Case Presentation

A 74-year-old man presented to the emergency department with a non–ST-segment elevation myocardial infarction and congestive heart failure. The patient consented to have his medical information included in this report. His history included chronic atrial fibrillation (AF), a right-sided colon cancer, and hypertension. Anticoagulation was being withheld because of ongoing anemia and the risk for further bleeding from the colon cancer. He did not have a history of transient ischemic attack or stroke. He was admitted for treatment and investigation. Coronary angiography revealed severe triple-vessel coronary artery disease, and it was felt that coronary artery bypass was indicated prior to definitive surgical management of the colon cancer. Preoperative transthoracic echocardiography (TTE) revealed a left ventricular ejection fraction of 39%, a mildly hypokinetic right ventricle, and mild mitral regurgitation. No atrial masses were noted.

The patient was brought to the operating room, invasive monitors were placed, and stable induction of anesthesia was carried out. A transesophageal echocardiographic probe was placed for intraoperative monitoring. Prebypass comprehensive transesophageal echocardiography (TEE) confirmed the preoperative findings on TTE, with the exception of an unexpected large mass, measuring approximately 1.8 × 1.0 cm, in the left atrial (LA) appendage (LAA) ( Figure 1 , [CR] ). Spontaneous echo contrast was also evident in the LAA. Given the clinical scenario and the echocardiographic findings, the mass was presumed to be thrombus. Careful examination of the left atrium showed enlargement but no additional masses. The left ventricle demonstrated hypokinetic inferior, septal, and lateral walls, with a calculated ejection fraction of 36%. The surgeon was informed about the unexpected finding, and arrangements were made for bicaval cannulation with planned thrombus extraction via a left atriotomy once cardiopulmonary bypass was initiated. Until that time, manipulation of the heart was kept to a minimum to avoid embolization.

Zoomed midesophageal 2-chamber view shows a large thrombus in the LAA (arrow) . LA , Left atrium; LV , left ventricle.

On the initiation of bypass, the transesophageal echocardiographic reevaluation of the LAA revealed that the thrombus had become dislodged and was freely floating in the left atrium ( Figure 2 , [CR] ). Fortunately, the aortic cross clamp had been applied and cardioplegic arrest of the heart had occurred, and there was no flow through the heart, keeping the thrombus within the left atrium. At this point, it was necessary to remove the mobile thrombus. The interatrial groove was dissected, and the left atrium was incised. As it was incised, the clot spontaneously extruded through the incision. This was confirmed with transesophageal echocardiographic examination of the atrium ( [CR] ). The clot came out in a single piece and was sent for pathologic examination, which later confirmed thrombus. The atrium was then closed, and the heart was reflected into the wound to expose the LAA, which was oversewn. The remainder of the operation was unremarkable. Postbypass TEE revealed improved left ventricular function, with an ejection fraction estimated at 50% to 55%. The LAA had been oversewn with no evidence of residual flow by color Doppler. The patient awoke in the intensive care unit 2 hours after his arrival with no neurologic deficits and was extubated shortly thereafter.

Zoomed midesophageal 4-chamber view shows a mobile thrombus (arrow) in the left atrium (LA) after the initiation of cardiopulmonary bypass.

Discussion

This case demonstrates the risk for embolization of LAA thrombi during cardiac surgery. Despite careful attention by the surgeon after being informed of its presence, the thrombus became mobile on the initiation of cardiopulmonary bypass, likely as a result of decompression of the left atrium. Several cases of accidental dislodgement of incidentally discovered LA and LAA thrombi have been previously reported. Matsuyama et al reported real-time transthoracic echocardiographic embolization of a thrombus from the LAA into the systemic circulation. The present report is the first to demonstrate by way of echocardiography the dislodgement of a thrombus from the LAA on cardiopulmonary bypass initiation and its subsequent extraction.

LA thrombus is a common finding in patients with nonrheumatic AF, with an incidence of approximately 7% in one series of nonselected outpatients. Additional risk factors for LA thrombus formation include mitral stenosis, an enlarged left atrium, advanced age, and congestive heart failure. The vast majority of the thrombi that form in nonrheumatic AF are found within the LAA. These thrombi are a source of embolic stroke, and as such Johnson et al termed the LAA the “most lethal human attachment.” Surgical and percutaneous closure devices have been advocated as a means to reduce thrombus formation and subsequent stroke.

