Echocardiography in detection of cardiac and aortic sources of systemic embolism

INTRODUCTION — Intracardiac sources of embolism account for 15 to 20 percent of the 500,000 strokes that occur annually in the United States (in addition to other embolic issues such as organ infarction [eg, renal infarct, splenic infarct, mesenteric infarct] or acute limb ischemia) [1]. Although transthoracic echocardiography (TTE) remains the cornerstone of noninvasive cardiac imaging, transesophageal echocardiography (TEE) has been shown to be a superior method for the identification of most potential cardiac sources of emboli [2,3], although cardiovascular magnetic resonance (CMR) imaging is superior for left ventricular (LV) thrombus detection [4]. (See "Transesophageal echocardiography: Indications, complications, and normal views".)

The potential cardiac and arterial sources of emboli, along with the role of TTE and TEE in their detection, will be reviewed here. The acute management of embolic disease and the secondary prevention of additional embolization are discussed in the appropriate clinically-oriented topics. (See "Early antithrombotic treatment of acute ischemic stroke and transient ischemic attack".)

OUR APPROACH TO IDENTIFYING THE SOURCE OF EMBOLISM — The choice between TTE and TEE as the initial imaging test to identify a source of embolism should be individualized on a case-by-case basis. For most patients, TEE yields higher quality images and has a greater sensitivity and specificity than TTE, but a few conditions (eg, LV thrombus) are better seen on TTE with contrast or by CMR imaging. While TEE may have a higher yield for small intracardiac shunts/patent foramen ovale, TTE will identify large shunts, and patients undergoing TTE can generally more readily cooperate with provocative maneuvers (eg, cough and Valsalva) due to the lack of sedation. Because of widespread availability, ease of performance, and moderate to high yield in many patients, we typically begin with TTE with agitated saline contrast at rest and with maneuvers for younger patients (under age 65 years), although for selected patients, TEE as the initial test may be preferred.

TTE as the initial test — We choose TTE as the initial test for the majority of patients with a suspected cardiac or aortic source of emboli, including:

●Patients ≥45 years with a neurologic event and no identified cerebrovascular disease

●Any patient with an abrupt occlusion of a major peripheral or visceral artery

●Patients with a high suspicion of LV thrombus

●Patients in whom TEE is contraindicated (eg, esophageal stricture, unstable hemodynamic status) or who refuse TEE

TEE as the initial test — We choose TEE as the initial test to localize the source of embolism in the following circumstances:

●Patients <45 years without known cardiovascular disease (ie, absence of infarction or valvular disease history) for whom a patent foramen ovale is strongly suspected.

●Patients with a high pretest probability of a cardiac embolic source in whom a negative TTE is likely to be falsely negative and likely to lead to a TEE.

●Patients with atrial fibrillation and suspected left atrial or left atrial appendage thrombus, especially in the absence of therapeutic anticoagulation, but only if the TEE would impact management (eg, cardioversion is desired).

●Patients with a prosthetic mitral or aortic valve.

●Patients with suspected aortic pathology.

We rarely choose CMR imaging as the initial test but may obtain this study if the TTE has suboptimal image quality, if TTE with contrast is nondiagnostic, or if LV thrombus is strongly suggested, as in patients with an LV aneurysm.

POTENTIAL SOURCES OF EMBOLI — A systemic (arterial) embolus generally arises from an intracardiac or arterial location, although a venous mass/thrombus may paradoxically cause a systemic (arterial) embolus in a patient with intracardiac shunting (eg, patent foramen ovale).

Intracardiac sources of emboli — Intracardiac sources of emboli (table 1) can originate from a variety of chambers and endocardial surfaces, though, in the absence of intracardiac shunting, the mass must be left-sided. While there are a variety of ways to classify intracardiac masses, we generally divide the masses into two main categories: thrombus and nonthrombotic masses (table 1).

Thrombus — Intracardiac thrombus typically arises in the setting of low (or relatively low) flow, leading to blood stasis, or in the setting of a mechanical prosthetic heart valve and subtherapeutic anticoagulation. A thrombus may develop in the LV, left atrium (LA), or on a left-sided prosthetic valve. Additionally, systemic venous thrombi may paradoxically embolize in patients with right-to-left intracardiac shunting.

