Purpose: To retrospectively evaluate the contribution of MRI to the diagnosis of arrhythmogenic right ventricular dysplasia (ARVD).
Patients and methods: Thirty two men and 18 women (mean age: 48.7 years) were imaged using gated spin echo scans and short axis cine MR. References were McKenna criteria (gold standard), and also what we called "strong presumption" which may correspond to early or localized patterns, but correspond to negative McKenna scores.
Results: One patient was claustrophobic; another one was lost to follow-up. In reference to McKenna score, diagnosis of ARVD was established in 12 patients (2 of whom had familial dilated biventricular cardiomyopathy with rhythmic expression). Sensitivity, specificity, positive predictive value, negative predictive value and prevalence were respectively: 75%, 75%, 50%, 90% and 25%. Using "strong presumption" criteria, we observed 14 true positives (with sensitivity of 82%, specificity of 87%, PPV of 78%, NPV of 90% and prevalence of 35%).
Conclusion: In our group, MRI was always performed before angiography. In our series, right ventricular wall T1W hyperintensity was the most frequent finding.