Transesophageal Echocardiography and Cardiac Masses

Abstract

A former president of the American College of Physicians once opined that the most compassionate thing physicians can do for a patient is to make the right diagnosis. Seven years in medical practice have convinced me of the wisdom of his words and have suggested but one addendum: the most compassionate thing physicians can do for a patient is to make the right diagnosis compassionately. Pursuing a correct diagnosis compassionately involves answering, on our patients' behalf, many questions beyond simply, “What's wrong, doctor?” For example, what combinations of tests will help determine the differential diagnosis? In what order should they be done? If several approaches are possible, which produces the least discomfort and is least invasive? Which minimizes risk? Which is most expeditious? Which is most cost-effective? Balancing this montage is the daily fare of the physician with adjustments ever necessary to accommodate new technologies. Experience acquired during the past decade has given most of us a reasonable understanding of the power and also the limitations of transthoracic two-dimensional echocardiography (TTE) in initial diagnosis and subsequent assessment of patients with cardiac masses. In this issue of Mayo Clinic Proceedings (pages 1101 to 1109), Reeder and associates attempt to determine the role of transesophageal ultrasound studies. The subject of cardiac masses encompasses a wide variety of lesions. The authors chose to exclude valvular vegetations, leaving for their study a potpourri of primary cardiac and extracardiac neoplasms and cysts, thrombi, and metastatic deposits. Intracavitary thrombi are common and clinically important as a treatable cause of substantial morbidity and mortality, especially in selected groups of patients. In contrast, primary cardiac tumors and cysts are fascinating oddities found in only 0.001 to 0.28% of unselected autopsies and vary in nature from left atrial myxomas (the most common) to extraordinarily rare primary cardiac malignant lesions.1McAllister Jr, HA Primary tumors and cysts of the heart and pericardium.Curr Probl Cardiol. May 1979; 4: 1-51Abstract Full Text PDF PubMed Scopus (210) Google Scholar Primary cardiac tumors, like thrombi, may cause considerable morbidity or death when diagnosis is delayed and are often amenable to curative surgical resection, facts that secure for them a prominent place in discussions of cardiac masses despite their rarity. Although lesions metastatic to the heart are much more common than primary cardiac neoplasms in autopsy series, they receive less attention in the cardiovascular literature both because many produce no symptoms and because the poor prognoses of the affected patients are seldom improved by cardiovascular interventions other than pericardiocentesis for tamponade. In his extensive review of cardiac tumors published in 1951, Prichard2Prichard RW Tumors of the heart: review of the subject and report of one hundred and fifty cases.Arch Pathol. 1951; 51: 98-128Google Scholar pointed out that no one had diagnosed an atrial myxoma ante mortem. Two developments in which Mayo scientists and physicians played major roles brought the entire spectrum of cardiac masses into the province of the clinician. The first was the discovery during the 1940s of the clinical role of orally administered anticoagulants,3Butt HR Allen EV Bollman JL A preparation from spoiled sweet clover [3,3′-methylene-bis-(4-hydroxycoumarin)] which prolongs coagulation and prothrombin time of the blood: preliminary report of experimental and clinical studies.Proc Staff Meet Mayo Clin. 1941; 16: 388-395Google Scholar and the second was the development of the extracorporeal pump-oxygenator during the 1950s.4Jones RE Donald DE Swan HJC Harshbarger HG Kirklin JW Wood EH Apparatus of the Gibbon type for mechanical bypass of the heart and lungs: preliminary report.Proc Staff Meet Mayo Clin. 1955; 30: 105-113PubMed Google Scholar Thereafter, thrombosis could be inhibited in appropriate patients, and even intracavitary and intramural tumors could be approached by the surgeon. Using cardiopulmonary bypass, Crafoord5Crafoord C Discussion [the technique of mitral commis-surotomy].in: Lam CR Henry Ford Hospital International Symposium on Cardiovascular Surgery: Studies in Physiology, Diagnosis and Techniques. WB Saunders Company, Philadelphia1955: 202-203Google Scholar performed the first successful removal of a left atrial myxoma in 1954. While methods of treatment were evolving, diagnosis and careful study of cardiac masses lagged for want of diagnostic tools. Thrombi became evident through an embolic event and could be studied systematically only at autopsy. Symptomatic patients with cardiac tumors challenged the clinician with typically nonspecific auscultatory, roentgenographic, and laboratory abnormalities. Even the well-studied left atrial myxoma usually eluded diagnosis, masquerading as rheumatic mitral stenosis, a collagen-vascular disease, or some other more common malady until impaled by an unsuspecting surgeon or disclosed at autopsy. Goldberg and colleagues6Goldberg HP G'enn F Dotter CT Steinberg I Myxoma of the left atrium: diagnosis made during life with operative and postmortem findings.Circulation. 