Role of transesophageal echocardiography in the management of thromboembolic stroke

Abstract

Cardiac causes of stroke account for approximately 20% of strokes occurring in the United States. Transthoracic echocardiography (TTE) remains the cornerstone of non-invasive cardiac imaging, but transesophageal echocardiography (TEE) is superior for identifying potential cardiac sources of emboli, including left atrial thrombi, valvular vegetations, thoracic aortic plaque, patent foramen ovale, and spontaneous left atrial echocardiographic contrast. The diagnostic yield of TEE for potential cardiac causes of thromboembolism exceeds 50%. The impact of TEE on the clinical management of this group, however, remains undefined for most TEE-specific diagnoses. Thus, routine use of TEE in these patients has been questioned. The diagnostic yield is highest if the clinical history/physical examination suggests a cardiac source. However, the clinical scenario often dictates patient management, and TEE data are used to "validate" the clinical impression. Data from large, prospective, randomized (aspirin/warfarin) studies, in which TEE data are obtained from patients with suspected cardiac thromboembolism, are needed. If specific TEE diagnoses can be identified in which defined therapies are beneficial, "source of embolism" will continue to be the most common indication for TEE referral. In this paradigm, TEE (without initial TTE) will probably become a more direct diagnostic pathway. However, if these studies demonstrate that all patients with suspected cardiac source benefit from one (or no) therapy, independent of TEE data, referrals for TEE will decline. Results of ongoing randomized trials to evaluate the efficacy of TEE in patients with cryptogenic stroke or transient ischemic attack are awaited.