Quantification of left ventricular fibrosis in arrhythmic mitral valve prolapse patients: comparison of different semi-automated techniques assessed by cardiac magnetic resonance

Abstract Background Left ventricular (LV) fibrosis has a key role in arrhythmogenesis mitral valve prolapse (MVP) patients. Cardiac magnetic resonance (CMR) demonstrated able to clearly identify LV with the post-contrast late gadolinium enhancement (LGE) images. Despite pivotal of arrhythmogenesis, quantification and identification reproducible method accurately measure LGE arrhythmic MVP patients not been recognized. Purpose We aimed compare different semi-quantitative methods for assessed by CMR, order most one, Methods 66 normal systolic function without significant regurgitation were enrolled. Semi-automated gray-scale thresholding technique using full at half maximum (FWHM) 2, 3 5 standard deviation (SD) above remote myocardium used compared visual assessment (Fig. 1). Results was identified 41 (62%) quantified 2). The mean quantity visually 2.40±1.07% or 1.40±0.82 g. With FWHM, resulted 3.56±1.23% 1.99±1.13 Using thresholding, 9.2±3.1% 4.82±2.28 g 2-SD, 5.72±1.75% 3.06±1.47 3-SD 2.36±0.99% 1.29±0.79 5-SD. 5-SD measurement percentage good correlation (2.402±1.075 vs 2.363±0.9909, p: 0.543). all intra inter-observer agreement, threshold quantification, both grams, revealed less intra-observer (respectively, ICC: 0.976 0.966) variability (respectively 0.948 0.935) when assessment. Conclusion This is first study that CMR semi-automated population best an optimal reproducibility. Funding Acknowledgement Type funding sources: None.