Home SVCC                                                  Area: English - Español - Português

TEE and Source of Emboli

Stefano De Castro, MD; Federica Papetti, MD;
Anna Toso, MD; Alessandro Aiello, MD;
Adnan Hachem, MD; Annalisa Celotto, MD;
Daniela Padovani, MD; Francesco Fedele, MD

Cardiovascular Imaging Laboratory, Dpt of Cardiovascular and Respiratory
Sciences 'La Sapienza', University Policlinico Umberto I, Rome, Italy

   Stroke continues to represent a major cause of mortality and morbidity in the industrialized countries and it recognizes multiple etiologies, including atherosclerotic cerebral vascular disease, and cardiac and hemathologic disorders. Although the vast majority of cerebral infarctions are secondary to cerebrovascular atherosclerotic disease, up to 20% of all strokes are thought to be secondary to cerebral embolism of cardiac origin.

   The advent of TEE has offered a method for examining regions of the heart often implicated in cerebral embolic events of known or suspected cardiac origin. These include superior resolution of the left and right atrium and appendages, interatrial septum, aorta, patent foramen ovale, atrial septal aneurysm, vegetations, and spontaneous contrast, thus providing increased sensitivity for detection of various cardiac abnormalities associated with embolism of cardiac origin. Thrombi and vegetative lesions associated with prosthetic or native cardiac valves are also visualized to a much greater extent with TEE.

   The cause of ischemic stroke (IS) and transient ischemic attacks (TIA) often remain an unsolved problem (the clinical evaluation does not always discriminate between embolic and atherosclerotic causes). Transthoracic echocardiography has poor diagnostic sensitivity in finding the possible cardiac causes of emboli in patients with IS, and, in particular, those without underlying cardiac abnormalities. TEE is a relatively new diagnostic method which produces real time cardiac images with extremely high resolution. Recent studies have demonstrated its potential in recognizing possible cardioembolic sources in patients with IS or TIA (1,2).

   TEE identified potential cardiac sources of emboli in approximately 60% of the overall population compared with only 15% by transthoracic echocardiography. Even in patients without clinical heart disease, TEE identified abnormalities in around 35% of patients with cerebral ischemic events. TEE appears superior to TTE in detecting atrial septal aneurysm, PFO, left atrial thrombus, left atrial spontaneous contrast, and left appendage abnormalities.

   Identifiable cardiac sources of emboli may be divided into abnormalities with definite evidence of cause and effect, and lesions that have been found with increased prevalence in patients with cryptogenic cerebral ischemia.

   We briefly report the most common indications to perform echocardiographic examinations in a daily clinical practice.

   Echocardiography permits dynamic assessment of intracardiac masses evaluating both anatomic extent and pathophysiologic sequelae.

   First of all, the user has to know that some artifacts can be caused by electrical interference or various physical factors influencing image formation from the reflected ultrasound signals. That may cause artifacts so making interpretation of intracardiac abnormal signals very difficult. Below are reported the most common structures that may be mistaken for abnormal cardiac masses.

Left atrial thrombus
   Several studies have demonstrated the superiority of TEE in identification of left atrial thrombus. The obvious advantage of this technique is the proximity of the left atrium to the transducer location in the esophagus. In standard transthoracic echocardiography, the left atrium is in the far field, and attenuation of the ultrasound signal hampers identification of such masses. In addition TEE allows visualization of regions of the left atrium not seen by standard TTE. This includes, importantly, the left atrial appendage, in which a large number of atrial thrombi can be found. Although it is well known that patients with atrial fibrillation and left atrial enlargement have a higher incidence of left atrial thrombi, several investigators have recently shown that, in some cases, left atrial thrombus may be identified in individuals in sinus rhythm with no detectable morphologic or functional cardiac disease. Furthermore, decisions regarding anticoagulation treatment, timing for cardioversion, and assessment for percutaneous balloon mitral valvuloplasty, are now more comprehensibly assessed with TEE.

Left ventricular thrombus
   Two-dimensional echocardiography remains the procedure of choice for identification of LV thrombus. The sensitivity and specificity of this technique is in the 80 to 90% range. There are, however, occasional patients in whom technically adequate transthoracic studies may not be available.

   Myxoma is by far the most common tumor of cardiac origin. This lesion is most commonly encountered in the left atrium, although right atrial myxoma is found in a proportion of these patients. Identification of the attachment of such lesions, as well as differentiation of myxomas from atrial thrombus, can best be accomplished by application of this technique. In addition, small recurrent lesions as well as multiple myxomas can be identified with higher accuracy by superior imaging resolutions afforded by TEE. Other tumors that have been associated with cardiac source of embolus include also papillomas, fibromas and others.

