Arrhythmogenic right ventricular cardiomyopathy (ARVC) is a potentially fatal disease, which is often difficult to diagnose. As a non-invasive test cardiac magnetic resonance imaging (CMR) has become an important tool in establishing the diagnosis. The aim of this study was to evaluate the diagnostic and prognostic value of CMR in patients with suspected ARVC and to assess the long-term outcome of patients with CMR-diagnosed ARVC. Thirty-six patients with suspected ARVC (26 male, 10 female, median age 41 years) underwent non-invasive and invasive clinical tests as gold standard for ARVC diagnosis. ARVC was clinically diagnosed in 19 patients and excluded in 17 patients. Both groups underwent CMR, and diagnosis was confirmed by CMR in 16/18 patients with clinically diagnosed ARVC (sensitivity 89%), and correctly excluded in 14/17 of patients with clinically excluded ARVC (specificity 82%). This result indicates a positive predictive value of the CMR of 84%, and a negative predictive value of 88%, respectively (p < 0.0001). Using a scoring system, multiple CMR parameters were compared in the two groups in regard of the clinical diagnosis. By univariate analysis, right ventricular fatty tissue infiltration (p = 0.0003) was predictive for diagnosis. Compared by outcome, 37% of patients with clinically and by CMR-diagnosed ARVC had an arrhythmic event during a mean follow-up of 16 +/- 11 months. These data suggest that CMR is a highly sensitive and specific method to diagnose or exclude ARVC, and thus, has an important prognostic impact on predicting arrhythmia free survival.