Coronary artery disease (CAD) represents the leading cause of morbidity and mortality in western countries. Many non-invasive stress tests are currently available for detecting CAD and assessing prognosis. Non-invasive stress techniques comprise exercise electrocardiographic (ECG) stress testing and imaging techniques, namely exercise/pharmacological stress echocardiography (SEcho) and exercise/pharmacological stress myocardial perfusion imaging (single-photon-emission computed tomography [SPECT]). Other available imaging techniques, less frequently used, include positron emission tomography and cardiovascular magnetic resonance imaging. All of them use either exercise or pharmacological stress to produce heterogeneity of blood flow between myocardial regions supplied by stenotic vessels and those that are perfused by normal coronary arteries in order to detect ischemia and to evaluate viability.
Pathophysiological Mechanisms of Ischemia
In the absence of a flow-limiting coronary stenosis, physiological stress results in an increase in heart rate and myocardial contractility, leading to augmentation of systolic wall thickening, endocardial excursion, global contractility, and myocardial oxygen demand. In the presence of a coronary stenosis, the increase in oxygen demand is not matched by an appropriate increase in blood supply. The imbalance between myocardial energy requirements and oxygen delivery results in ischemia.
The temporal sequence of events that occur after the onset of ischemia, usually called the ischemic cascade, is predictable and includes the presence of a perfusion defect, diastolic dysfunction (decreased relaxation and increased diastolic stiffness), electrocardiogram (ECG) changes, systolic dysfunction, and, finally, chest pain. This sequence is translated clinically into a gradient of sensitivity of the different clinical markers of ischemia. Since regional contractile abnormalities occur soon after development of the perfusion defect, echocardiographic evaluation during stress is extremely useful for diagnosing CAD.
The earliest change provoked by ischemia is delayed onset and termination of systolic thickening, which cannot be identified by visual inspection but may be detected by better temporal discriminators, such as myocardial deformation analysis. Immediately afterwards, ischemia reduces endocardial excursion and myocardial thickening.
Recently, Reant et al.1 demonstrated in an open-chest animal model of flow-limiting and non-flow-limiting coronary artery stenosis that longitudinal and circumferential systolic function abnormalities (implied in endocardial excursion) precede the decrease in radial deformation (implied in wall thickening). This observation may eventually be explained by the predominant longitudinal orientation of subendocardial myocardial fibres, which are much more sensitive to ischemia and therefore affected earlier in the ischemic cascade.
During regional myocardial ischemia, non-ischemic segments may demonstrate compensatory hypercontractility, preserving global ventricular systolic function. In the setting of CAD, the absence of improvement of global systolic function during stress associated with left ventricular enlargement is a marker of severe ischemia, suggesting multivessel involvement. A global decrease in left ventricular function in response to stress, however, may be due to other causes such as hypertension or cardiomyopathy. In the vast majority of patients, once the stressor is eliminated, ischemia resolves and wall motion abnormalities recover rapidly. Sometimes contractile dysfunction may persist for 30 minutes or longer, however, suggesting the presence of more severe ischemia.
The combination of stress testing and echocardiography—SEcho—has assumed an important role in the diagnosis of CAD. The appearance of a new stress-induced regional wall motion abnormality, readily identified by echocardiography, allows the detection of ischemia, and its location may be used to predict the stenosed coronary vessel. The types of stress employed fall into two basic categories: exercise and pharmacological. SEcho with pharmacological stress can be performed with either dobutamine or vasodilator stressors (dipyridamole or adenosine), dobutamine being the best studied and most widely clinically available. Other forms of stress, such as atrial pacing in patients with a permanent pacemaker, are less frequently used.
Exercise Stress Echocardiography
SEcho with exercise testing can be performed either on a treadmill or by stationary (upright or supine) cycle ergometry. Exercise induces an increase in heart rate, blood pressure, myocardial contractility, and cardiac work, increasing oxygen demand.
Treadmill exercise is the most common form of stress testing in the US. In this context, echo imaging is performed before and immediately after treadmill exercise without affecting the exercise portion of the test. Ischemia may resolve quickly after termination of the exercise, so all post-exercise images should be obtained within one to two minutes.
The primary advantage of bicycle exertion is the ability to acquire images during the exercise protocol, particularly at peak stress, avoiding the potential problem of rapid recovery after termination of exercise. The attainable workload with a bicycle is lower than with a treadmill, however, since some patients find cycling in the supine position very difficult.
