Harrison's Cardiovascular Medicine 2 ed.


Rick A. Nishimura image Panithaya Chareonthaitawee image Matthew Martinez

Cardiovascular imaging plays an essential role in the practice of cardiology. Two-dimensional (2D) echocardiography is able to visualize the heart directly in real time using ultrasound, providing instantaneous assessment of the myocardium, cardiac chambers, valves, pericardium, and great vessels. Doppler echocardiography measures the velocity of moving red blood cells and has become a noninvasive alternative to cardiac catheterization for assessment of hemodynamics. Transesophageal echocardiography (TEE) provides a unique window for high-resolution imaging of posterior structures of the heart, particularly the left atrium, mitral valve, and aorta. Nuclear cardiology uses radioactive tracers to provide assessment of myocardial per-fusion and metabolism, along with ventricular function, and is applied primarily to the evaluation of patients with ischemic heart disease. Cardiac MRI and CT can delineate cardiac structure and function with high resolution. They are particularly useful in the examination of cardiac masses, the pericardium, the great vessels, and ventricular function and perfusion. Gadolinium enhancement during cardiac MRI adds information on myocardial perfusion. Detection of coronary calcification by CT as well as direct visualization of coronary arteries by CT angiography (CTA) may be useful in selected patients with suspected coronary artery disease (CAD). This chapter provides an overview of the basic concepts of these cardiac imaging modalities as well as the clinical indications for each procedure.



Basic principles

2D echocardiography uses the principle of ultrasound reflection off cardiac structures to produce images of the heart (Table 12-1). For a transthoracic echocardiogram (TTE), the imaging is performed with a handheld transducer placed directly on the chest wall. In selected patients, a TEE may be performed, in which an ultra-sound transducer is mounted on the tip of an endoscope placed in the esophagus and directed toward the cardiac structures.

TABLE 12-1



Current echocardiographic machines are portable and can be wheeled directly to the patient’s bedside. Thus, a major advantage of echocardiography over other imaging modalities is the ability to obtain instantaneous images of the cardiac structures for immediate interpretation. Thus, echocardiography has become an ideal imaging modality for cardiac emergencies. A limitation of TTE is the inability to obtain high-quality images in all patients, especially those with a thick chest wall or severe lung disease, as ultrasound waves are poorly transmitted through lung parenchyma. Technology such as harmonic imaging and IV contrast agents (which traverse the pulmonary circulation) can be used to enhance endocardial borders in patients with poor acoustic windows.

Chamber size and function

2D echocardiography is an ideal imaging modality for assessing left ventricular (LV) size and function (Fig. 12-1). A qualitative assessment of the ventricular cavity and systolic function can be made directly from the 2D image by experienced observers. 2D echocardiography is useful in the diagnosis of LV hypertrophy and is the imaging modality of choice for the diagnosis of hypertrophic cardiomyopathy. Other chamber sizes are assessed by visual analysis, including the left atrium and right-sided chambers.



Two-dimensional echocardiographic still-frame images from a normal patient with a normal heart. Upper: Parasternal long-axis view during systole and diastole (left) and systole (right). During systole, there is thickening of the myocardium and reduction in the size of the left ventricle (LV). The valve leaflets are thin and open widely. Lower: Parasternal short-axis view during diastole (left) and systole (right) demonstrating a decrease in the left ventricular cavity size during systole as well as an increase in wall thickening. LA, left atrium; RV, right ventricle; Ao, aorta.

Valve abnormalities

2D echocardiography is the “gold standard” for imaging valve morphology and motion. Leaflet thickness and mobility, valve calcification, and the appearance of subvalvular and supravalvular structures can be assessed. Valve stenosis is reliably diagnosed by the thickening and decreased mobility of the valve. 2D echocardiography is also the gold standard for the diagnosis of mitral stenosis, which produces typical tethering and diastolic doming, and the severity of the stenosis can be ascertained from a direct planimetry measurement of the mitral valve orifice. The presence and often the etiology of stenosis of the semilunar valves can be made by 2D echocardiography (Fig. 12-2), but evaluation of the severity of the stenosis requires Doppler echocardiography (discussed later). The diagnosis of valvular regurgitation must be made by Doppler echocardiography, but 2D echocardiography is valuable for determining the etiology of the regurgitation, as well as its effects on ventricular dimensions, shape, and function.



Two-dimensional echocardiographic still-frame images from a patient with aortic stenosis. Parasternal long-axis view shows a heavily calcified aortic valve. RV, right ventricle; LV, left ventricle; Ao, aorta; LA, left atrium.

Pericardial disease

2D echocardiography is the imaging modality of choice for the detection of pericardial effusion, which is easily visualized as a black echolucent ovoid structure surrounding the heart (Fig. 12-3). In the hemodynamically unstable patient with pericardial tamponade, typical echo findings include a dilated inferior vena cava, right atrial collapse, and then right ventricular collapse. Echocardiographically guided pericardiocentesis has now become a standard of care.



