Intermediate-risk treadmill score (11 less than score less than 5)
3. Stress-induced moderate perfusion defect without LV dilation or increased lung intake (thallium-201)
4. Limited stress echocardiographic ischemia with a wall motion abnormality only at higher doses of dobutamine involving less than or equal to two segments
Low-Risk (less than 1% annual mortality rate)
1. Low-risk treadmill score (score greater than or equal to 5)
2. Normal or small myocardial perfusion defect at rest or with stress*
3. Normal stress echocardiographic wall motion or no change of limited resting wall motion abnormalities during stress*
Techniques used to Assess Myocardial Viability
Rest and delayed redistribution is the most common radionuclide method used to assess viability. Uptake of Tl is related not only to blood flow, but also to membrane integrity. Myocardial stunning or hibernation does not result in a reduction in Tl extraction as long as the sarcolemmal membrane does not sustain irreversible ischemic damage. 60 to 70% of asynergistic segments will show > a 50% improvement after revascularization.
Same as above, as usual a better signal with Tc, can also assess regional wall thickening. If thickening is present, then viability is likely.
- Considered by many to be the gold standard. Can be used to assess perfusion and metabolism simultaneously. If there is mismatch in perfusion and metabolism, then the tissue is viable. If there is a match, then there is scar.
- Dobutamine: Enhanced systolic contractility with low dose dobutamine is associated with recovery.
Emergent Stress Testing in Young People
Stress testing has frequently been used to assess adult patients with suspected or known coronary artery disease (CAD) based on pre-test probability. Pre-test probability is the assessment of a patient and their likelihood of CAD based on clinical history and symptoms. Stress testing to diagnose myocardial ischemic syndrome is usually indicated only in patients with an intermediate pre-test probability.
The average age of a patient who undergoes a stress test is typically between 45-60 years. Increasing age is one of many positive risk factors for CAD. However, there have been several cases in which young adults and adolescents have presented with chest pain and were found to have had a myocardial infarction (MI).  Since chest pain can be a complaint among children, the question becomes whether or not an emergent stress test is needed.
The most common reason for stress testing is chest pain. All patients who present with acute or chronic chest pain need to be evaluated to determine the course or urgency of further non-invasive vs. invasive testing. Inpatient stress testing can be done if a recent MI or an acute unstable coronary syndrome has been excluded.
Among children presenting with chest pain, the symptoms often tend to be benign.  Given the fact that the majority of children have no probable cardiac risk factors, their pre-test probability is already very low. Yet there are several conditions that can cause ischemic chest pain and other cardiac abnormalities so a thorough careful history and physical examination should always be performed. The presenting symptom can be secondary to congenital defects as well as acquired diseases. Kawasaki disease has a common manifestation of coronary artery aneurysms which can progress to coronary stenosis.  Acute MI is one of the main causes of death in children with Kawasaki disease. Another acquired condition is sickle cell disease in which children can frequently present with chest pain, have an MI and have normal coronary arteries.  Other issues that could cause ischemic chest pain are coronary vasospasm, pericarditis or myocarditis, cocaine use, or other conditions causing anatomic congenital cardiovascular abnormalities.
Acute symptoms in children should be dealt with accordingly to rule out an MI, congenital defects or diseases. Based on above indications, an emergent stress test may not be warranted. To help determine the etiology of the symptom, ECG, echocardiogram, MRI, cardiac enzymes, drug screening, blood testing for hypercoagulability and coronary angiograms may be more useful. Or for chronic chest pain associated with exertion, an outpatient stress test could also be helpful.
Whether or not stress testing is emergent in children should again be considered similarly to adult emergent stress testing. Comprehensive assessment of acute or chronic problems and the consideration of the child’s pre-test probability being significantly low are compelling points that an emergent stress test may not be necessary.
T Wave Alternans for Risk Stratification during Exercise Stress Testing
Across the past decade, a sizeable body of evidence has been amassed indicating that measurement of T wave alternans (TWA), a beat-to-beat fluctuation in the morphology of the T wave, during exercise is useful in assessing risk for life-threatening arrhythmias.