    Non primary cardiac tumors are about 20 times more common than the primary ones. Cardiac malignancies most often involve the pericardium and epicardium. Myocardial involvement by metastatic disease is less common, but can occur with lymphoma or melanoma. Endocardial involvement, conversely, is rarely seen.

   Vegetations secondary to infective endocarditis are identified with higher degree of accuracy by TEE. The diagnostic yield of TTE is approximately 70% while it is 95% by application of TEE. In addition, TEE can visualize complications of infective endocarditis and assess, with higher degree of accuracy, vegetation size and mobility. The role of echocardiography, both TTE and TEE, in predicting embolic events is still controversed.

Protruding Aortic Debris
   It has been reported that extensive atherosclerotic plaque in the ascending aorta or aortic arch may be responsible for both cerebral and systemic embolization. Protruding and mobile debris appear to have the highest embolic potential. Lesions of > 4 mm thickness or with mobile components superimposed have been seen in multiple studies as related to otherwise unexplained thoracic stroke. TEE, as high definition imaging technique, is the method of choice for studying aorta abnormalities. Toyoda et al. (3) found TEE to have a sensitivity and specificity of 74% and 94% respectively, in diagnosing such "complicated" aortic lesions. Multiplane TEE gives more detailed informations compared to CT-scan and aortography, probably because it can observe the inner surface of the plaque closely and continuously in a multitude of cutting planes (4).

   PFO and ASD are known to occur in a significantly higher percentage of patients with cerebral ischemic events than in those without cerebral ischemia. It has been hypothesized that paradoxical emboli are responsible for the events in such patients. It is clear that TEE is a superior method for the evaluation of intracardiac shunt at the atrial level; with the injection of 10 ml of agitated saline solution into an arm vein, the sensitivity and specificity of this technique approach to 100%. In the presence of PFO, microbubbles may appear in the left-side cavities, depending on the pressure gradient between right and left atrium. Normally, the mean left atrial pressure is higher than right. However, a transient inversion of this pressure gradient often occurs during early systole. This explains the spontaneous appearance of bubbles in the left heart cavities in the presence of PFO. The normal interatrial pressure gradient can also be transiently reversed by provocative maneuvers such as coughing or the Valsalva maneuver. With cardiac catheterization used as reference method, contrast echocardiography performed during the Valsalva maneuver has been reported to detect the patency of patent foramen ovale with a sensitivity of 60% and a specificity of 78%. With a cough test, sensitivity and specifity of the method were 100% and 78%, respectively .

   We performed a study (5) that showed a higher risk for recurrent brain embolism in PFO patients, when high membrane mobility is associated with shunting at rest. We studied 350 patients, within one week at the admission, with acute ischemic Stroke/TIA. Of these, 74 (21%) were identified by contrast TEE to have a PFO; 86 (25%) were cryptogenic stroke patients and 163 patients were excluded because of the presence of a definite or possible arterial or clinical evidence of source of emboli or small vessel disease. Thirteen PFO subjects, without history of embolism, were considered as control group. All PFO and criptogenic stroke patients were followed by neurological visits. We found that, compared to controls, PFO patients with cerebrovascular ischemic event, more frequently presented a right-to-left shunt at rest and a higher membrane mobility (P<.05). Patients with these characteristics were considered at higher risk. During a median follow-up period of 31 months, 8 PFO and 18 cryptogenic patients experienced recurrent cerebrovascular event. The cumulative estimate of risk of cerebrovascular event recurrence at 3 years was 4.3% (95% CI, 0 to 10.2) for 'low risk' PFO patients, 12.5% (95% CI, 0 to 26.1) for 'high risk' PFO patients, and 16.3% (95% CI, 7.2 to 25.4) for cryptogenic stroke patients (high risk PFO vs low risk PFO, P=.05).

   The prevalence of a PFO as assessed by transesophageal contrast echocardiography varies considerably among echocardiographic laboratories (ranging from 8 to 54%). This striking difference may be due to several reasons: (1) the population studied (age, healthy persons versus patients, history of embolism); (2) the technique used (contrast agent, provocative maneuvers, ultrasonic instrumentation including transducer frequency); and (3) the criteria applied for diagnosis. A large autopsy series has shown that foramen ovale is patent in 20 to 34% of otherwise normal human hearts.