●Left ventricle – Thrombi in the LV can occur in patients with a myocardial infarction (usually anteroapical), resulting in severe hypokinesis, akinesis, or dyskinesis/aneurysm of the corresponding wall segments and reduced LV systolic function. In addition, patients with a dilated cardiomyopathy (ischemic or nonischemic) and severely reduced LV systolic function may rarely develop apical LV thrombus. Patients with transient apical dysfunction (eg, stress cardiomyopathy) are also at risk for thrombus formation. (See 'Left ventricular thrombi' below and "Left ventricular thrombus after acute myocardial infarction" and "Antithrombotic therapy in patients with heart failure", section on 'Left ventricular thrombus' and "Clinical manifestations and diagnosis of stress (takotsubo) cardiomyopathy".)

●Left atrium – The LA, and specifically the left atrial appendage (LAA), is the most common location for atrial thrombus formation. The majority of LAA thrombi occur in the setting of atrial fibrillation or atrial flutter. Significant mitral valve stenosis leading to LA dilation is also associated with LA/LAA thrombi. Patients with cardiac amyloid in whom there is markedly impaired atrial mechanical activity are at risk for LA/LAA thrombus formation despite apparent electrical sinus rhythm. (See 'LA/LAA thrombi' below and "Prevention of embolization prior to and after restoration of sinus rhythm in atrial fibrillation" and "Rheumatic mitral stenosis: Clinical manifestations and diagnosis", section on 'Thromboembolism' and "Cardiac amyloidosis: Epidemiology, clinical manifestations, and diagnosis", section on 'Cardiac involvement'.)

●Mechanical prosthetic valve – Patients with a mechanical prosthetic valve in the mitral or aortic position are at risk for thrombus formation on the valve with subsequent embolization, primarily in the setting of subtherapeutic warfarin anticoagulation. Even when a discrete thrombus is not identified by imaging, a mechanical valve should always be considered a potential source of embolus, particularly if anticoagulation has been suboptimal or interrupted. For these patients, TEE is often preferred as the initial test. (See 'Prosthetic valve sources' below and "Mechanical prosthetic valve thrombosis or obstruction: Clinical manifestations and diagnosis".)

●Paradoxical embolization – Patients with an atrial septal defect or patent foramen ovale (with or without associated atrial septal aneurysm) may have intermittent right-to-left intracardiac shunt flow across the defect. In such instances, patients may have paradoxical systemic embolization of a venous thrombus or mass, with large shunts more likely to be a source than small shunts. (See 'Abnormalities of the interatrial septum' below and "Atrial septal abnormalities (PFO, ASD, and ASA) and risk of cerebral emboli in adults".)

Nonthrombotic masses — As with thrombi, nonthrombotic masses can arise in a variety of locations within the heart, including the LV, LA, or valvular structures. While mitral annular calcification (MAC) may be a rare cause of emboli due to overlying thrombus, most nonthrombotic masses can be categorized as either tumors or vegetations.

●Tumors – A variety of benign and malignant intracardiac tumors may present with primary or secondary cardiac manifestations; however, in terms of cardiac masses that may potentially embolize, cardiac myxomas and papillary fibroelastomas are the most common candidates. The vast majority of myxomas present in the LA attached to the interatrial septum, while papillary fibroelastomas are typically found on the aortic or mitral valves. (See "Cardiac tumors".)

●Vegetations – Endocarditis most commonly presents with vegetations involving either native or prosthetic valves, although any endocardial surface may be involved (eg, complex congenital heart disease). In addition to infectious vegetations, noninfectious endocarditis (also called nonbacterial thrombotic endocarditis or NBTE) with valvular vegetations may occur in patients with an underlying malignancy, systemic inflammatory illness (eg, systemic lupus erythematosus, rheumatoid arthritis, etc), or in patients being treated for sepsis or severe burn injuries. (See "Clinical manifestations and evaluation of adults with suspected left-sided native valve endocarditis" and "Nonbacterial thrombotic endocarditis".)

Arterial sources of emboli — Complex (>4 mm radial thickness and/or mobile) aortic atheroma, typically in the ascending aorta and aortic arch, are an important potential source of both thromboemboli and atheroemboli in patients with unexplained stroke, transient ischemic attack, and arterial embolization. Embolization from aortic plaques is discussed in greater detail separately. (See "Thromboembolism from aortic plaque" and "Embolism from atherosclerotic plaque: Atheroembolism (cholesterol crystal embolism)".)

CHOOSING BETWEEN TTE AND TEE — While TTE and TEE have advantages and disadvantages, both are effective diagnostic tests with a role in the evaluation of suspected cardioaortic source of embolism. In most patients, TEE yields higher quality images and has a greater sensitivity and specificity than TTE for common sources of embolism, but a few conditions (eg, LV thrombus) are better seen on TTE (with an endocardial border definition agent) or CMR. However, because it is less invasive and readily available in most institutions, TTE, often with intravenous agitated saline at rest and with maneuvers (eg, cough and Valsalva release), is often reasonable as the initial test of choice, and in patients with adequate imaging windows, TTE can have a sensitivity and specificity approaching that of TEE [5] (see 'Our approach to identifying the source of embolism' above).