TWA is a marker of repolarization instability and an indicator of a vulnerable myocardial substrate. This electrocardiographic phenomenon parallels the beat-to-beat oscillation of action potential duration (APD) at the level of cardiac myocytes. The cellular mechanism has been linked primarily to an aberration in intracellular calcium, which results in fluctuation of calcium transients from one beat to the next. This oscillation in APD can be solely a temporal event (concordant alternans) or both a temporal and spatial occurrence (discordant alternans). Discordant alternans has the potential to create steep repolarization gradients leading to transient unidirectional block, a pre-requisite for reentrant arrhythmias  
Until recently, TWA analysis has largely involved frequency-domain based spectral analysis. The spectral method (SM) requires provocative testing to raise and plateau the heart rate. The level of TWA detected is in the range of a few microvolts and thus cannot be observed by visual inspection. SM is the first and most widely studied commercially available algorithm (Cambridge Heart, Inc.). It employs a fast Fourier transformation of the electrocardiogram (ECG) across 128 consecutive beats into the frequency domain and employs specialized electrodes to minimize noise. The power of the spectrum at 0.5 cycle per beat (occurring on every other beat) between the JT segment is defined as the alternans power. An alternans level (Valt) >1.9 μV, greater than 3 times the standard deviation of noise (k score), and sustained for at least one minute at stable heart rates <110 beats per minute is considered a positive test, indicating that TWA is present. A negative test is defined as one that has a Valt of <1.9 μV at a heart rate >105 bpm without significant noise or premature beats. Tests that do not strictly meet the positive or negative test definitions are referred to as indeterminate  and occur in 20 to 40% of all cases. Most recent studies using SM have grouped positive and indeterminate tests together as “abnormal” or “non-negative,” since the risk of death or sustained ventricular arrhythmias in patients with indeterminate tests due to patient factors is as high as that of patients with positive tests. 
A recently developed, FDA cleared commercial method (GE Healthcare, Inc.) is time domain modified moving average (MMA) developed at Beth Israel Deaconess Medical Center (Nearing and Verrier 2002). This technique was developed to circumvent the stationarity requirements of SM, which mandates stabilization of heart rate for several minutes given the fast Fourier transform. The requirement for specialized electrodes is also eliminated through the use of advanced noise reduction algorithms. The MMA method separates odd and even beats into separate bins and creates median templates for both the odd and even complexes every 15 seconds.  These templates are then superimposed and the entire JT segment is analyzed for alternation. The peak difference between the odd and even median complexes at any point within the JT segment is defined as the TWA value. These templates of superimposed complexes may be examined visually to verify TWA presence and magnitude. Noise measurements are in part derived from mismatch of the median templates outside of the JT segment. The moving average allows control of the influence of new incoming complexes on the median templates with an adjustable update factor. A lower update factor provides greater sensitivity in detecting transient surges in TWA.
Results from SM and MMA are highly comparable, although the TWA values reported with the MMA algorithm are consistently larger by a factor of 4 to 10. This difference is mainly attributable to the fact that SM reports the average TWA level across the entire JT segment for 128 beats that is above the noise level, while MMA method reports the peak TWA level at any point within the JT segment for each 15-second beat stream, with the noise level reported separately.
The majority of clinical studies focusing on TWA as a risk stratification tool have enrolled CAD patients with EF 40% and employed the SM. In 2005, Gehi and colleagues  conducted a meta-analysis of 19 prospective studies of exercise-based TWA testing with SM that enrolled a total of 2608 patients. The majority of these patients had CAD, and half had depressed EF, but only a small percentage had a history of ventricular arrhythmias. Positive TWA test results conveyed an average 3.77-fold risk of future ventricular tachyarrhythmic events when compared to patients with negative TWA test results. The negative predictive value (NPV) of TWA was 97.2%. However, its positive predictive value (PPV) was quite poor, generally <30% for all subgroups.
By virtue of its excellent NPV, TWA testing has been presented as a means of identifying those patients who are least likely to experience a future ventricular tachyarrhythmic event and thus least likely to benefit from ICD implantation.
Only one large prospective observational trial has investigated TWA in a broader population. The incidence of SCD in this subgroup of patients, however, is relatively low; rendering it even more difficult to identify those most likely to benefit from ICD implantation even though the absolute number of SCD events is higher in this population than in those with depressed EF.  Nieminen and coworkers (2007) provided evidence that TWA is also suitable as a screening tool in the general population of patients with preserved ejection fraction and can be performed during routine exercise stress testing. They applied the MMA method in 1037 consecutive patients referred for exercise testing and reported that TWA 65μV recorded in the precordial leads predicted all-cause death (RR= 3.3), cardiovascular mortality (RR=6.0), and sudden cardiac death (RR=7.4) across the 44 ±7 month follow-up. The analysis window was restricted to heart rates 125 beats/min in order to minimize the effects of noise.
Most recently, the REFINE study  enrolled 322 post-MI patients with ejection fraction 50% and measured TWA at 10 to 14 weeks. Spectral analysis was performed during the specialized exercise protocol, and MMA was employed during post-exercise recovery. Exner and colleagues (2007) determined that the predictivity of the spectral and MMA methods for TWA analysis is similar, with hazard ratios in the range of 2.75-2.94 for cardiac death or arrest during 47 months following the index event. Combining the TWA test results with heart rate turbulence, a noninvasive marker of autonomic tone, accurately predicted risk of cardiac death or arrest with a hazard ratio of 5.2 and identified the majority of patients destined to suffer serious events.
Collectively, sound scientific and clinical evidence support the utility of TWA testing for sudden death risk stratification during exercise. With the advent of time-domain based TWA analysis, this measurement can be performed seamlessly during the course of routine clinical exercise stress testing as well as ambulatory ECG monitoring. While TWA testing has been focused largely on guiding ICD implantation for primary prevention, there may be a greater role for TWA analysis in screening the broader, low-risk population and for evaluating the effectiveness of medical therapy.
Since sudden cardiac death results from diverse pathologic mechanisms, involving derangements in myocardial substrate and altered autonomic function, it is unlikely that any single parameter will adequately represent the complex factors that lead to lethal ventricular arrhythmias. Therefore, it will be valuable to examine whether combinations of several risk stratification parameters may be more effective than any individual parameter as observed in the REFINE study.
- Myocardial infarction
- Emergent Stress Testing in Young People 
- T Wave Alternans for Risk Stratification during Exercise Stress Testing 
- Modified Gervino Protocol
- Emergent Stress Testing in Young People 
- T Wave Alternans for Risk Stratification during Exercise Stress Testing 
- ↑ Sabatine, Marc (February 15, 2000). Pocket Medicine. Lippincott Williams & Wilkins. pp. 256 pages.
- ↑ 2008.10.005 J. Am. Coll. Cardiol. 2009;53;530-553.
- ↑ Gibbons, RJ, Balady, GJ, Bricker, JT, et al. ACC/AHA 2002 guideline update for exercise testing: summary article: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2002; 106:1883.
- ↑ Kocis, KC. Chest pain in pediatrics. Pediatr Clin North Am 1999; 46:189.
- ↑ Lane, JR, Ben-Shachar, G. Myocardial Infarction in Healthy Adolescents. Pediatrics 2007; 120 No.4: 938
- ↑ Taubert, KA, Shulman, ST. Kawasaki Disease. Am Fam Physician 1999; 59 No.11: 3093
- ↑ Martin, CR, Johnson, CS, Cobb, C, et al. Myocardial infarction in sickle cell disease. J Natl Med Assoc 1996; 88:428.
- ↑ Narayan SM: T-wave alternans and the susceptibility to ventricular arrhythmias. J Am Coll Cardiol 2006, 47: 269-281.
- ↑ Nearing BD, Verrier RL: Tracking heightened cardiac electrical instability by computing interlead heterogeneity of T-wave morphology. J Appl Physiol 2003, 95:2265-2272.
- ↑ Bloomfield DM, Hohnloser SH, Cohen RJ: Interpretation and classification of microvolt T wave alternans tests. J Cardiovasc Electrophysiol 2002, 13:502-512.
- ↑ Kaufman ES, Bloomfield DM, Steinman RC, et al: “Indeterminate” microvolt T-wave alternans tests predict high risk of death or sustained ventricular arrhythmias in patients with left ventricular dysfunction. J Am Coll Cardiol 2006, 48:1399-1404.
- ↑ Nearing BD, Verrier RL: Modified moving average method for T-wave alternans analysis with high accuracy to predict ventricular fibrillation. J Appl Physiol 2002, 92:541-549.
- ↑ Gehi AK, Stein RH, Metz LD, Gomes JA: Microvolt T-wave alternans for the risk stratification of ventricular tachyarrhythmic events: a meta-analysis. J Am Coll Cardiol 2005, 46:75-82.
- ↑ Huikuri HV, Castellanos A, Myerburg RJ: Sudden death due to cardiac arrhythmias. N Engl J Med 2001, 345:1473-1482.
- ↑ Exner DV, Kavanagh KM, Slawnych MP, et al, for the REFINE Investigators: Noninvasive Risk Assessment Early After a Myocardial Infarction. The Risk Estimation Following Infarction, Noninvasive Evaluation (REFINE) Study. J Am Coll Cardiol 2007, 50:2275-84.
- ↑ Kumar K, Kwaku KF, Verrier RL. Treatment Options for Patients with Coronary Artery Disease Identified as High-Risk by T-Wave Alternans Testing. In: Current Treatment Options in Cardiovascular Medicine 2008, in press.