Two-dimensional echocardiographic still-frame image of a patient with a pericardial effusion. Pericardial effusion (PE) is shown as a black echo-free space surrounding the heart. LV, left ventricle.

Intracardiac masses

Intracardiac masses can be visualized on 2D echocardiography, provided that image quality is adequate. Solid masses appear as echo-dense structures, which can be located inside the cardiac chambers or infiltrating into the myocardium or pericardium. LV thrombus appears as an echo-dense structure, usually in the apical region associated with regional wall motion abnormalities. The appearance and mobility of the thrombus are predictive of embolic events. Vegetations appear as mobile linear echo densities attached to valve leaflets. Atrial myxoma can be diagnosed by the appearance of a well-circumscribed mobile mass with attachments to the atrial septum (Fig. 12-4). The high-resolution images provided by TEE may be required for further delineation of myocardial masses, especially those <1 cm in diameter.



Transesophageal still-frame echocardiographic images of a patient with a left atrial myxoma. There is a large echo-dense mass in the left atrium, attached to the atrial septum. The mass moves across the mitral valve in diastole. LV, left ventricle; RV, right ventricle.

Aortic disease

2D echocardiography can provide extremely useful information on diseases of the aorta. The proximal ascending aorta, the arch, and the distal descending aorta can usually be visualized via the transthoracic approach. The definitive diagnosis of a suspected aortic dissection usually requires a TEE, which can rapidly provide high-resolution images of the proximal ascending and descending thoracic aorta (Fig. 12-5).



Transesophageal still-frame echocardiographic view of a patient with a dilated aorta, aortic dissection, and severe aortic regurgitation. The arrow points to the intimal flap that is seen in the dilated ascending aorta. Left: The long-axis apex-down view of the black-and-white two-dimensional image in diastole. Right: Color-flow imaging that demonstrates a large mosaic jet of aortic regurgitation. Ao, aorta; RV, right ventricle; AR, aortic regurgitation.


Basic principles

Doppler echocardiography uses ultrasound reflecting off moving red blood cells to measure the velocity of blood flow across valves, within cardiac chambers, and through the great vessels. Normal and abnormal blood flow patterns can be assessed noninvasively. Color-flow Doppler imaging displays the blood velocities in real time superimposed upon a 2D echocardiographic image. The different colors indicate the direction of blood flow (red toward and blue away from the transducer), with green superimposed when there is turbulent flow. Pulsed-wave Doppler measures the blood flow velocity in a specific location on the 2D echocardiographic image. Continuous-wave Doppler echocardiography can measure high velocities of blood flow directed along the line of the Doppler beam, such as occur in the presence of valve stenosis, valve regurgitation, or intracardiac shunts. These high velocities can be used to determine intracardiac pressure gradients by a modified Bernoulli equation:

Pressure change = 4 times (velocity)2

Tissue Doppler echocardiography measures the velocity of myocardial motion. Myocardial velocities can be used to determine myocardial strain rate, which is a quantitative measure of regional myocardial contraction and relaxation.

Valve gradients

In the presence of valvular stenosis, there is an increase in the velocity of blood flow across the stenotic valve. A continuous-wave Doppler can be used to determine the pressure gradient across the valve (Fig. 12-6). A valve area can also be calculated from the Doppler velocities.



Continuous-wave Doppler of mitral valve velocities in a patient with mitral stenosis. The mean gradient calculated from Doppler (DOPP) of 10 mmHg is similar to the mean gradient of 11 mmHg from simultaneous cardiac catheterization in this patient.

Valvular regurgitation

Valvular regurgitation is diagnosed by Doppler echo-cardiography when there is abnormal retrograde flow across the valve. Color-flow imaging is the Doppler method used most frequently to detect valve regurgitation by visualization of a high-velocity turbulent jet in the chamber proximal to the regurgitant valve (Fig. 12-7). The size and extent of the color-flow jet into the receiving cardiac chamber provide a semiquantitative estimate of the severity of regurgitation.



Left: Transesophageal echocardiographic view of a patient with severe mitral regurgitation due to a flail posterior leaflet. The arrow points to the portion of the posterior leaflet that is unsupported and moves into the left atrium during systole. Right: Color-flow imaging demonstrating a large mosaic jet of mitral regurgitation during systole. LA, left atrium; LV, left ventricle; AV, aortic valve.

Intracardiac pressures

These can be calculated from the peak continuous-wave Doppler signal of a regurgitant lesion, which reflects the pressure gradient between two cardiac chambers. This approach is commonly applied to a tricuspid regurgitant jet, from which the systolic pressure gradient between the right atrium and right ventricle can be calculated, yielding an accurate measurement of pulmonary artery systolic pressure (Fig. 12-8).