1952; 6: 762-767Crossref PubMed Scopus (115) Google Scholar reported the angiographic characteristics of left atrial myxoma in 1952, and descriptions of other lesions followed. Besides diagnostic limitations, however, the expense, risk, and invasive nature of angiography made it unattractive as a method of screening for cardiac masses, which regularly are included in the differential diagnoses of myriad patients. M-mode echocardiography was introduced in Sweden in 1954, and within 5 years, its potential for noninvasive diagnosis of intracardiac thrombi and tumors was recognized and reported. During the next 15 years, descriptions of the M-mode findings for various cardiac masses were published. Included in these reports were, from the Mayo Clinic, the first English-language descriptions of left and right atrial myxomas, which were published in 19687Schattenberg TT Echocardiographic diagnosis of left atrial myxoma.Mayo Clin Proc. 1968; 43: 620-627PubMed Google Scholar and 1973,8Harbold Jr, NB Gau GT Echocardiographic diagnosis of right atrial myxoma.Mayo Clin Proc. 1973; 48: 284-286PubMed Google Scholar respectively. The limitations of the M-mode examination also became apparent. Although the M-mode images of a typical left atrial myxoma became familiar, these “ice-pick” views of the heart precluded adequate examination of some regions and allowed only incomplete characterization of masses relative to shape, dimensions, consistency, deformability, mobility, and point of attachment—often critical clues to an accurate diagnosis. Nonetheless, although the diagnostic limitations might approach those of angiography, safety, simplicity, and reasonable cost made M-mode echocardiography an acceptable screening test in the clinician's armamentarium. Most patients with myxomas were still referred for angiography preoperatively. By the late 1970s, two-dimensional echocardiography was becoming widely available. Its clear superiority in revealing those anatomic characteristics inapparent with M-mode studies made possible the accurate noninvasive diagnosis of most lesions. In a review of the Mayo Clinic experience with all cardiac tumors from the time of introduction of TTE in 1977 through 1983, 30 of 32 intracavitary and intramural neoplasms subjected to echocardiographic examination were detected.9Fyke III, FE Seward JB Edwards WD Miller Jr, FA Reeder GS Schattenberg TT Shub C Callahan JA Tajik AJ Primary cardiac tumors: experience with 30 consecutive patients since the introduction of two-dimensional echocardiography.J Am Coll Cardiol. 1985; 5: 1465-1473Abstract Full Text PDF PubMed Scopus (128) Google Scholar The frequency of diagnosis of cardiac tumors increased profoundly from previous eras. Because excellent correlation was found between the echocardiographic characteristics and those confirmed at operation, diagnostic angiography was deemed unnecessary for most patients. Two-dimensional echocardiography also proved reasonably well suited for the systematic study of left ventricular thrombi.10Reeder GS Tajik AJ Seward JB Left ventricular mural thrombus: two-dimensional echocardiographic diagnosis.Mayo Clin Proc. 1981; 56: 82-86PubMed Google Scholar Despite the greater cost of the equipment and the study itself, the superior sensitivity and specificity of two-dimensional over M-mode echocardiography led quickly to the elimination of the latter as a means of screening patients for possible cardiac masses. This new technique, however, was not without limitations either. Interference from the chest wall, lungs, ribs, prosthetic valves, calcified valves and valve annuli, and other structures so degraded the quality of the images in some patients that excluding a cardiac mass was impossible. Especially in adults, penetration to distant cardiac structures such as the high left atrium and the left atrial appendage necessitated the use of relatively low-frequency transducers (2.25 and 3.5 MHz), whose longer wavelengths provided inferior resolution. The effect of these limitations is illustrated in the previously cited experience,9Fyke III, FE Seward JB Edwards WD Miller Jr, FA Reeder GS Schattenberg TT Shub C Callahan JA Tajik AJ Primary cardiac tumors: experience with 30 consecutive patients since the introduction of two-dimensional echocardiography.J Am Coll Cardiol. 