With sensitivity and specificity approaching 100%, TEE is considered the gold standard for detecting LAA thrombi. Some authors have recommended the use of intraoperative TEE prior to heart manipulation in any case in which LA thrombus is likely. A complete LA examination is performed from the esophageal views with complete scanning of the left atrium. By carefully altering rotation, the omniplane angle, and depth in a stepwise fashion, the entire body of the left atrium and the interatrial septum can be visualized. TEE has lower sensitivity (81%) for detecting thrombus in the main LA cavity but maintains specificity near 100%. The LAA is best visualized in the midesophageal 2-chamber view. Stepwise advancement of the omniplane angle with slight rotation of the probe can offer a more comprehensive view. The LAA can also be visualized in the transgastric 2-chamber view. Stasis of blood, a predisposing factor to thrombus formation, is manifested as “smoke” or spontaneous echo contrast in the LAA. An LAA pulse-wave Doppler emptying velocity < 0.2 m/s is also associated with thrombus formation. Assessment of the LAA for thrombi may be complicated; false-negative results may occur because the LAA is usually multilobed, and a thrombus may be overlooked. False-positive results may occur, as prominent trabeculations of pectinate muscles can often be mistaken for thrombi. Differences in echodensity help distinguish thrombus from adjacent cardiac structures. LAA thrombi have a similar appearance, but assessment is complicated by LAA trabeculations. Trabeculations move with and are confluent with the LA walls, whereas thrombi have independent motion. Given the clinical scenario, it was immediately suspected that the mass in this case was thrombus. The differential diagnosis of any LA mass also includes neoplasm, the most likely being an atrial myxoma. However LAA myxomata have been infrequently reported, and other primary tumors of the left atrium are rare.

Although some have advocated routine LAA exclusion during cardiac surgery, this practice is not standard of care. If closure is performed, TEE has an important role in the assessment of success of surgical exclusion of the LAA. This is typically confirmed by the absence of color flow Doppler between the left atrium and LAA. Recent reports suggest a high incidence of failed surgical closure of the LAA, with suture or stapler exclusion of the LAA being less successful than excision.

In this case, TTE 10 days prior to the operation failed to reveal a thrombus. This may have been due to the poor sensitivity of TTE for LAA thrombus detection or the possibility that the thrombus developed in the interval between the TTE and the operation. The unexpected intraoperative finding of this large thrombus changed surgical management in this case, specifically bicaval cannulation and an atriotomy. A mobile LA thrombus is generally considered an indication for urgent preventive surgery. A large mobile thrombus may fragment and embolize or cause acute mitral valve orifice occlusion. It is not clear whether or not a large nonmobile thrombus should be managed surgically or medically. Likewise, the role of LAA thrombectomy for the incidentally discovered LAA thrombus in a patient who is scheduled for bypass surgery is largely unknown. An extensive literature search reveals a paucity of reports or recommendations regarding the best course of action in these cases. This may be a result of the rarity of this incidentally finding or the relatively new use of TEE as a standard monitor for straightforward bypass surgery. If surgical removal of a thrombus is not undertaken, manipulation of the heart should be kept to a minimum. If a thrombus becomes mobile, as in this case, it is necessary to remove it to prevent embolization or mitral valve obstruction. In any case, the discovery of a LAA thrombus in a patient presenting for cardiac surgery necessitates continued reassessment of the thrombus until the aortic cross clamp is placed and even after cardiopulmonary bypass if the thrombus is not removed.

This case highlights the importance of TEE in those patients presenting for cardiac surgery who are at risk for LA thrombus formation and supports its routine use these patients. This patient had risk factors for the development of atrial thrombus: advanced age, congestive heart failure, and chronic AF. In addition, his anticoagulation had been withdrawn months earlier because of the risk for bleeding. In this case, routine intraoperative TEE revealed a potentially devastating situation of a LAA thrombus that mobilized on the initiation of cardiopulmonary bypass. As in other studies, this case demonstrates that intraoperative TEE for cardiac surgery often results in unexpected findings that lead to a change in surgical management. This case also supports the removal of an incidentally discovered LAA thrombus in a patient presenting for cardiac surgery, as the minimally increased risks of the change in surgical approach are outweighed by the potential benefits of avoiding an accidental dislodgement and embolization.