Continuous-wave Doppler of tricuspid regurgitation in a patient with pulmonary hypertension. There is an increase in the velocity to 5.4 m/s. Using the modified Bernoulli equation, the peak pressure gradient between the right ventricle and right atrium during systole is 116 mmHg. Assuming a right atrial pressure of 10 mmHg, the right ventricular systolic pressure is 126 mmHg. In the absence of right ventricular outflow tract obstruction, this indicates there is severe pulmonary hypertension with a pulmonary artery systolic pressure of 126 mmHg.

Cardiac output

Volume flow rates (or stroke volume and cardiac output) can be reliably measured noninvasively by Doppler echocardiography. Flow is calculated as the product of the cross-sectional area of the vessel or chamber through which blood moves and the velocity of blood flow as assessed by Doppler.

Diastolic filling

Doppler echocardiography allows noninvasive evaluation of ventricular diastolic filling. The transmitral velocity curves reflect the relative pressure gradients between the left atrium and ventricle throughout diastole and are influenced by the rate of ventricular relaxation, the driving force across the valve, and the compliance of the ventricle. In the early phase of diastolic dysfunction there is primarily an impairment of LV relaxation, with reduced early transmitral flow and a compensatory increase in flow during atrial contraction (Fig. 12-9). As disease progresses and ventricular compliance declines, left atrial pressure rises, resulting in a higher early transmitral velocity and shortening of the deceleration of flow in early diastole. Analysis of Doppler tissue velocities of annular motion and myocardial strain provides further information concerning the diastolic properties of the heart.



High-fidelity left ventricular (LV) pressure curves superimposed on a mitral inflow velocity curve obtained by Doppler echocardiography. The ratio of early and late diastolic flows is termed the E:A ratio. The deceleration time (DT) measures the rate of decline of early velocity and reflects the effective operative compliance of the left ventricle. Left: In early stages of diastolic dysfunction, there is an abnormality of relaxation. There is a decrease in the early diastolic filling and an increase with filling at atrial contraction, resulting in a low E:A ratio of 0.5, with a deceleration time (DT) of 250 ms. In this instance, the LV diastolic pressure is low at 6 mmHg. Right: As diastolic dysfunction progresses, there is a restriction to filling, in which there is a high early diastolic velocity and low velocity at atrial contraction resulting in a high E:A ratio of 3.0, with DT of 150 ms. In this instance, the LV diastolic pressure is markedly elevated to 34 mmHg.

Congenital heart disease

2D and Doppler echocardiography have been useful in the evaluation of patients with congenital heart disease. Congenital stenotic or regurgitant valve lesions can be assessed. The detection of intracardiac shunts is possible by 2D and Doppler echocardiography. Patency of surgical shunts and conduits can also be evaluated.


2D and Doppler echocardiography are usually performed with the patient in the resting state. Further information can be obtained by reimaging during either exercise or pharmacologic stress. The primary indications for stress echocardiography are to confirm the suspicion of ischemic heart disease and determine the extent of ischemia.

A decrease in systolic contraction of an ischemic area (segment) of myocardium, termed a regional wall motion abnormality, occurs before symptoms or electrocardiographic changes (Fig. 12-10). New regional wall motion abnormalities, a decline in ejection fraction, and an increase in end-systolic volume with stress are all indicators of myocardial ischemia. Exercise stress testing is usually done with exercise protocols using either upright treadmill or bicycle exercise. In patients who are not able to exercise, pharmacologic testing can be performed by infusion of dobutamine to increase myocardial oxygen demand. Dobutamine echocardiography has also been used to assess myocardial viability in patients with poor systolic function and concomitant CAD; when used for this purpose, dobutamine is administered at lower doses than standard pharmacologic stress doses.


FIGURE 12-10

Systolic still-frame two-dimensional echocardiographic images of a patient undergoing a stress echocardiogram. During rest (left), there is contraction of all segments of the myocardium. During exercise (right), there are regional wall motion abnormalities in the anterior and anteroapical segments (arrows). 4 ch = four-chamber view, 2 ch = two-chamber views, LV = left ventricle, RV = right ventricle.

Doppler echocardiography can be used at rest and during exercise in patients with valvular heart disease to determine the hemodynamic response of valve gradients and pulmonary pressures (Fig. 12-11). In patients with low-output, low-gradient aortic stenosis, the response of the gradient to dobutamine stimulation is of diagnostic and therapeutic value.


FIGURE 12-11

Continuous-wave Doppler echocardiogram across the mitral valve of a patient with mitral stenosis. In the resting state (left), there is a mean gradient of 8 mmHg. During exercise (right), the mean gradient rises to 29 mmHg, indicating a hemodynamically significant mitral stenosis.


