Could we use cardiac MRIfor diagnosis of infective endocarditis

Abstract Funding Acknowledgements Type of funding sources: None. Introduction Echocardiography is a reliable imaging tool in diagnosing infective endocarditis (IE) according to "modified Dukes criteria"; however, it lacks tissue characterization of cardiac masses, and transthoracic echocardiography (TTE) has low sensitivity in detection of vegetations. Moreover, transesophageal echo (TEE) is considered semi-invasive and intolerable in some situations .On the other hand, cardiac magnetic resonance (CMR) is a powerful true 3D imaging tool with wide field of view, and superior capability of tissue characterization of cardiac masses; however, modest information is available about utility of CMR in diagnosing IE. Purpose To assess the feasibility of CMR to identify vegetations and complications of IE, and compare obtained findings with those obtained from echocardiography regarding: presence, number, size of vegetations, and detection of cardiac complications (e.g. aortic root abscess, periannular abscess, and shunts). Materials and Methods Eighty consecutive patients with suspected IE were enrolled in the study. All patients underwent TTE; only those with left-sided lesions underwent TEE.  When clinical situation allowed, CMR examination using 1.5 Tesla magnet were performed using the following sequences: Steady-state free precession (SSFP), T1W (± fat suppression) and T2W, first-pass perfusion, and delayed hyperenhancement. Moreover, chest and abdominal survey was done. Results Sixty-one patients (45 males and 16 females) were able to undergo and complete CMR study. Affection of tricuspid valve was seen in 39.3%, mitral 31.1%, aortic 24.6%, and pulmonary 4.9% of cases. All vegetations visualized by echocardiography were also clearly detected and confirmed by CMR. The sensitivity, specificity, accuracy and Kappa agreement of CMR with echocardiography in depicting >0.5cm vegetations were all 100%. By tissue characterization, vegetations resemble features of thrombi (with variation in signal intensity according to age of vegetation). However, in some cases, masses have a unique pattern that is different from vegetations of IE, and subsequently other diagnoses were suggested (e.g. fibroelastoma, Libman-Saks endocarditis). In 22 patients, CMR provided more information than echocardiography, and in 6 patients, CMR made a paradigm shift of diagnosis. Conclusion Cardiac MRI is a powerful imaging tool in diagnosis of IE and its complications. In comparison with echocardiography, CMR can identify the presence, numbers, and size of vegetations accurately. Moreover, with its unique ability of tissue characterization, CMR helps distinguish vegetations from other masses; therefore changes the diagnosis of IE and subsequent management in some patients.