1985; 5: 1465-1473Abstract Full Text PDF PubMed Scopus (128) Google Scholar in which a 0.5-cm papillary fibroelastoma identified at autopsy on the tricuspid valve of a 65-year-old patient was not evident even on retrospective review of the previously obtained video images. To circumvent the shortcomings of transthoracic imaging, physicians and scientists from the United States, Europe, and Japan turned to the esophageal window to the heart.11Seward JB Khandheria BK Oh JK Abel MD Hughes Jr, RW Edwards WD Nichols BA Freeman WK Tajik AJ Transesophageal echocardiography: technique, anatomic correlations, implementation, and clinical applications.Mayo Clin Proc. 1988; 63: 649-680Abstract Full Text Full Text PDF PubMed Scopus (736) Google Scholar M-mode studies were reported in 1976, but this early device provided limited views and few clinical applications. A crude two-dimensional probe was developed in 1977 by Japanese investigators. The practical value of two-dimensional transesophageal echocardiography (TEE) in the surgical and outpatient settings was demonstrated in the early 1980s primarily through the work of Dr. Peter Hanrath and associates in Germany. The equipment and techniques for performing both monoplanar11Seward JB Khandheria BK Oh JK Abel MD Hughes Jr, RW Edwards WD Nichols BA Freeman WK Tajik AJ Transesophageal echocardiography: technique, anatomic correlations, implementation, and clinical applications.Mayo Clin Proc. 1988; 63: 649-680Abstract Full Text Full Text PDF PubMed Scopus (736) Google Scholar and biplanar12Seward JB Khandheria BK Edwards WD Oh JK Freeman WK Tajik AJ Biplanar transesophageal echocardiography: anatomic correlations, image orientation, and clinical applications.Mayo Clin Proc. 1990; 65: 1193-1213Abstract Full Text Full Text PDF PubMed Scopus (302) Google Scholar TEE as used at the Mayo Clinic and elsewhere have been described in detail. The shorter wavelength of the commercially available 5-MHz transducer and the absence of interference from the chest wall and lungs allow spectacular, high-resolution images of regions heretofore poorly seen, such as the atrial appendages, high left atrium, and tissues adjacent to the heart. In this retrospective review, the authors study the sensitivity of TTE and provide a semiquantitative analysis of the clinical value of the TEE findings in a group of 80 patients with 83 cardiac masses evident on TEE. The distribution of tumors, thrombi, and cysts is representative of that in other reviews. Surgical or autopsy materials were available for some cases; in the others, the diagnosis of cardiac mass was based on the TEE findings alone or in combination with those from other imaging techniques (TTE, computed tomography, or magnetic resonance imaging). Although both monoplanar and biplanar probes were in use at the Mayo Clinic during the last year of this study,12Seward JB Khandheria BK Edwards WD Oh JK Freeman WK Tajik AJ Biplanar transesophageal echocardiography: anatomic correlations, image orientation, and clinical applications.Mayo Clin Proc. 1990; 65: 1193-1213Abstract Full Text Full Text PDF PubMed Scopus (302) Google Scholar the authors do not state whether both were used in some of these patients. Overall, the findings confirm what one might anticipate from the additional capabilities of TEE and corroborate those from a growing number of other reports in the literature. Transesophageal imaging revealed several masses in the left atrial appendage and even in the body of each atrium and also smaller masses, such as the small left and right atrial tumors, that were undetectable by TTE. For atrial masses in general, transthoracic imaging performed poorly for left atrial thrombi (detecting only 27%) and performed reasonably well only for relatively large left atrial myxomas. Similar findings were reported by Mügge and co-workers13Mügge A Daniel WG Haverich A Lichtlen PR Diagnosis of noninfective cardiac mass lesions by two-dimensional echocardiography: comparison of the transthoracic and transesophageal approaches.Circulation. 1991; 83: 70-78Crossref PubMed Scopus (248) Google Scholar and Pearson and associates.14Pearson AC Labovitz AJ Tatineni S Gomez CR Superiority of transesophageal echocardiography in detecting cardiac source of embolism in patients with cerebral ischemia of uncertain etiology.J Am Coll Cardiol. 1991; 17: 66-72Abstract Full Text PDF PubMed Scopus (501) Google Scholar Studies of infective endocarditis have emphasized the inadequacy of TTE for detecting small masses.15Erbel R Rohmann S Drexler M Mohr-Kahaly S Gerharz CD Iversen S Oelert H Meyer J Improved diagnostic value of echocardiography in patients with infective endocarditis by transesophageal approach: a prospective study.Eur Heart J. 1988; 9: 43-53PubMed Google Scholar Although the cumulative experience to date has shown no clear advantage with use of TEE for ventricular tumors and thrombi, two such masses identified with TEE were not visualized with TTE. TEE provided superior imaging of the extracardiac masses, which were overlooked entirely or poorly characterized with precordial imaging. Usually, the physical characteristics of all types of masses were much better defined with TEE than with TTE. The single exception was identification of the attachment site of a lesion metastatic to the right ventricle, which was seen only with TTE. The study has a few weaknesses and limitations, some of which were pointed out by Reeder and colleagues in their discussion. Confirmation of the presence of a mass by surgical intervention, autopsy, or imaging procedure other