When limited information is obtained from a TTE due to poor imaging windows, TEE can be useful. Diseases of the aorta, such as aortic dissection, can be readily diagnosed by TEE. Defining the source of embolism is a common indication for TEE, as abnormalities such as atrial thrombi, patent foramen ovale, and aortic plaques can be detected. Other masses, particularly those in the atria, can be visualized. The presence of vegetations for the diagnosis of infective endocarditis and its complications can be assessed by TEE. This technique has been used before cardioversion in patients with atrial fibrillation to rule out a thrombus in the left atrium or left atrial appendage.



Nuclear (or radionuclide) imaging requires intravenous administration of radiopharmaceuticals (isotopes or tracers). Once injected, the isotope traces physiologic processes and undergoes uptake in specific organs. During this process, radiation is emitted in the form of photons, generally gamma rays, generated during radioactive decay when the nucleus of an isotope changes from one energy level to a lower one. A special camera detects these photons and creates images via a computer interface. The two most commonly used technologies in clinical nuclear cardiology are single-photon emission computed tomography (SPECT) and positron emission tomography (PET). These technologies differ in instrumentation, acquisition, resolution, and nuclides used.


Assessment of myocardial perfusion and coronary artery disease

Nuclear myocardial perfusion imaging (MPI) using SPECT and more recently PET has an established role in the evaluation and management of patients with known or suspected coronary artery disease (CAD). Both SPECT and PET MRI require the injection of isotopes at rest and during stress to produce images of regional myocardial uptake proportional to regional blood flow. Normally, myocardial blood flow can be increased up to fivefold above the resting state to meet the increased myocardial oxygen demand during stress. In the presence of a fixed coronary stenosis, the inability to increase myocardial perfusion in the territory supplied by the stenosis creates a flow differential and inhomogeneous myocardial tracer uptake. In patients unable to exercise, pharmacologic agents are used to increase blood flow and create similar inhomogeneities.

The most commonly used SPECT perfusion tracers are thallium-201 (201Tl) and technetium-99m (99mTc) labeled isonitriles. 99mTc isonitriles have higher photon energies and shorter physical half-lives than 201Tl, permitting injection of higher doses with less radiation exposure while concurrently producing higher-quality images. The FDA-approved PET tracers are rubidium-82 (82Rb) and 13N ammonia (13NH3) for high-dose administration and shorter imaging protocols.

Both SPECT and PET myocardial perfusion images are commonly interpreted by visual analysis, which may be supplemented with quantitative software. Normal myocardial perfusion images demonstrate uniform tracer uptake throughout the LV myocardium (Fig. 12-12). In contrast, regions with reduced myocardial blood flow demonstrate varying degrees of reduced tracer uptake (Fig. 12-13), which can be graded on a semiquantitative scale. Reduced tracer uptake in a myocardial region on both resting and stress images is called a fixed defect and is consistent with infarction. Reduced tracer uptake on the stress image with relatively preserved or improved uptake on the rest image is called a reversible defect and indicates ischemia. PET has the ability to quantify myocardial blood flow and flow reserve in absolute terms.


FIGURE 12-12

Exercise technetium-99m sestamibi images in a 65-year-old man with atypical angina. Images are shown in three standard views; stress (left) and rest (right) in each panel. There is uniform tracer uptake throughout the left ventricular myocardium at rest and peak stress in all three views.


FIGURE 12-13

Exercise technetium-99m sestamibi and rest thallium-201 images in a 72-year-old woman with typical angina. Images are shown in three standard views, with stress (left) and rest (right) in each panel. Stress images demonstrate reduced tracer uptake in the apical, mid-anterior, mid-lateral, and mid-inferior regions (white arrowheads) with normal or near-normal tracer uptake in the corresponding regions on the rest images (white arrowheads), signifying a reversible defect consistent with ischemia. The lack of complete normalization (or reversibility) of tracer uptake on the rest images at the mid-inferior and mid-lateral regions represents associated infarction in that area (yellow arrowheads). On both stress and rest images, the basal inferior and basal lateral regions exhibit severely reduced tracer uptake, signifying a fixed defect consistent with infarction (red arrowheads). Subsequent invasive coronary angiography demonstrated severe stenosis of the mid-left anterior descending coronary artery and occlusion of the left circumflex coronary artery with collaterals.

For the diagnosis of angiographically significant CAD, SPECT using 201T1 and 99mTc isonitriles and either exercise or pharmacologic stress has an average sensitivity of 87% and specificity of 73%. In comparison, PET MPI has higher accuracy (average sensitivity 90%; specificity of 89%). The robust methods for attenuation correction with PET improve the specificity, particularly in obese populations and women, while the superior resolution and higher extraction fraction of PET tracers increase the sensitivity (Fig. 12-14). PET has not been as widely used as SPECT due to decreased availability and less local experience, but PET scanners are becoming more widely available (Table 12-2).