LA/LAA thrombi — Thrombus occurring in the left atrial appendage (LAA) or body of the left atrium (LA), most often associated with atrial arrhythmias (most often atrial fibrillation [AF] or atrial flutter) and/or rheumatic mitral stenosis, accounts for nearly half of cardiogenic thromboemboli. LAA thrombi are also reported in patients with cardiac amyloid due to the associated LA mechanical dysfunction despite apparent electrical sinus rhythm. The incidence of LA/LAA thrombus in AF is highly variable depending on the population studied [6-8]. TEE offers detailed visualization of the LA, including the LAA, which is the main site of thrombus formation in patients with AF or mitral stenosis (movie 1 and movie 2 and movie 3 and figure 1 and movie 4 and movie 5). In one intraoperative study, for example, TEE was compared with direct visualization of LA contents at surgery; the sensitivity and specificity of TEE for thrombi were 100 and 99 percent, respectively [9]. By comparison, conventional two-dimensional TTE has a sensitivity of only 39 to 65 percent for LA thrombi due to poor visualization of the LAA [9].

Left atrial spontaneous echo contrast — Spontaneous echo contrast seen during echocardiography is believed to represent erythrocyte aggregation in low shear rate conditions [10]. Spontaneous echo contrast is most commonly seen in the LA and LAA on TEE, but it may also be seen by TTE in the LV with the use of high frequency transducers in patients with severe LV dysfunction.

Spontaneous echo contrast or "smoke-like" echoes seen within the LA during TEE is a common finding among patients referred for evaluation of a possible cardiac source of embolism, especially in those with AF or LA enlargement, and should raise concern about the possibility of an underlying thrombus (movie 4 and movie 5 and movie 1 and movie 2 and image 1) [11,12]. Its appearance is highly variable and can be accentuated or minimized depending on the gain settings used while imaging.

There is a very strong association between depressed LA appendage ejection velocity, LA spontaneous echo contrast, and LAA thrombus. Among patients with AF and LA thrombi, for example, spontaneous echo contrast is seen in almost 80 percent of cases. This finding is an independent predictor of thromboembolic risk and, in patients with AF, is associated with an increase in embolic rate from 3 to 12 percent per year. Mitral regurgitation appears to lessen the frequency of spontaneous echo contrast. (See "Atrial fibrillation in adults: Use of oral anticoagulants".)

Left ventricular thrombi — LV thrombus is one of the complications of myocardial infarction, usually involving the anteroapical region. TTE can document LV wall motion abnormalities that may underlie thrombus, the presence of thrombus, and risk factors for embolization. The use of TTE with an endocardial border definition contrast agent (eg, Definity, Lumason, Optison) can also be helpful in identifying thrombus in patients with difficult imaging windows (movie 6). One limitation of TEE is difficulty in imaging the LV apex. As a result, TTE is generally preferred for the assessment of LV systolic function and identification of LV apical thrombi (movie 7) [13]. For patients with suboptimal TTE even with an endocardial contrast agent, cardiovascular magnetic resonance (CMR) with gadolinium contrast should be considered and is superior to native TTE, TTE with contrast, and TEE for detection of intraventricular thrombus [4].

Issues related to the diagnosis, prevention, and treatment of LV thrombus are discussed in detail separately. (See "Left ventricular thrombus after acute myocardial infarction".)

Valvular vegetations — Native heart valves, particularly those with underlying structural abnormalities, may serve as a nidus for infectious and noninfectious vegetations. When an infectious vegetation is suspected, the risk of subsequent embolization is related to vegetation size and mobility, although echocardiography is relatively poor at predicting risk of embolization [14,15]. Nonbacterial thrombotic endocarditis (NBTE), or marantic endocarditis, refers to a spectrum of lesions ranging from microscopic aggregates of platelets to large vegetations on the heart valves (most often aortic and mitral), usually in patients with advanced malignancy. (See "Nonbacterial thrombotic endocarditis".)

We begin with TTE in patients with suspected endocarditis and native heart valves because it is less invasive than TEE. However, TTE has relatively low sensitivity in infective endocarditis. Thus, a negative study does not preclude the diagnosis and should be followed by TEE, which has superior spatial resolution and improved image quality (movie 8 and movie 9 and movie 10) in moderate or high risk patients.