than echocardiography was absent for some cases. The study design dictated that any mass not identified with TEE, even if visualized with TTE, would be excluded. One assumes that no such cases existed. This clarification, however, is important; in its absence, whether TEE detects more lesions than TTE is unclear. From another perspective, this approach disadvantages TTE, which at best may equal but never surpass TEE in identifying cardiac masses. Of interest, but of no importance to the conclusions of this report, is whether cardiac masses were identified by other imaging techniques but not detected by either TEE or TTE. Some of the criteria by which the authors judged TEE to “have added useful data” or altered the management of patients seem somewhat contrived. Perhaps more discrimination existed in their application than could be stated in a review of this nature. For example, was size actually “useful” in all cases for which it could be ascertained? Although perhaps clinically valid, use of TEE to exclude other masses seems logically tenuous. By what measure can one conclude that “nothing” is present? To use the technique under scrutiny as the gold standard for diagnosis begs the question a bit. How the authors used TEE in managing patients with atrial fibrillation is not specifically stated although I understand that they used TEE to select patients for cardioversion by excluding atrial thrombus. In light of this study and others, what is the current status of the diagnostic approach to cardiac masses? If TEE is clearly superior to TTE in achieving the right diagnosis in patients with cardiac masses, should it be an integral part of a complete cardiac ultrasound examination? As physicians committed to making right diagnoses compassionately, we are confronted with two groups of patients for whom the answer is pertinent. The first are those patients in whom a cardiac mass is detected unexpectedly during an examination performed to address other concerns. In our experience with TTE,9Fyke III, FE Seward JB Edwards WD Miller Jr, FA Reeder GS Schattenberg TT Shub C Callahan JA Tajik AJ Primary cardiac tumors: experience with 30 consecutive patients since the introduction of two-dimensional echocardiography.J Am Coll Cardiol. 1985; 5: 1465-1473Abstract Full Text PDF PubMed Scopus (128) Google Scholar most cardiac tumors were discovered unexpectedly during examinations done with other diagnoses in mind; in only 3 of 20 patients with left atrial myxomas were the echocardiograms ordered with a presumptive diagnosis of cardiac tumor. In accord with the authors' closing comments, I cannot recommend that TEE be universally performed even though more masses might be discovered thereby. The cost-to-benefit and risk-to-benefit ratios are unknown. TEE is less convenient because a physician must participate directly in the study; is approximately twice as expensive as TTE; involves small but real risks of esophageal perforation (perhaps 1 in 3,000 procedures), drug reactions, aspiration, and other complications not associated with TTE; and is more invasive and less comfortable for the patient. Some of these disadvantages may diminish as probes decrease in size, as equipment becomes less expensive, and as experience with the procedure increases. Because of the added information provided by TEE, most patients in whom masses are identified would benefit from a subsequent transesophageal examination. The second group of patients are those in whom the physician has cause to seek a cardiac mass aggressively. Included would be patients with chest roentgenographic findings suggestive of a mass such as an aneurysm of the left atrial appendage or pericardial cyst, patients with emboli to multiple circulations, and patients with pulmonary or systemic emboli in whom no other source is likely. In such patients, the physician would rarely be satisfied with a negative TTE because of what may be missed and would almost always follow a positive TTE with TEE because of the additional useful information often revealed. Because TEE seems to be the single most sensitive diagnostic tool in these patients, inclusion of this procedure in a complete echocardiographic examination seems appropriate. A discussion of this subject is incomplete until the considerable distance between purchasing TEE equipment and using it expertly has been emphasized. As most cardiologists can attest from personal experience, competence with a TTE transducer does not make one a capable transesophageal ultrasonographer. Obtaining good studies is extremely operator-dependent. The results described herein reflect the vast experience and careful training typical of the Mayo Clinic and cannot be extrapolated to laboratories with limited experience and less expertise. If physicians aspire to seek the right diagnosis compassionately in patients with cardiac masses, they must learn not only the role of this new procedure but also the discipline of a careful, gentle, methodical, and complete examination. For the present, the latter may be the greater challenge.