FIGURE 12-14

SPECT and PET images in a 67-year-old woman with atypical angina. Images are shown in short-axis views, with stress (top) and rest (bottom) in each panel. Shifting breast position between the rest and stress SPECT acquisitions produced an apparent reversible apical, anterior, and anterolateral attenuation artifact (arrowheads) resembling ischemia. With PET and its built-in attenuation correction in the same patient, the defect was not present. SPECT, single-photon emission computed tomography; PET, positron emission tomography.

TABLE 12-2



Both SPECT and PET MPI have powerful prognostic value. In patients with normal SPECT MPI results, the annual rate of cardiac death or myocardial infarction is generally very low (<0.7%). Annual death/event rates increase with the extent and severity of imaging abnormalities and are generally about 3% in those with mild to moderate abnormalities and about 7% in those with severe abnormalities; rates are higher in specific populations such as diabetics and those with high-risk exercise treadmill results. High-risk SPECT MPI findings include severe resting or poststress LV systolic dysfunction, large or multiple stress-induced defects, or a large fixed defect with LV dilation or increased 201Tl lung uptake. The incremental prognostic value of SPECT MPI has been established in many clinical settings, including populations with known CAD, prior myocardial infarction and/or revascularization, and acute chest pain in the emergency department.

Assessment of myocardial metabolism and viability

PET has traditionally been regarded as the gold standard technique for the assessment of myocardial viability. The positron-emitting tracer F-18 fluorodeoxyglucose (FDG) assesses myocardial glucose metabolism and is an indicator of myocardial viability. Because uptake is heterogeneous in normal myocardium in the fasting state, oral glucose loading or a combination of insulin and glucose infusions is used to enhance myocardial uptake. With reduced myocardial blood flow and ischemia, substrate utilization switches from fatty acids and lactate toward glucose, leading to enhanced myocardial FDG uptake. This pattern of enhanced FDG uptake in regions of decreased perfusion (termed flow/metabolism “mismatch”) identifies areas of ischemic or hibernating myocardium that are likely to improve in function after revascularization (Fig. 12-15). This mismatch has a sensitivity and specificity of 92% and 63%, respectively, for regional contractile recovery after revascularization. The SPECT radiopharmaceuticals, 201Tl and 99mTc isonitriles, require an intact (viable) cell membrane for uptake and also provide an assessment of myocardial viability in addition to perfusion. However, PET identifies ischemic or hibernating myocardium in 10–20% of regions otherwise classified as fibrotic (infarcted) by SPECT perfusion tracers. Patients with ischemic heart failure, who have viable myocardium identified by PET or SPECT and undergo revascularization, have a better survival than those who do not have viable myocardium or do not undergo revascularization.


FIGURE 12-15

PET viability study in a 63-year-old woman with heart failure, severe lV systolic dysfunction, and severe coronary artery disease. Images are shown in three standard views, with perfusion (left) and glucose metabolism (right) in each panel. The N-13 ammonia images show a very large apical, septal, anterior, and lateral perfusion defect (arrowheads), but F-18 fluorodeoxyglucose (18FDG) images demonstrate relatively preserved glucose uptake in the corresponding segments (arrowheads). This PET perfusion-metabolism mismatch is consistent with hibernating myocardium. The patient underwent coronary artery bypass grafting surgery with improvement in left ventricular systolic function (ejection fraction increased from 26% pre- to 45% postoperatively). All regions identified as viable recovered contractile function after revascularization.

Assessment of ventricular function

In addition to perfusion and metabolic information, LV systolic function and volumes are now routinely obtained with gated SPECT and PET acquisitions, as long as the heart rate is relatively constant. An automated technique determines the endocardial borders of the LV cavity, and a geometric model is used to calculate the LVEF and volumes with a high level of reproducibility. Regional wall motion can also be assessed by visual examination. The combined variables of perfusion and function are more effective in risk stratification than either alone.

Another established but less widely available nuclear technique for assessing LV function and volumes is equilibrium radionuclide angiography RNA, also known as multiple-gated blood pool acquisition (MUGA). This technique involves imaging of 99mTc-labeled albumin or red cells that are uniformly distributed throughout the blood volume. LV volumes throughout the cardiac cycle are calculated from a time-activity curve generated using regions of interest.

Innovations in hybrid imaging technology, especially PET/CT and SPECT/CT, are occurring rapidly and contribute to their emerging role in the combined assessments of anatomy and physiology in patients with suspected or known CAD. The diagnostic literature is evolving for these hybrid technologies but radiation exposure is a concern and large-scale clinical trials are still needed to validate their clinical applications, determine their prognostic value, and address their cost-effectiveness and appropriateness.