As an example, one prospective study of 96 patients with suspected endocarditis reported a sensitivity for the detection of vegetations of 100 percent with TEE compared with only 63 percent with TTE [16]. In a second report, TTE and TEE were compared in 66 episodes of suspected endocarditis; TEE had a sensitivity of 94 percent versus only 44 percent with TTE. Both methods had specificity (the likelihood of a negative result in a patient without the disease) approaching 100 percent, indicating very few false positives [17].

Mobile, filamentous echo densities (sometimes referred to as Lambl excrescences) (image 2) have been described on native aortic and mitral leaflets among patients (especially older adults) who undergo TEE for indications other than "source of embolism" or "endocarditis" [18-20]. These echo filamentous densities are felt to represent fibrin strands. An increased prevalence of these thread-like echo densities has been described in patients with a recent cerebral ischemic event [18-20]. Similar TEE findings have been described on apparently normal functioning prosthetic valves in the absence of an embolic history.

Prosthetic valve sources — Thrombus formation and subsequent embolization are common in patients with mechanical valve prostheses, especially in those with a mitral or tricuspid prosthesis and suboptimal anticoagulation (movie 11) [21]. Patients with prosthetic valves are also at risk for infectious endocarditis with vegetations as well as pannus formation on the prosthesis. (See "Mechanical prosthetic valve thrombosis or obstruction: Clinical manifestations and diagnosis" and "Antithrombotic therapy for mechanical heart valves".)

Evaluation of prosthetic valves, especially mitral and tricuspid prostheses, is best performed by TEE because of the orientation of the valve relative to the imaging probe, which eliminates interference from reverberations off of the metallic valve. Aortic and mitral prostheses often require a combination of both TTE and TEE to comprehensively assess the valve. In patients with a prosthetic heart valve, unless another source is identified, the valve must be considered the presumed embolic source even if imaging does not demonstrate residual thrombus. The risk of thrombus formation on prosthetic valves is high, particularly if there have been periods of suboptimal anticoagulation. The risk is higher for mechanical valves, but thromboembolism can occur with bioprosthetic valves as well. In addition, residual thrombi may no longer be present after embolization, may be small, or may be obscured by artifacts due to the prosthetic valve material.

Cardiac tumors — Cardiac myxomas and papillary fibroelastomas are the cardiac tumors most associated with thromboembolism.

●Myxomas – Myxomas, the majority of which present in the LA or right atrium and often in association with the interatrial septum. Over half present with thromboembolism, felt to be the result of embolization of tumor material itself or of overlying thrombi. Both TTE and TEE are highly sensitive in detecting myxomas (movie 12 and movie 13 and movie 14 and movie 15 and movie 16). TEE may provide more accurate anatomical details, such as the site of attachment and thereby guide resection technique, and, in selected patients, may help to differentiate these tumors from thrombi [22]. (See "Cardiac tumors", section on 'Myxomas'.)

●Papillary fibroelastomas – Papillary fibroelastomas are tumors that are usually pedunculated and most frequently located on the aortic or mitral valves, although they may also be present on endocardial surfaces [23]. Echocardiographically, papillary fibroelastomas appear speckled with echolucencies near the edges [24]. These tumors may be a source of systemic embolization due to migration of thrombus from the tumor surface or tumor embolization and are best detected by TEE [23,24]. (See "Cardiac tumors", section on 'Papillary fibroelastomas'.)

Abnormalities of the interatrial septum — Abnormalities of the interatrial septum are associated with thromboembolism via two pathophysiologic mechanisms: transient right-to-left shunting and interatrial septal aneurysm (movie 17 and movie 18). Both TTE and TEE, combined with intravenous injection of agitated saline contrast, are highly effective for documenting an interatrial shunt, particularly when performed with provocative maneuvers (eg, cough and Valsalva). TEE is generally better for detecting the precise location and measuring the physical size of any defect and smaller defects. Atrial septal aneurysms are generally better visualized on TEE.

Patients with an atrial septal defect or patent foramen ovale may have intermittent flow of blood from the right atrium to the left atrium (movie 19 and movie 20 and movie 21 and movie 22). This can be induced on a daily basis by coughing or during the release phase of the Valsalva maneuver (as with bowel movements); either maneuver will transiently increase right atrial pressure. A thrombus in the venous system may therefore cross to the left side of the heart, resulting in a systemic thromboembolism. Such a "potential" communication has been found in over 25 percent of normal hearts at autopsy [25]. (See "Patent foramen ovale" and "Atrial septal abnormalities (PFO, ASD, and ASA) and risk of cerebral emboli in adults".)

