Value and limitations of transesophageal echocardiography in infective endocarditis

Echocardiography has contributed considerably to the evolution in the management of patients with infective endocarditis. There is a clear hierarchy with respect to sensitivity of the different methods is superior when compared to 2-D and M-mode echocardiography in identifying both vegetations and perivalvular complications e.g. abscess formation, aneurysms, mural endocardial lesions. For patients with suspected endocarditis, in whom vegetations can not be clearly identified or in whom abscess formation is suspected with transthoracic echocardiography (TTE), transesophageal echocardiography (TEE) with mono- or multiplane scans has become the standard diagnostic procedure. For the examination of prosthetic valves it is the method of choice. It has even been suggested that it is employed as routine measure in all patients with suspected infective endocarditis. TEE is a safe semi-invasive technique with an extremely low complication rate and high sensitivity. Its specificity depends largely on the patient group which is examined. In patients with indicative clinical symptoms the specificity and the predictive value of vegetations are high. When used as screening method to assess echodense formations at cardiac valves particularly in the elderly, in whom degenerative changes prevail, its specificity and positive predictive value of vegetation-like structures are much lower. The negative predictive value of a negative transesophageal echocardiogram remains high, however. Valve abscesses are detected rarely by transthoracic echocardiography. It is the domain of TEE to assess them particularly in the aortic and mitral valve area. For them the specificity and positive predictive value of TEE in the diagnosis of infective endocarditis is high again. Of further importance was the observation that patients with vegetations of > 10 mm were more likely to suffer embolic complications. It should be noted, however, that infective endocarditis remains a clinical diagnosis: neither is the demonstration of a vegetation already the equivalent of endocarditis, nor does missing vegetations completely rule out the possibility of it. But without doubt, the presence of vegetations, of abscess formation or a concomitant pericardial effusion add valuable information to clinical diagnosis of infective endocarditis, which still needs a "high index of suspicion".