DIAGNOSTIC YIELD OF CARDIOVASCULAR MAGNETIC RESONANCE (CMR) SCREENING FOR ARRHYTHMOGENIC RIGHT VENTRICULAR CARDIOMYOPATHY (ARVC) STRATIFIED BY BASELINE ECHOCARDIOGRAPHY FINDINGS OF THE RIGHT VENTRICLE

Abstract

Arrythmogenic right ventricular cardiomyopathy (ARVC) is a rare inherited cardiomyopathy but an important cause of sudden death. Cardiovascular Magnetic Resonance (CMR) imaging is considered the gold-standard for exclusion of its structural phenotype, however, limited availability poses challenge in many communities. We aimed to study the relative diagnostic yield of CMR in patients with versus without abnormal right ventricular (RV) findings by routine 2D echocardiography. 447 patients referred for the screening of ARVC between February 2015 and December 2017 were identified from the Cardiovascular Imaging Registry of Calgary (CIROC). A standardized CMR imaging and analysis protocol was used to assess chamber volumes, function and architectural changes in accordance with the 2010 modified Task Force Criteria (mTFC). All patients having echocardiographic imaging within 12 months prior to CMR were identified and a blinded review of reports was performed. A final clinical diagnosis was derived based on the assimilation of all clinical information, and was classified as meeting or not meeting mTFC for ARVC. Of 447 enrolled patients, 387 (76.6%) had echocardiogram performed and were studied. A final diagnosis of definite ARVC was established in nine (2.3%) patients. Echocardiography reported no RV abnormalities in 283 (73.1%) patients (Group A), while 104 patients (26.9%) had at least one abnormal finding (Group B), defined as; reduced RV function, regional RV wall motion abnormality, global or focal dilatation, aneurysm or “hyper-reflective” trabeculations. In Group A, the yield of any mTFC by CMR was 4 (1.4%) with three patients having a minor and one having a major criterion. Two Group A patients (0.7%) ultimately received a diagnosis of ARVC, neither showing any imaging-based abnormalities. In Group B, the yield of any mTFC by CMR was 13 (12.5%) with nine subjects having a major and 4 having a minor criterion. Seven Group B patients (6.7%) were eventually diagnosed with ARVC, 5 (4.8%) meeting a CMR-based mTFC and 2 (1.9%) with normal CMR. Overall, the sensitivity and specificity of CMR for ARVC was 0% and 98.6% in Group A, and 71.4% and 91.7% in Group B. Among patients without RV abnormality by echocardiography referred to CMR for suspected ARVC, the yield of a CMR-based mTFC was low (1.4%) and did not lead to a diagnosis of ARVC in any subject. In contrast, patients with an RV abnormality had a CMR-based yield of 12.5%, leading to a diagnosis of ARVC in 4.8% of patients.