Basic principles

MRI is a technique based on the magnetic properties of hydrogen nuclei. In the presence of a large magnetic field, nuclear spin transitions from the ground state to excited states can be induced by an electric field, and as the nuclei relax and return to their ground state, they release energy in the form of electromagnetic radiation that is detected and processed into an image. Although the large vascular vessels can be visualized on MRI without contrast agents, gadolinium is frequently employed as a contrast agent to produce magnetic resonance angiograms (MRAs). Contrast agents also provide enhanced soft tissue contrast as well as the opportunity to obtain rapid angiographic images during the first pass of contrast through the vascular system.

Cardiac MRI is challenging because of the rapid motion of the heart and coronary arteries. However, both static and cine images can usually now be obtained using electrocardiographic triggering, often within short breath holds of 10–15 s. Cine images can be acquired in any plane with excellent blood-myocardial contrast. These images can be used to quantify accurately ejection fraction, end-systolic and end-diastolic volumes, and cardiac mass with high accuracy, reliability, and reproducibility, and without the need for ionizing radiation.

Clinical utility

The multiplanar capabilities of MRI, coupled with excellent contrast and spatial resolution, provide superb images of the myocardium and great vessels. MRI is of great value in defining anatomic relationships in patients with complex congenital heart disease (Fig. 12-16) and cardiomyopathies (Fig. 12-17). Cardiac masses can be characterized and distinguished from thrombus (Fig. 12-18). In addition to defining their relationship to normal anatomic structures, MRI can determine whether a mediastinal or pulmonary mass has invaded the pericardium or heart. The entire pericardium can be visualized in multiple planes, and MRI has proved useful in characterizing pericardial effusions, pericardial thickening, and inflammation. Specialized pulse sequences can measure the velocity of blood in each pixel of the image, so that flow across valves and within blood vessels may be determined with accuracy, thereby aiding in the evaluation of valvular disease and intracardiac shunts.


FIGURE 12-16

MRA scan of a patient with partial anomalous pulmonary venous drainage of the right lung into the inferior vena cava (scimitar syndrome). MRA is able to define the abnormal anatomic relationships of cardiac structures and great vessels in patients with congenital heart disease.


FIGURE 12-17

MRI scan of a patient with hypertrophic cardiomyopathy, showing the severe increase in left ventricular wall thickness. Cardiac MRI is an ideal imaging modality for diagnosing cardiomyopathies.

MRA is a standard technique for imaging the aorta and large vessels of the chest and abdomen, with results essentially identical to conventional angiography. MRA of the coronary arteries is a much more difficult challenge, both because of the small size of these vessels and because of their rapid and complex motion during the cardiac cycle; thus, coronary MRA is not yet a reliable clinical technique.

MRI is now an accepted technology for the evaluation of patients with suspected or known coronary disease. Ventricular function and wall motion can be assessed at rest and during infusion of inotropic agents. Assessment of myocardial perfusion can be performed by injecting a bolus of gadolinium contrast and then continuously scanning the heart as the gadolinium passes through the cardiac chambers and into the myocardium. Relative perfusion deficits are reflected as regions of low signal intensity within the myocardium. Pharmacologic stress (typically achieved with vasodilators) can be applied during perfusion imaging to detect physiologically significant coronary artery lesions. Myocardial per-fusion imaging with cardiac MRI is more sensitive than SPECT imaging for detecting subendocardial ischemia due to its enhanced spatial resolution.

Myocardial viability and infarction may be determined by imaging the heart 10–20 min after gadolinium injection, known as delayed enhancement magnetic resonance imaging. In normal myocardium, gadolinium cannot penetrate the membranes of the densely packed myocytes. Abnormal myocardial tissue accumulates excess gadolinium following intravenous injection, as ruptured myocyte membranes allow gadolinium to passively diffuse into intracellular space. In chronic infarction, the tissue concentration of gadolinium is increased due to an expansion of the intracellular space from collagenous scar (Fig. 12-18). Thus, delayed enhancement is indicative of nonviable or infarcted myocardium, the subendocardial versus transmural extent of which is accurately assessed by the high spatial resolution of MRI. The presence and pattern of gadolinium enhancement not only is useful for determining viability but also has prognostic value in the patient with an ischemic cardiomyopathy. “Myocardium at risk” following myocardial infarction can be assessed by examining the amount of myocardial edema, using T2-weighted sequences (Fig. 12-19).


FIGURE 12-18

MRI scan with delayed gadolinium enhancement in a patient with a large anteroapical infarction. The gadolinium (white area) accumulated in the extracellular space in the presence of cell death from myocardial infarction.