   Stollberger et al. (6) demonstrated a high interobserver variability among different laboratories, which in part may account for the large differences observed. The rate of positive studies is strongly influenced by the criteria used for PFO diagnosis. Interlaboratory disagreement concerning transesophageal contrast echocardiography exists and should be taken into account when assessing the prevalence of PFO by this method. There is a need for standardization of which technique and criteria should be applied for diagnosis of PFO, correlating the final results to the anatomic findings in a larger number of patients.

   Multiple factors, including atrial septal aneurysm and clinical risk factors associated with early and marked shunting have all been shown to risk stratify those events with patent foramen ovale and potential paradoxical events.

   New therapeutic trends include the PFO closure by atrial septal device, with the benefit to reduce thromboembolic risk; many studies are addressed to evaluate long-term results.

   Spontaneous echo contrast is a demonstration of smokelike swirling reflectances of intracardiac blood flow, most commonly seen in low flow states. The most common clinical setting in which spontaneous echo contrast is found is characterized by the presence of enlarged left atrium, tipically in patients with atrial fibrillation, with or not mitral dysfunction. Spontaneous echocontrast is also seen in the left ventricle in patients with severe global left ventricular dysfunction as well as on the right side of the heart. Its frequent association with left atrial thrombus suggests that this finding is a marker for blood stasis and subsequent thrombus formation. Studies have demonstrated that this finding in patients with non valvular atrial fibrillation identifies a group of patients at high risk for subsequent embolic events. Anyway, left atrial spontaneous echo contrast is: 1) a relative common finding in patients undergoing TEE, 2) associated with conditions favoring stasis of left atrial blood flow, and 3) a probable marker of previous thromboembolism in patients with nonvalvular atrial fibrillation and those with mitral stenosis or mitral valve replacement.

   It is defined as: "a redundant interatrial septum usually located in the region of the fossa ovalis or, sometimes, involving the entire interatrial septum. It may bulge 15 mm or more beyond the plane of the atrial septum, at some time during the cardiac cycle, into either the right or the left atrium. This abnormality is reportedly found in only a small percentage of all TTE. Using TEE, atrial septal aneurysms are identified in approximatively 4% of all patients studied. In patients with unexplained cerebral ischemia, however, it may be identified in up to 15% of this population. It has been postulated that perhaps these abnormalities provide a nidus for thrombus formation and blood stasis and therefore constitute a risk factor for cerebral embolization. In patients with unexplained cerebral ischemia, atrial septal aneurysms are associated with atrial septal defects of patent foramen ovale in up to 75% of the cases.

   Multiple factors, including atrial septal aneurysm and clinical risk factors associated with early and marked shunting have all been shown to risk stratify those events with patent foramen ovale and potential paradoxical events.



1. Palazzuoli A, Ricci D, Lenzi C, Lenzi J, Palazzuoli V. Transesophageal echocardiography for identifying potential cardiac sources of embolism in patients with stroke. Neurol Sci 2000; 21(4):195-202

2. Okura H, Inoue H, Tomon M, Nishiyama S, Yoshikaua T, Yoshida K. Transesophageal echocardiographic detection of cardiac sources of embolism in elderly patients with ischemic stroke. Intern Med 1999; 38(19): 766-772

3. Toyoda K, Yasaka M, Nagata S, Yamaguchi T. Aortigenic embolic stroke: a transesophageal echocardiographic approach. Stroke 1992; 23(8): 1056-61

4. Mariano MC, Gutierrez CJ, Alexander J, Roth F, Katz S, Kohl RD. The utility of transesophageal echocardiography in determining the source of arterial embolization. Am Surg 2000; 66(9): 901-4

5. De Castro S, Cartoni D, Fiorelli M, Rasura M, Anzini A, Zanette EM, Beccia M, Colonnese C, Fedele F, Fieschi C, Pandian NG. Morphological and functional characteristics of patent foramen ovale and their embolic implications. Stroke 2000; 31(10): 2407-13

6. Stollberger C, Schneider B, Abzieher F, Wollner T, Meinertz T, Slany J. Diagnosis of patent foramen ovale by transesophageal contrast echocardiography. Am J Cardiol 1993; 71(7): 604-6



Your questions, contributions and commentaries will be answered
by the lecturer or experts on the subject in the Echocardiography list.
Please fill in the form (in Spanish, Portuguese or English) and press the "Send" button.

or commentary
Name and Surname:
E-Mail address:


2nd Virtual Congress of Cardiology

Dr. Florencio Garófalo
Steering Committee
Dr. Raúl Bretal
Scientific Committee
Dr. Armando Pacher
Technical Committee - CETIFAC

Copyright© 1999-2001 Argentine Federation of Cardiology
All rights reserved


This company contributed to the Congress: