Value of cardiovascular MR in diagnosing left ventricular non-compaction cardiomyopathy and in discriminating between other cardiomyopathies

OBJECTIVES: To analyse the value of cardiovascular magnetic resonance (CMR)-derived myocardial parameters to differentiate left ventricular non-compaction cardiomyopathy (LVNC) from other cardiomyopathies and controls. METHODS: We retrospectively analysed 12 patients with LVNC, 11 with dilated and 10 with hypertrophic cardiomyopathy and compared them to 24 controls. LVNC patients had to fulfil standard echocardiographic criteria as well as additional clinical and imaging criteria. Cine steady-state free precession and late gadolinium enhancement (LGE) imaging was performed. The total LV myocardial mass index (LV-MMI), compacted (LV-MMI(compacted)), non-compacted (LV-MMI(non-compacted)), percentage LV-MM(non-compacted), ventricular volumes and function were calculated. Data were compared using analysis of variance and Dunnett's test. Additionally, semi-quantitative segmental analyses of the occurrence of increased trabeculation were performed. RESULTS: Total LV-MMI(non-compacted) and percentage LV-MM(non-compacted) were discriminators between patients with LVCN, healthy controls and those with other cardiomyopathies with cut-offs of 15 g/m(2) and 25 %, respectively. Furthermore, trabeculation in basal segments and a ratio of non-compacted/compacted myocardium of ≥3:1 were criteria for LVNC. A combination of these criteria provided sensitivities and specificities of up to 100 %. None of the LVNC patients demonstrated LGE. CONCLUSIONS: Absolute CMR quantification of the LV-MMI(non-compacted) or the percentage LV-MM(non-compacted) and increased trabeculation in basal segments allows one to reliably diagnose LVNC and to differentiate it from other cardiomyopathies. KEY POINTS: Cardiac magnetic resonance imaging can reliably diagnose left ventricular non-compaction cardiomyopathy. Differentiation of LVNC from other cardiomyopathies and normal hearts is possible. The best diagnostic performance can be achieved if combined MRI criteria for the diagnosis are used.