Dobutamine Stress Echocardiography
The underlying principle of dobutamine SEcho (DSE) is that adrenoreceptor stimulation will augment heart rate, blood pressure, contractility, and myocardial oxygen demand. Hemodynamic response to dobutamine and exercise are not absolutely identical. In fact, changes in venous return and heart rate response are more pronounced with exercise, while dobutamine provokes a greater augmentation in myocardial contractility.
Current state-of-the-art DSE exam involves various stages of either low- or high-dose protocols, with increments from 5 to 40μg/kg/min, with each stage lasting three minutes. Echocardiographic second harmonic images are acquired at each stage to determine new wall motion abnormalities, worsening of pre-existing abnormalities, or enhanced wall motion. The protocol should not be stopped because of the induction of minor wall motion abnormalities, as the chance to identify multivessel CAD increases when maximal stress testing is pursued.
One of the limitations of DSE is the failure to achieve 85% of age-predicted maximum heart rate. Several reports have shown that the addition of atropine to peak-dose dobutamine (in doses of 0.25mg up to a maximum of 1mg) is safe and beneficial in patients without symptoms and echo signs of ischemia who developed inadequate heart rate responses.2 It is especially beneficial in those taking beta-blockers and those in whom second-degree heart block develops during tachycardia.2 Moreover, anti-anginal medical therapy (in particular beta-blockers) strongly reduces the diagnostic accuracy of all forms of stress; therefore, discontinuation of the drug at the time of testing is recommended to avoid a false-negative result.
Side effects of dobutamine include anxiety, flushing, and palpitations due to premature ventricular or atrial contractions and brief episodes of non-sustained ventricular tachycardia. Occasionally, dobutamine induces paradoxical hypotension and left ventricular outflow obstruction with systolic anterior motion of the mitral valve. The half-life of dobutamine is short, so side effects can be readily reversed through termination of the infusion. In severe cases, a short-acting intravenous beta-blocker such as esmolol or metoprolol is effective. Despite the use of high-dose protocols, DSE is an exceptionally safe procedure in appropriate patients, with serious complications such as life-threatening arrhythmias or myocardial infarction occuring in approximately three in 1,000 patients.3
Dipyridamole and Adenosine Stress Echocardiography
Dipyridamole inhibits adenosine uptake, inducing endogenous adenosine accumulation. The stimulation of adenosine receptors induces potent regional vasodilatation, which is significantly less pronounced in those areas supplied by stenotic coronary arteries. Flow is diverted away from abnormal regions (coronary steal), and this phenomenon of blood flow maldistribution produces ischemia in those regions with a precarious flow balance. Vasodilator stressors have a negative chronotropic and dromotropic effect and no relevant hypertensive response occurs. Despite the different pathophysiological mechanisms, the vasodilator at appropriately high doses shows ischemic stressor potency similar to exercise or dobutamine.4 Moreover, the lower peak heart rate in those protocols compared with exercise or dobutamine facilitates the acquisition of images.
The standard dipyridamole dose is 0.56mg/kg administered over four minutes. If no ischemia is induced after four minutes without perfusion, an additional 0.28mg/kg is administered over two minutes, followed if necessary by atropine (in doses of 0.25mg up to a maximum of 1mg). The standard adenosine protocol consists of an intravenous infusion of 0.14mg/kg/minute over four to six minutes.
The safety of dipyridamole and adenosine SEcho is well established. The risk of major adverse reactions is lower than dobutamine SEcho, at approximately one in 1,000.5 The most common side effects include headache and dyspnea. Since they have a bronchoconstrictor activity, both techniques are contraindicated in patients with untreated atrioventricular block and bronchospastic disorders. Abstinence from xanthene-containing foods and beverages is required before the test.
Aminophylline (240mg intravenously) reverses dipyridamole-related adverse effects and should be available for immediate use during the test. Although reversal of dipyridamole with aminophylline is not otherwise usually needed, some SEcho laboratories propose that aminophylline should be routinely given at the end of the test, independent of the result.
Interpretation of Stress Echocardiography
Standard SEcho protocol requires a full echocardiographic evaluation in resting condition. After this, the echocardiogram is continuously monitored and a digital record is intermittently captured at each stage of the protocol and during the recovery phase or administration of the antidote. Parasternal long-axis and short-axis views, as well as standard apical views (four-chamber, two-chamber, and long-axis) are the echocardiographic projections usually used. The interpretation of SEcho is usually based on a subjective assessment of regional wall motion, comparing wall thickening and endocardial excursion in a sideby- side display of baseline and stress images. Clear endocardial definition is crucial for optimal regional function evaluation, which is performed using a five-point wall-motion scoring system (1 = normokinesis; 2 = mild hypokenesis; 3 = moderate or severe hypokinesis; 4 = akynesis; and 5 = dyskenesis) for the 16- or 17- segment model of the left ventricle.