FIGURE 12-19

Left: Normal delayed enhancement and “edema”-sensitive images. Top (left): Delayed enhancement image illustrating normal black myocardium without infarction/fibrosis. Bottom (left): A triple inversion recovery sequence that is T2-weighted demonstrating normal homogenous-appearing gray myocardium. Right: A patient postmyocardial infarction and early revascularization without evidence of an infarction and an edematous myocardium in the septum. Top (right): Delayed enhanced with normal black myocardium without infarction or fibrosis. Bottom (right): A triple inversion recovery sequence illustrating edema in the septum without infarction. This is the area of “salvaged” myocardium.

Limitations of MRI

Relative contraindications to MRI include the presence of pacemakers, internal defibrillators, or cerebral aneurysm clips. A small percentage of patients are claustrophobic and unable to tolerate the examination within the relatively confined quarters of the magnet bore. Examination of clinically unstable patients and those undergoing stress testing is problematic, since close hemodynamic and electrocardiographic monitoring is difficult. Image quality in patients with significant arrhythmias is often limited. Patients with renal disease receiving gadolinium contrast may be at risk of developing nephrogenic systemic fibrosis, characterized by increased tissue deposition of collagen in the skin and development of fibrosis in skin and other organs.


Basic principles

CT is a fast, simple, noninvasive technique that provides images of the myocardium and great vessels with excellent spatial resolution and good soft tissue contrast. The development of electron-beam CT and multidetector-row CT have led to improved temporal resolution and routine imaging of the beating heart. Motion-free high-spatial-resolution images are now possible with multidetector CT technology (≥64 channel) that allows imaging of the coronary arteries.

Clinical applications

Cardiac CT has important clinical applications. Pericardial calcification is easily detected by CT (Fig. 12-20). CT is useful in characterizing cardiac masses, particularly those containing fat or calcium. The ability to detect small amounts of fat with high spatial resolution makes CT an attractive technique for imaging patients with suspected arrhythmogenic right ventricular dysplasia. Cine images can be used to evaluate wall motion and to determine ejection fraction, end-diastolic and end-systolic volumes, and cardiac mass.


FIGURE 12-20

CT scan showing pericardial calcification, seen as a white linear density anterior to the myocardium.

CT angiography (CTA) has demonstrated accuracy similar to MRA in imaging the aorta and great vessels, and CTA is the examination of choice in the evaluation of patients with suspected pulmonary embolus. CTA is an excellent imaging modality for the diagnosis of aortic dissection or penetrating ulcers. Complete visualization of the entire aorta and its branches is possible by CTA using a single contrast medial bolus injection.

Coronary calcification

Calcium in the coronary arteries occurs in atherosclerosis and is absent in the normal coronary artery (Fig. 12-21). CT is very sensitive for the detection of coronary artery calcification, and the absence of coronary calcification excludes significant epicardial coronary disease. The quantity of coronary calcification (coronary calcium score) is related to the severity of CAD and prognosis. However, the utility of CT calcification score in clinical practice in the asymptomatic patient is limited to those with a moderate risk of coronary heart disease in whom the result will change management.


FIGURE 12-21

CT scans of three patients showing the ability to detect coronary calcification. Left: Normal coronary arteries without calcification. Middle: Calcification in the left anterior artery (LAD). Right: Severe calcification in the LAD and circumflex (CX) arteries.

Contrast-enhanced CT angiography

With the high temporal and spatial resolution of multislice spiral CT, accurate assessment of luminal narrowing in the major branches of the coronary arteries is possible in selected patients. Studies at experienced centers have shown a sensitivity and specificity of >90% for detecting coronary artery lesions as compared to cardiac catheterization. The highest accuracy has been noted in the left main and the proximal portions of the left-sided coronary arteries with decreased sensitivities in the more distal segments and in the more rapidly moving right coronary artery (Fig. 12-22).


FIGURE 12-22

Three-dimensional volume rendered image of a contrast-enhanced CT angiogram demonstrating a normal left main coronary artery arising from the aorta and its two branches, the left anterior descending artery (left) and the circumflex artery (right).

The concept of “noninvasive coronary angiography” has generated great interest in CTA. However, as with any imaging modalities, CTA has technical limitations requiring proper patient selection and preparation. The integration of CTA into clinical practice requires knowledge of pretest diagnostic and prognostic data and the incremental information that will alter management. The well-accepted indication for coronary CTA is in the evaluation of suspected coronary artery anomalies for which CTA not only confirms the diagnosis but also shows the course of the arteries related to the great vessels (Fig. 12-23). For patients with chest pain syndromes, CTA is best used to rule out significant coronary disease, given its high negative predictive value. Thus, it is the patient with an intermediate pretest probability of CAD who cannot exercise or has uninterpretable or equivocal results on prior testing who would be best suited for CTA. The benefit of CTA in other groups of patients is still unclear.


FIGURE 12-23

Three-dimensional volume rendered image of a contrast-enhanced CT angiogram illustrating an anomalous left coronary artery arising from the right coronary artery and traveling posterior to the aorta.