In a retrospective review of 1162 consecutive patients who underwent TTE with agitated saline contrast injection, paradoxical right-to-left shunting was seen in 403 patients (35 percent), although 195 patients (48 percent of those with shunting) showed paradoxical shunting only following a provocative maneuver (ie, Valsalva maneuver) [26]. There was very good agreement between TTE and TEE, with 99 percent sensitivity and 85 percent specificity of TTE for making the diagnosis. Saline contrast TEE may have reduced sensitivity in more heavily sedated patients due to the patient's inability to cooperate with maneuvers in the sedated state.

An atrial septal aneurysm (movie 17 and movie 18 and movie 23) is a congenital malformation due to redundant atrial septal tissue, typically involving the region of the fossa ovalis [27]. Prevalence in the normal population is estimated to be 0.5 percent by TTE and up to 5 percent by TEE [28-30]. Embolic events can result from an associated patent foramen ovale (which may be present in over 75 percent of cases), or from direct thrombus formation in the neck of the aneurysm [31].

Aortic atherosclerosis — For imaging of the aorta in most patients, TEE is a far better diagnostic test than TTE. While the aortic root at the level of the sinuses of Valsalva, a small portion of the aortic arch, and part of the descending thoracic aorta may be seen during standard TTE imaging, large portions of the aorta are not visible during TTE due to anatomic proximity of the probe or other intervening structures. Conversely, with TEE imaging, the thoracic aorta can generally be imaged continuously from its origin to the diaphragm (with the exception of a small portion posterior to the trachea in most patients).

TEE is a sensitive method for visualization of the aortic intima (movie 24). Aortic atherosclerotic plaques, which can be seen on TEE, are an important source of both thromboemboli and atheroemboli in patients with unexplained stroke, transient ischemic attack, and arterial embolization. Complex (≥4 mm radial thickness and/or mobile) atheromas have greater thrombogenic potential.

SUMMARY AND RECOMMENDATIONS

●Intracardiac sources of embolism account for 15 to 20 percent of strokes (in addition to other embolic issues such as organ infarct [eg, renal infarct, splenic infarct, mesenteric infarct] or acute limb ischemia). Echocardiography plays an important role in the diagnosis of intracardiac sources of embolism. (See 'Introduction' above.)

●The choice between transthoracic echocardiography (TTE) and transesophageal echocardiography (TEE) as the initial imaging test to identify a source of embolism needs to be individualized on a case-by-case basis. (See 'Our approach to identifying the source of embolism' above.)

•We choose TTE, often with intravenous injection of agitated saline at rest and with maneuvers as the initial test for the majority of patients <65 years of age and without agitated saline for older patients with a suspected cardiac or aortic source of emboli, including:

-Patients ≥45 years with a neurologic event and no identified cerebrovascular disease

-Any patient with an abrupt occlusion of a major peripheral or visceral artery

-Patients with a high suspicion of left ventricular (LV) apical thrombus

-Patients in whom TEE is contraindicated (eg, esophageal stricture, unstable hemodynamic status) or who refuse TEE

•We choose TEE as the initial test to localize the source of embolism in the following circumstances:

-Patients <45 years without known cardiovascular disease (ie, absence of infarction or valvular disease history) for whom a patent foramen ovale is strongly suspected.

-Patients with a high pretest probability of a cardiac embolic source in whom a negative TTE would be likely to be falsely negative.

-Patients with atrial fibrillation and suspected left atrial or left atrial appendage thrombus, especially in the absence of therapeutic anticoagulation, and presuming information from the TEE would impact management (eg, cardioversion is desired).

-Patients with any prosthetic heart valve.

-Patients with suspected aortic pathology.

●Intracardiac sources of emboli can originate from a variety of chambers and endocardial surfaces, though in the absence of intracardiac shunting the mass must be left-sided. While there are a variety of ways to classify intracardiac masses, we generally divide the masses into two main categories: thrombus and nonthrombotic masses (table 1). (See 'Intracardiac sources of emboli' above.)

●While TTE and TEE have advantages and disadvantages, both are effective diagnostic tests with a role in the evaluation of suspected cardioaortic source of embolism. In most patients, TEE yields higher quality images and has a greater sensitivity and specificity than TTE, but a few conditions (eg, LV thrombus) are better seen on TTE (with an endocardial definition agent) or cardiovascular magnetic resonance (CMR) imaging. However, because it is less invasive and readily available in most institutions, TTE with an endocardial definition agent is often reasonable as the initial test of choice, and in patients with adequate imaging windows, TTE can have a sensitivity and specificity approaching that of TEE. For patients with LV systolic dysfunction with suboptimal TTE with subendocardial contrast agent, a CMR imaging study with gadolinium contrast should be considered. (See 'Choosing between TTE and TEE' above.)