Pathological CMR findings and their clinical value in patients with high grade ventricular arrythmias without previously known cardiac conditions

Methods Patients with VAs, without known cardiac condition were included (n=71). CMR, echocardiography and 12-lead ECGs were performed in all patients. Most patients underwent electrophysiologic testing (EPS). CMR findings were correlated with ECG and/or EPS findings. A morphological substrate was defined as 1) underlying cardiac condition detected by CMR, known to cause associated arrhythmia, or 2) presence of an electro-anatomic correlate, if the location of a structural abnormality/scar tissue, matched the origin of the VA as defined by ECG and/or EPS. A positive CMR finding was defined as clinically relevant if 1) it resulted in a new diagnosis, or 2) a change of treatment, or 3) additional diagnostic procedures. Results Pathological CMR findings were found in 42 patients. The main findings were wall motion abnormalities (n=22), ventricular hypertrophy (n=10), dilatation (LV or RV or atrium, total=28), wall thinning (n=3), pericardial and/or pleural effusion (n=6), myocardial scar/structural abnormalies (n=14). Based on CMR a diagnosis of heart disease was established in 19 patients (27%; DCM n = 3, HCM n = 1, criteria for ARCM n = 2, post-myocarditis scar n = 7, post-myocardial infarction scar n = 2, myocarditis n = 4). In 15/71 patients (21%) the definite diagnosis was obtained by CMR exclusively. All CMR based diagnoses were classified as morphological substrates. In 8 patients the location of the relevant CMR finding (myocardial scar) directly matched the origin of the VA as defined by ECG or EPS. In none of these patients the relevant finding was detected by echocardiography. The diagnoses resulted in a change of treatment in 11 patients (15%) and a change of treatment plus additional diagnostic procedures in 8 patients (11%).