Limitations of CT

Limitations of CT include its dependence on ionizing radiation (in contrast to MRI) and the need for iodinated contrast. Techniques to lower radiation doses continue to evolve, as the radiation doses for coronary CTA generally exceed those delivered during standard diagnostic cardiac catheterization. Fast or irregular heart rhythms and body motion limit the accuracy of CTA. Heavy calcification and artifacts from stents preclude accurate assessment of the severity of a stenosis.


TABLE 12-3




The choice of the optimal imaging modality for a particular patient should be based upon the major problem being addressed, other concomitant clinical questions, as well as the local expertise and equipment available in an institution. The clinical urgency and costs of each test also need to be considered. To ensure the effective use of cardiovascular imaging tools, Appropriateness Criteria have been developed by the national societies to examine the incremental clinical benefit of imaging modalities.


Left ventricular size and function

2D echocardiography is the primary imaging modality obtained for assessment of LV cavity size, systolic function, and wall thickness. Echocardiography can also provide concomitant information on valve function, pulmonary artery pressures, and diastolic filling, which are valuable in the patient presenting with possible heart failure. The disadvantage is poor endocardial resolution in some patients and the lack of reproducible quantitative measurements.

Equilibrium radionuclide angiography can provide an accurate quantitative measurement of LV volumes and function but is not widely available and cannot be used in patients with irregular rhythms. Gated SPECT and PET measure LV systolic function and volumes as a part of myocardial perfusion and/or viability imaging but also require relatively regular rhythm. Both MRI and CT scanning provide the highest quality resolution of the endocardial border and, thus, are the most accurate of all modalities. However, they are of higher cost, lack portability, and do not provide concomitant hemodynamic information as echocardiography does.

Valvular heart disease

2D and Doppler echocardiography provide both anatomic and hemodynamic information regarding valve disease, and are the first test of choice. MRI can also visualize valve motion and determine abnormal flow velocities across valves, but there is less validation of quantitative hemodynamic measurements in comparison to echocardiography.

Pericardial disease

Echocardiography is the first imaging modality of choice in patients with suspected pericardial effusion and tamponade owing to its rapid image display and portability. For patients with suspected constrictive pericarditis, either MRI or CT scanning is the imaging modality that best delineates pericardial thickness. Hemodynamic analysis of the enhancement of ventricular interaction that occurs in pericardial constriction can be assessed by Doppler echocardiography.

Aortic disease

Both CT scanning and MRI are the imaging modalities of choice for the evaluation of the stable patient with suspected aortic aneurysm or aortic dissection. In the acutely ill patient with suspected aortic dissection, either TEE or CT scanning is a reliable imaging modality.

Cardiac masses

2D TTE is the first test to rule out an intracardiac mass; masses >1.0 cm in diameter are usually well visualized. Intracardiac masses of smaller size may be visualized by TEE. CT scanning and MRI are optimal for evaluating masses extrinsic to the heart or involving the myocardium.


The choice of an initial test should be based on the evaluation of the patient’s resting electrocardiogram, the ability to perform exercise, the clinical features, the patient’s body habitus, and the available local expertise and technology (Fig. 12-24). For the standard assessment of CAD, the exercise electrocardiographic test should be the initial consideration in patients with an interpretable electrocardiogram who are able to exercise. If there are resting electrocardiographic abnormalities, or if the patient has had prior coronary revascularization, an imaging modality (either nuclear imaging or echocardiography) should be used for initial evaluation. Imaging tests can add prognostic information to a standard exercise electrocardiographic test and, thus, are especially useful when the initial results fall into an intermediate risk category. Pharmacologic stress testing with imaging should be used in patients who are unable to exercise. The utility of CT coronary angiography is evolving.


FIGURE 12-24

Flow diagram showing selection of initial stress test in a patient with chest pain. Patients who are able to exercise, without previous revascularization, and with an interpretable resting ECG can be tested with an exercise ECG. The appropriate imaging study for other patients depends on multiple factors (see text). LBBB, left bundle branch block; Prev MI-Reg ischemia, previous MI with a need to detect regional ischemia; nuc, SPECT nuclear imaging study; Pharm, pharmacologic. *Consider PET if morbidly obese or female with large/dense breasts.

While the patient is often best evaluated with the imaging modality for which most experience and expertise are available, there are additional considerations and certain situations where one imaging modality has an advantage over another. Echocardiography provides structural information. Therefore, if there is a question of concomitant valve disease, pericardial disease, or aortic disease, stress echocardiography should be considered. In patients with previous infarction and/or LV systolic dysfunction on the basis of CAD, nuclear imaging, particularly PET, or MRI, is the preferred modality as it also establishes viability. In general, nuclear imaging is more sensitive and less specific than echocardiography for the detection of myocardial ischemia and viability.