Location, size, and morphologic characteristics of left atrial thrombi as assessed by transesophageal echocardiography in relation to systemic embolism in patients with rheumatic mitral valve disease

Abstract

A systemic embolism (SE) is one of the most important complications in patients with rheumatic mitral stenosis; the documented frequency of SE in patients with mitral stenosis varies from 10% to 45%.1,2 In surgical and postmortem clinicopathologic series, left atrial (LA) thrombus has been associated with SE.1,2 Recently, in transthoracic and transesophageal echocardiographic studies, the presence of spontaneous echocardiographic contrast (SEC) and/or thrombus within the left atrium has been found to increase the risk of SE. There are a limited number of studies evaluating the association between the risk of SE and the location, size, and morphologic characteristics of LA thrombi.3,5 Recently, it was reported that both the size and mobile character of LA thrombi were associated with increased risk of SE.14 We aimed to assess transesophageal echocardiographic correlates of recent SE, with specific reference to size, location, and morphologic characteristics of LA thrombi, and to investigate whether the intensity of left atrial SEC is related to the characteristics of concomitant LA thrombi in patients with rheumatic mitral valve disease. • • • The study population comprised 474 consecutive patients (320 women and 154 men, mean age 40 16 years) with rheumatic mitral valve disease who underwent transthoracic and transesophageal echocardiography 1 to 5 days before mitral valve surgery. Patient characteristics are listed in Table 1. Pure or predominant mitral stenosis was diagnosed in 333 patients (70.2%) and mitral regurgitation in the remaining patients. Two hundred sixty-seven patients (56.3%) had atrial fibrillation, and the remaining patients were in sinus rhythm. A history of anticoagulation was noted in 72 patients with history of SE; 34 patients with a history of silent LA thrombi were diagnosed by previous routine transthoracic echocardiographic studies. The mean value of the international normalized ratio was 2.5 1.2 in patients with SE and 2.6 1.1 in patients without SE. Mean time interval between the SE and the transesophageal echocardiographic study was 19 6 days. Patients with potential cardiac sources of SE, such as prosthetic heart valves; infective endocarditis; severe left ventricular dysfunction with or without left ventricular mural thrombi; and LA mass such as myxoma, protruding atheroma, aortic dissection, patent foramen ovale, and intracardiac shunts were excluded from the study. In patients without a history of SE, the presence of severe carotid artery disease was excluded by absence of systolic murmur over the carotid arteries. Transthoracic and transesophageal echocardiographic examinations were performed by a 3.25-MHz transthoracic and a 5-MHz multiplane transesophageal probes connected to a Vingmed CFM800 echocardiographic system (Horten, Norway). Written informed consent was obtained from each patient before entry into the study, and the protocol was approved by the institutional review board. Transesophageal echocardiography was performed after 4 hours of fasting, under topical anesthesia with 10% lidocaine and conscious sedation with intravenous midazolam. The mitral valve area was measured by pressure halftime and planimetric methods.15 Pure or predomFrom the Departments of Cardiology and Cardiovascular Surgery, Kosuyolu Heart and Research Hospital, Istanbul, Turkey. Dr. Ozkan’s address is: Department of Cardiology, Kosuyolu Heart and Research Hospital, Kadikoy, 81020, Istanbul, Turkey. E-mail: memoozkan@ superonline.com. Manuscript received June 29, 2002; revised manuscript received and accepted November 12, 2002. TABLE 1 Clinical and Echocardiographic Characteristics of Patients