MIVA method (mitral region velocity analysis) for mitral regurgitation quantification

Abstract Background Mitral regurgitation (MR) quantification remains a challenging aspect in clinical practice. Actual recommendations emphasize the importance of quantitative methods, but some limitations have emerged, including low reproducibility and low applicability in routine clinical practice. Purpose To evaluate a novel method for estimating MR severity through the semiautomatic analysis of 2D color Doppler images from transthoracic echocardiography (TTE). Methods Prospective and single-center study including 193 patients with at least mild MR assessed with TTE. The MIVA (MItral region Velocity Analysis) method is based on the velocity information contained in each pixel of the Doppler colour map. The standard deviation of pixel velocities within a region of interest (ROI) including the mitral valve region is obtained in every systolic frame, and the frame-by-frame mean is calculated. MIVA’s (Method 1) diagnostic performance was compared with two additional methods (Figures A-C): PISA-derived effective regurgitant orifice area (EROA, Method 2) and the continuous wave Doppler (CWD) average pixel intensity (API, Method 3). Receiver Operating Characteristic (ROC) curves were generated using expert integrative criteria as the reference standard. Primary (PMR) and secondary (SMR) subsets were also analyzed. Results Most cases (56%) were SMR, and 53% were graded as significant (moderate-severe or severe). In the overall sample, PISA-EROA showed better diagnostic performance as compared to MIVA and API (Figure D). Regarding MR mechanisms, all 3 methods had a similar diagnostic performance (Figures E-F), whereas PISA-EROA remained slightly better in PMR cases. The optimal cutoffs for the detection of significant MR for MIVA and API were different in SMR and PMR subsets (MIVA: 0.32 vs 0.30, and API: 150 vs 115, for SMR and PMR, respectively). Conclusions MIVA is proposed as a new parameter for assessing MR severity, that does not require previous experience or high quality images. This pilot study suggests a high applicability and high diagnostic accuracy as compared to PISA-EROA, and API, another semi-automated method based on the CWD signal analysis. MIVA could improve consistency in MR quantification, and extend its applicability in scenarios where EROA is either unfeasible or unreliable.