Author + information
- ↵*Reprint requests and correspondence:
Dr. Elyse Foster, Professor of Clinical Medicine and Anesthesia, University of California San Francisco, 505 Parnassus Avenue, M314, San Francisco, California 94143-0214, USA.
In 1951, the feeling of frustration regarding the inadequacy of the antemortem diagnosis of cardiac masses was summarized by Pritchard (1)in a review of tumors of the heart: “Tumors of the heart are rarely diagnosed before autopsy. Surgical treatment of these neoplasms is virtually unheard of, and the present state of diagnosis is far behind the therapeutic possibilities. Several authors have wholeheartedly subscribed to the view that the diagnosis of cardiac tumors is either impossible or a matter of chance. Seen from the threshold of an era of ever bolder cardiac surgery, these tumors present a dismal diagnostic prospect.” Considerable progress in cardiac imaging modalities over the past 50 years has resulted in an improved understanding of the prevalence and natural history of various cardiac masses, and advances in surgery have enabled resection of most cardiac tumors. Echocardiography is the most widely used application in the initial investigation of cardiac masses. However, a major limitation of all current imaging modalities is their inability to accurately distinguish thrombus from benign and malignant tumors. This distinction has major therapeutic implications.
Before the advent of transthoracic echocardiography (TTE), the means of detecting cardiac masses were limited to angiography or direct inspection at surgery or postmortem. Early M-mode studies were able to detect intracardiac masses as small as 1 mm. Two-dimensional echocardiography made it possible to define the spatial extent, structural relationships, and mobility of both intracardiac and extracardiac masses (2). Although TTE is useful in the initial evaluation of cardiac masses, transesophageal echocardiography (TEE) frequently is required for a more comprehensive and accurate assessment, especially for the visualization of masses within the atria and the atrial appendages or those associated with the heart valves (3). Transesophageal echocardiography is particularly important in detecting smaller masses not seen on TTE. Features that aid in formulating a differential diagnosis of a mass include location, size, shape, “texture,” mobility, sites of attachment, and the presence or absence of myocardial infiltration. Unfortunately, the accuracy of the echocardiographic characterization is less than ideal, and the use of echocardiography provides only an “educated guess” as to the diagnosis. Moreover, both TTE and TEE are limited in evaluating masses in the mediastinum and paracardiac structures. Magnetic resonance imaging (MRI) has proved to be the gold standard for the assessment of these masses (4). However, the widespread availability, portability, and additional functional information provided by echocardiography makes it the initial investigation of choice for the assessment of cardiac masses.
Recent advances in echocardiographic imaging include contrast-enhanced imaging and power modulation (5,6). At present, the major clinical indication for contrast-enhanced echocardiography is endocardial border delineation for the evaluation of global and segmental left ventricular (LV) function. Emerging applications include the evaluation of myocardial perfusion and the improvement of the definition of intracavitary structures (7). In this issue of the Journal, Kirkpatrick et al. (8)have extended the use of contrast-enhanced imaging to determine the relative perfusion of cardiac masses and thereby aid in the differentiation of the sparse vascularity of a benign tumor, neovascularization of a malignancy or vascular tumor, and an avascular thrombus. Although the number of patients in the study was small (n = 16), the addition of contrast perfusion imaging enabled the correct diagnosis to be made by a single experienced observer in all cases and improved the diagnostic accuracy of less experienced observers compared with the pathological diagnosis.
A number of limitations of the Kirkpatrick et al. (8)study need to be considered. No clinical data were provided by the authors. The most likely diagnosis of a cardiac mass can often be made by a combination of the echocardiographic appearance and location of the mass together with clinical data. For example, 7 of the 16 cardiac masses, including 5 of the 6 masses located in the LV apex, were intracardiac thrombi, reflecting the relatively high incidence of intracardiac thrombus compared with cardiac tumors. Although in distinction to the malignant masses the thrombi failed to enhance with contrast, it is likely that the clinical setting may have substantially aided the diagnosis without the need for contrast. Left ventricular thrombi typically occur in conditions associated with stasis of blood flow and/or regional wall motion abnormalities, including myocardial infarction (especially the cardiac apex), LV aneurysm, and dilated cardiomyopathy (2,9). A history of a myocardial infarction and/or apical wall motion abnormality on echocardiography would make the diagnosis of an apical mass far more likely to be a thrombus than the much rarer tumor. In support of the use of contrast, LV thrombi have been reported in patients with normal regional LV function, and although there are other distinguishing echocardiographic characteristics of LV thrombus, these features may be variably present (9). Both left atrial masses were myxomas. A limitation of the use of contrast for differentiating thrombus from myxoma was pointed out by the authors: whereas myxomas appeared to demonstrate partial enhancement to visual inspection, thrombi and myxomas could not be differentiated objectively on the basis of average pixel intensity in the mass relative to the myocardium. As for LV masses, the clinical setting usually aids in the distinction of thrombus from myxoma in the left atrium. For example, left atrial thrombi typically occur in patients with a clinical condition associated with stasis of blood in the left atrium such as atrial fibrillation, mitral stenosis, prosthetic mitral valve, left atrial enlargement, and low cardiac output (9). A further distinctive feature is that compared with myxomas that are typically (but not always) attached to the fossa ovalis by a narrow stalk, left atrial thrombi tend to have a broad-based attachment to the posterior and lateral walls, especially within the left atrial appendage (2).