As with rest echocardiography, patient-dependent factors such as obesity and lung disease may lead to poor acoustic windows and reduce diagnostic accuracy. In these patients, contrast-enhanced endocardial border detection may be used to improve visualization. Evaluation of wall motion abnormalities is also challenging in patients with previous myocardial infarction, in whom passive tethering motion is a confounding variable. Finally, signal dropout can cause suboptimal images, leading to misdiagnosis in some patients.
Regardless of the form of stress, the normal response is a global increase in contractility leading to hyperdynamic wall motion. Lack of hyperkynesis is abnormal and may be caused by regional myocardial ischemia, non-ischemic cardiomyopathy, beta-blocker treatment, left bundle branch block, or severe hypertension. An ischemic response is recognized whenever a segment with normokinesis at rest becomes hypokinetic, akinetic, or dyskinetic during stress. SEcho is considered positive if stress-induced abnormalities are found in at least two adjacent segments.
Regarding regions with abnormal resting function, four possible responses during stress may be found:
- biphasic response—at low levels of stress, systolic wall thickening increases and starts earlier, improving contractile function; at higher levels, however, the increase in myocardial demand cannot be matched by further increases in blood flow, leading to ischemia and systolic function deterioration;
- sustained functional improvement at low stress that persists or is further improved until peak stress;
- worsening of resting wall motion during stress without any improvement; and
- no change in function.
The SEcho sign of myocardial viability is a stress-induced improvement of contractile function during low levels of stress in a region that is abnormal at rest. The pattern of response is predictive of post-revascularization functional improvement. A biphasic response, indicating that the tissue is not only viable but also supplied by a stenosed artery, has greatest predictive accuracy for recovery. In a recent study,6 72% of segments with biphasic response recovered function. A uniphasic response with sustained improvement has limited specificity to predict functional recovery, since augmentation alone may occur not only with non-jeopardized myocardium but also in areas of non-transmural infarction without hibernating myocardium (subendocardial scar) or in remodeled myocardium.
New Technologies Applied to Stress Echocardiography
The state-of-the-art diagnosis of ischemia and myocardial viability in SEcho remains the qualitative analysis of regional wall motion. The major potential drawback for use of this index is semi-quantitative assessment of wall motion, which is limited by subjectivity and technical challenges. In fact, the clinical acuity of SEcho evaluation does not depend on the stress modality used, since this is appropriate to patient characteristics, but rather depends on the quality of the echocardiographic window and the experience of the echocardiographer. Considerable expertise is required to interpret SEcho images accurately. This learning curve precludes the use of SEcho by all but experienced operators. Even among expert readers, although the concordance of interpretations within the same echo laboratory is high, concordance among different centres may be less than 80%.7 Inter- and intraobserver wall motion score variability is even greater in patients with poor images and in those with previous myocardial infarction due to pre-existing wall motion abnormalities and intraventricular conduction defects.
In recent years, quantitative parameters have been studied to provide objective and reproducible information on global and regional wall function during stress. The most important methods aiming to improve SEcho diagnostic accuracy and reduce its operator dependency are:
- tissue Doppler imaging (TDI) and derived strain and strain rate measurements;
- 2D strain based on speckle tracking; and
- automatic contour techniques, including acoustic quantification and colour kinesis.
TDI measures the low-frequency, high-amplitude signals of myocardial tissue motion, allowing the assessment of myocardial velocities. Yamada et al.8 showed that normal segments have a substantial augmentation in myocardial systolic velocity during stress, which is significantly less pronounced in ischemic and scarred segments. Myocardial velocity profiles are, however, unable to discriminate passive motion from active deformation. The post processing of color-coded TDI allows quantification of myocardial deformation by measuring strain and strain rate. These TDI-derived techniques assess velocity gradients between different points in space, allowing the evaluation of active contraction of a given segment independent of local tethering of the neighboring regions.
Strain and strain rate reflect, respectively, the percentage of deformation and intrinsic rate of deformation of the analyzed myocardial segment. These parameters are less dependent on image quality and less subjective than the visual assessment of endocardial border motion. Moreover, TDI-derived techniques use high frame rates (>100 frames/second), allowing a temporal resolution in the assessment of systolic and diastolic deformation that is not achievable by other methods, namely 2D strain.
