Quantitation of mitral regurgitation after percutaneous MitraClip repair: comparison of Doppler echocardiography and cardiac magnetic resonance imaging.

Abstract

Percutaneous valve intervention for severe mitral regurgitation (MR) using the MitraClip is a novel technology. Quantitative assessment of residual MR by transthoracic echocardiography (TTE) is challenging, with multiple eccentric jets and artifact from the clips. Cardiovascular magnetic resonance (CMR) is the reference standard for left and right ventricular volumetric assessment. CMR phase-contrast flow imaging has superior reproducibility for quantitation of MR compared to echocardiography. The objective of this study was to establish the feasibility and reproducibility of CMR in quantitating residual MR after MitraClip insertion in a prospective study. Twenty-five patients underwent successful MitraClip insertion. Nine were excluded due to non-magnetic resonance imaging (MRI) compatible implants or arrhythmia, leaving 16 who underwent a comprehensive CMR examination at 1.5 T (Siemens Aera) with multiplanar steady state free precession (SSFP) cine imaging (cine CMR), and phase-contrast flow acquisitions (flow CMR) at the mitral annulus atrial to the MitraClip, and the proximal aorta. Same-day echocardiography was performed with two-dimensional (2D) visualization and Doppler. CMR and echocardiographic data were independently and blindly analyzed by expert readers. Inter-rater comparison was made by concordance correlation coefficient (CCC) with 95% confidence intervals (CIs), and Bland-Altman (BA) methods. Mean age was 79 years, and mean LVEF was 44%±11% by CMR and 54%±16% by echocardiography. Inter-observer reproducibility of echocardiographic visual categorical grading by expert readers was poor, with a CCC of 0.475 (-0.7, 0.74). Echocardiographic Doppler regurgitant fraction reproducibility was modest (CCC 0.59, 0.15-0.84; BA mean difference -3.7%, -38% to 31%). CMR regurgitant fraction reproducibility was excellent (CCC 0.95, 0.86-0.98; BA mean difference -2.4%, -11.9 to 7.0), with a lower mean difference and narrower limits of agreement compared to echocardiography. Categorical severity grading by CMR using published ranges had good inter-observer agreement (CCC 0.86, 0.62-0.95). CMR performs very well in the quantitation of MR after MitraClip insertion, with excellent reproducibility compared to echocardiographic methods. CMR is a useful technique for the comprehensive evaluation of residual regurgitation in patients after MitraClip. Technical limitations exist for both techniques, and quantitation remains a challenge in some patients.