Cardiac magnetic resonance imaging assessment:
Cardiac magnetic resonance imaging was performed with 1.5 tesla or 3.0-tesla unit scanners (Amira and Verio, Siemens Healthineers, Erlangen, Germany). Patients were scanned with the electrocardiogram (ECG), triggering a 16-channel surface phased array of body coils. Briefly, 10 to 12 consecutive short-axis images covering the entire LV and 2-, 3-, and 4-chamber long-axis images were acquired with a cine steady-state free precession sequence (SSFP) to assess myocardial function mass and quantification of non-compaction. Two-dimensional LGE images were acquired in short axis, 2, and 4 chamber views by phase-sensitive inversion recovery sequence (PSIR) ten to 15 min after 0.2 mmol/kg gadolinium contrast agent injection. Cardiac volume, function, and mass on cardiac images were analyzed with software; Medis medical imaging systems-Medis Suite 3.1 (Leiden, Netherlands) by three radiologists for disagreements on data between two readers. A consensus agreement was achieved with the third expert opinion. The non-compacted and compacted ratio calculated distal to the papillary muscle that any segment reveals the highest proportion except the left ventricular apex. LGE distribution was reviewed in long and short axis contrast-enhanced images, and LGE presence was accepted in short and long-axis imaging planes. LGE distributions are visually classified as sub-endocardial, mid-myocardial (mid-wall), sub-epicardial, and right ventricular insertion involvement. Two radiologists determined the LGE distribution pattern; a third senior radiologist was consulted when there was disagreement.