Of the 16 patients in the study by Kirkpatrick et al. (8), 3 patients had a mass in the pericardium, all of which were metastatic adenocarcinomas. The major differential diagnosis of a pericardial mass is a primary tumor versus metastatic disease (10). Metastatic tumors to the heart and pericardium occur in up to 15% of patients with malignant diseases and are 20 to 40 times more common than primary tumors (11). Carcinomas of the lungs and breast, because of their prevalence, are the most common malignant tumors that metastasize to the heart, whereas melanoma has the greatest propensity to metastasize to the heart (11). Cardiac metastases generally appear late in the course of the primary disease, and isolated cardiac involvement is rarely seen without dissemination to other organs or as the presenting symptom of a remote primary tumor (2). Nearly all primary malignant cardiac tumors are sarcomas, and angiosarcoma is the most frequent. Because angiosarcomas tend to occur in the right atrium and involve the pericardium, the most frequent clinical presentation is right-sided heart failure or tamponade (10). Operative intervention is usually unsuccessful, and the prognosis is poor (11). In the study by Kirkpatrick et al. (8), five masses were located in right-sided cardiac chambers: three were malignant and two were thrombi. Most tumors arising within the right atrium are benign (most commonly myxomas, lipomas, hemangiomas, and thrombi), whereas those extending into the right atrium from outside are malignant (most commonly hypernephroma, hepatoma, and uterine leiomyosarcoma). Right ventricular tumors are rare and when encountered are likely to be from a metastatic malignancy (12).
Magnetic resonance imaging has been a major advance in the assessment of cardiac masses. A recent article on the usefulness of MRI of cardiac and paracardiac masses concluded that there was no single feature that was both highly specific and highly sensitive for malignant tumors (13). Highly sensitive but less specific for malignant lesions were the location outside the left heart, inhomogeneity, and gadolinium enhancement. The specificity of gadolinium enhancement for malignancy increased when only moderate and strong enhancement were considered because mild enhancement was found in 8 of 19 benign tumors. Highly specific but less sensitive indicators for malignant lesions were the infiltration of adjacent compartments, tumor size >5 cm, and the presence of pericardial and/or pleural effusions. All inhomogeneous tumors located in the right heart with concurrent pericardial effusion were malignant, whereas homogeneous tumors located in the left heart without pericardial effusion were always benign.
Although advances in contrast-enhanced imaging may improve the distinction of thrombus and benign and malignant masses, they cannot provide a histological diagnosis that may present important information on prognosis and appropriate therapy. Transvenous biopsy of masses in the right heart has been performed with success and may guide further investigations and management, but it is not routinely performed for the more common left heart and pericardial masses (12). Future clinical applications for contrast-enhanced echocardiography will likely expand beyond perfusion imaging. There has been considerable progress in the past few years in the development of site-targeted microbubbles that attach to specific markers of disease, allowing noninvasive diagnostic ultrasound imaging of molecular and cellular processes (14). Similar targeting of microbubbles to tumor antigens may provide a useful method for diagnosing neoplasms with ultrasound, for detecting metastases, and for characterizing tumor phenotype (14). It has recently been shown that contrast-enhanced ultrasound with microbubbles targeted to markers on neovascular endothelium can noninvasively detect early tumor angiogenesis (15). Further investigations are needed to determine whether targeted imaging of angiogenic phenotype in different tumor types provides additional diagnostic and prognostic information to imaging methods already used in the clinical setting. The use of microbubbles and ultrasound for therapeutic purposes is also now being realized with the potential for both the accurate diagnosis and therapeutic intervention of cardiac masses (14). The specific role for contrast echocardiography in the diagnosis of cardiac and pericardial tumors will await further studies that include larger numbers of patients and comparison with MRI. Nonetheless, the study by Kirkpatrick et al. (8)provides clinicians with another tool to use in the preoperative diagnosis of cardiac masses.
↵* Editorials published in the Journal of the American College of Cardiologyreflect the views of the authors and do not necessarily represent the views of JACCor the American College of Cardiology.
- American College of Cardiology Foundation
- Prichard R.W
- Errichetti A,
- Weyman A.E
- Hoffmann U,
- Globits S,
- Frank H
- Kaul S
- ↵Kirkpatrick JN, Wong T, Bednarz JE, et al. Differential diagnosis of cardiac masses using contrast echocardiographic perfusion imaging. J Am Coll Cardiol 2004;43:1412–9.
- Ellegala D.B,
- Leong-Poi H,
- Carpenter J.E,
- et al.