Voigt and colleagues9 reported that the ratio of strain rate post-systolic shortening to maximal segmental deformation was the best TDI-derived parameter to detect ischemia during SEcho, with a sensitivity (82%) and specificity (85%) comparable to an experienced observer. This study only enrolled 44 patients, however, and subsequent studies did not confirm such a high value of post-systolic shortening for detection of stress-induced ischemia.10,11 In fact, Weidermann et al.10 evaluated the performance of TDI parameters during DSE in patients with intermediate coronary lesions, considering as a reference method of hemodynamic significance the invasive myocardial fractional flow reserve.
They found that pathological and normal flow reserve may be best differentiated using the change of peak systolic strain rate, which does not increase during stress in ischemic segments. Although ischemic response was characterized by an increase in post-systolic strain associated with a reduction in systolic strain, that parameter had lower acuity for the detection of subclinical ischemia in this study.
Speckle tracking is a new technique that tracks frame-to-frame movement of natural acoustic markers identified on standard 2D images. Deformation can be determined from temporal differences in the mutual distance of neighbouring speckles, allowing the evaluation of circumferential, radial, and longitudinal strain without the influence of the angle of incidence. 2D strain is particularly attractive for contractile function quantification in SEcho, since speckle tracking analysis may be performed on 2D images collected in standard clinical practice. There are some pitfalls, however, since speckle tracking is influenced by image quality and is limited to a frame rate of 70s-1, which may be insufficient at peak stress.
Although 2D strain has been validated in various circumstances, there are few studies supporting its use in SEcho. Recently, Hanekom et al.11 compared the feasibility and accuracy of TDI strain rate and 2D strain in 150 patients undergoing DSE and coronary angiography. In order to make the technique feasible in daily practice, they studied single sentinel segments at the distal territory of each epicardial coronary artery. The most feasible method was 2D strain at rest and TDI strain at peak stress. The parameter with the best accuracy for detection of CAD was peak systolic strain rate for both methods. The accuracy of 2D strain and TDI at peak stress was similar in left anterior descending (LAD) territory, but TDI parameters were substantially better in the right coronary and left circumflex territories. Neither 2D strain nor TDI strain rate provided incremental accuracy to wall motion conventional analysis.
More recently, Ingul and colleagues12 investigated myocardial deformation during DSE in 197 patients by automated analysis based on velocity gradient and segmental length methods of measuring longitudinal motion within a region of interest tracked through the cardiac cycle. They found that automated strain rate imaging is feasible and accurate, the performance in the posterior circulation not being so good as in the LAD territory. Moreover, it increased the sensitivity of DSE compared with expert conventional reading (sensitivity 75%): the velocity and segment length methods had respective sensitivities of 87 and 84% for strain rate, and 87 and 88% for end-systolic strain.
Technological advances in transducer and computer technology have led to introduction of realtime 3D echocardiography during stress.13,14 No data currently show the additional value of this technique over conventional SEcho.
Several studies have shown that assessment of coronary flow reserve (CFR) during vasodilator SEcho as a complement to wall motion analysis increases acuity for diagnosing CAD and adds power of prognostication. In fact, a normal CFR as evaluated by pulsed Doppler flowmetry of mid-distal LAD artery increases the predicted value of a negative test for wall motion criteria.15 Reduced CFR, however, is an additional parameter of severity in risk stratification.16 The European Society of Echocardiography has recommended assessment of CFR and wall motion during vasodilator SEcho as complementary strategies whenever suitable technology and dedicated expertise are available.5
Myocardial contrast echocardiography (MCE) is a technique that uses microbubbles during echocardiography. These microbubbles remain exclusively within the intravascular space and their presence within any myocardial territory denotes the status of microvascular perfusion within that region. Recently, several authors have shown that by adding MCE analysis to wall motion assessment during SEcho, the sensitivity in detecting relevant CAD increases, especially in patients who do not reach the target heart rate during stress and in patients with multivessel disease.17 Moreover, although analysis of MCE is commonly performed on a visual subjective basis, it is less vulnerable to interobserver variability than wall motion assessment.18
Assessment of Myocardial Viability
Systolic left ventricular dysfunction due to CAD is the complex result of necrosis and scarring, but also of functional and morphological adaptive abnormalities of the viable myocardium. In the setting of chronic left ventricular dysfunction, myocardium viability usually refers to the downregulation of contractile function in surviving myocardium in response to periodic or sustained reduction in coronary blood flow, which may be reversed if normal blood flow is restored. Viable myocardium exists as a spectrum, from complete transmural infarction with no viability to transmural hibernation or stunning with potential for full recovery. As roughly 40% of myocardial segments with resting wall motion abnormalities after acute myocardial infarction have viable tissue that may recover contractile function if revascularized, detection of viable myocardium is clinically relevant.
A number of non-invasive imaging procedures have been developed to evaluate myocardial viability and to identify markers of functional recovery, including DSE, MCE, SPECT, positron emission tomography, and cardiovascular magnetic resonance imaging.
SEcho is based on evaluation of contractile reserve, a characteristic feature of viable myocardium that may be elicited by catecholamine stimulation. The underlying principle is that adrenoreceptor stimulation by dobutamine will augment function before ischemia is engendered by increased myocardial work and metabolic demands. Typically, primarily inotropic response occurs at low doses. Tachycardia, potentially eliciting ischemia, usually develops only at higher doses.
Alternative protocols for echocardiographic assessment of myocardial viability include dipyridamole, low-level exercise, and, more recently, levosimendan. Among them DSE is the most extensively studied and the only one that has class 1 indication for viability assessment in guidelines.19,20
From experienced laboratories, DSE using visual wall motion assessment demonstrated a mean sensitivity of 85% and specificity of 79% in regional functional recovery prediction.20 Moreover, a linear relation was present between the number of viable segments and the likelihood of recovery of overall left ventricular function after revascularization. The identification of four or more viable segments accurately predicted LVEF improvement (e.g. ≥5%) after revascularization (sensitivity 86%, specificity 90%), improvement in heart failure symptoms, and reduction in event rate.21
Recently, TDI parameters have been used to better quantify regional myocardial function during DSE in patients undergoing revascularization. Aggeli and co-workers22 showed that pre-ejection longitudinal tissue velocity change and peak systolic longitudinal velocity change assessed by pulsed-wave TDI during low-dose dobutamine are reliable parameters of myocardial viability, predicting recovery after revascularization. Hanekom et al.23 demonstrated that strain rate imaging as an adjunct to routine visual wall-motion scoring during conventional DSE provides incremental value to predict regional and global functional recovery following revascularization, increasing sensitivity from 73 to 83% without affecting the specificity. Further experimental and clinical studies have validated strain rate imaging for the assessment of myocardial viability and suggest that strain rate is a better quantitative parameter for the prediction of functional recovery compared with strain.24
Although 2D strain is considered a promising tool to improve DSE accuracy in the assessment of myocardial viability, there are not yet clinical studies published in this setting. Becker and co-workers25 showed recently, however, that 2D strain in rest allows the assessment of transmurality, since radial and circumferential strain impairment is proportional to the extent of transmural scarring. Moreover, they also found that peak systolic radial strain identifies reversible myocardial dysfunction and predicts regional and global functional recovery at 9±2 months follow-up. Moreover, the predictive value was similar to that achieved by contrast-enhanced cardiac magnetic resonance.
Assessment of Coronary Artery Disease
Exercise ECG stress testing is the most widely used test for detection of CAD and assessment of prognosis. Although relatively inexpensive, its use is limited by suboptimal sensitivity for detecting single-vessel disease and poor specificity in the presence of resting ST abnormalities. Exercise or pharmacological SPECT and SEcho compare favorably with ECG stress testing since they provide enhanced accuracy for detecting CAD. The advantages of SPECT include a higher feasibility, higher sensitivity (especially for single-vessel disease involving the left circumflex), and higher accuracy in the presence of extensive resting wall motion abnormalities. The advantages of SEcho include higher specificity, wider availability, lower cost, and, most importantly, its radiation-free nature.5 Recently, the recommendations of the European Association of Echocardiography defended that SEcho compares favorably with SPECT, considering the integrated risk–benefit balance.5 All SEcho methods (treadmill exercise, bicycle exercise, high-dose dobutamine, and high-dose dipyridamole) have similar diagnostic and prognostic accuracy. The choice of one over the other is strictly due to relative contraindications and the ability of patients to exercise.5,26 Among the available stressors exercise is the most used, dobutamine the best test for viability, and dipyridamole the safest pharmacological stressor that is the most suitable for combined wall motion CFR assessment.5
In conclusion, SEcho has a very high diagnostic accuracy for detecting ischemia and viability. Its wide acceptance in clinical practice reflects its safety and prognostic value, which has also been proved in several large-scale multicenter trials. The recent introduction of other methods, such as myocardial deformation assessment and contrast echocardiography, have further expanded the use of SEcho as one of the main tools in the management of patients with ischemic heart disease.