To Include or Not to Include: Quantification of Mitral Regurgitation in Patients with Mitral Valve Prolapse Using Cardiac MRI

See also article by Otemuyiwa et al in this issue.

Jeesoo Lee, PhD, is a research assistant professor of radiology specializing in cardiovascular flow imaging and fluid dynamics. His research aims to improve the diagnostic precision of valvular heart diseases and our understanding of the associated abnormal hemodynamics by imaging. His current research focuses on developing novel in vivo imaging methods for advanced cardiovascular hemodynamic characterization using echocardiography, 4D flow MRI, in vitro flow modeling, and computational fluid dynamics.

Michael Markl, PhD, is the vice chair for research in the department of radiology at Northwestern University Feinberg School of Medicine. He is also the Lester B. and Frances T. Knight Professor of Cardiac Imaging in the departments of radiology and biomedical engineering at Northwestern University. His research program focuses on the diagnosis and management of cardiovascular disease, including the development of novel imaging techniques for the assessment of cardiac structure, function, and hemodynamics, and discovery of mechanisms underlying cardiovascular disease development and stroke. He is internationally recognized as the pioneer of four-dimensional flow MRI, and his work in this area has advanced the understanding of cardiovascular disease processes as well as enhanced patient care. Dr Markl is a recipient of the RSNA Research Trainee Prize, the I.I. Rabi Young Investigator Award of the ISMRM, and the Distinguished Investigator Award of the Academy of Radiology Research. He is an associate editor for Radiology: Cardiothoracic Imaging, a fellow of the ISMRM and SCMR, a member of the executive committee of the Board of Trustees of the SCMR, and the past president of the SMRA.

Mitral valve prolapse is the most common cause of primary mitral regurgitation (1) and requires surgical or minimally invasive mitral valve repair for severe mitral regurgitation, which is often associated with impaired left ventricular (LV) function (2). Accurate assessment of mitral regurgitation severity and LV size and function are thus imperative for patient management. Cardiac MRI is a valuable imaging modality in this patient cohort as the reference standard for the assessment of LV volumes and global cardiac function such as stroke volume (SV) and ejection fraction (EF). In addition, cardiac MRI can also be used to quantify mitral regurgitant volume (MRV) indirectly by subtracting aortic forward flow measured by two-dimensional (2D) phase-contrast MRI from LV SV measured by a stack of short-axis cine images. Indirect quantification of MRV using cardiac MRI has been shown to be reproducible and predictive of patient outcomes such as postsurgical LV reverse remodeling (3,4). The standard workflow for the quantification of LV volumes and function is based on the delineation of the LV endocardial contours on short-axis cine images from the apex to the base. The location of the LV base is defined at the myocardial border (ie, mitral annulus). In case of mitral valve prolapse, however, the bulging of the mitral leaflets into the left atrium during systole extends the LV cavity beyond the mitral annulus. The standard LV function method does not account for this additional LV volume during systole and may thus underestimate LV end-systolic volume and consequently overestimate LV SV. This would result in a larger MRV and possible overestimation of mitral regurgitation severity. There is an ongoing debate as to whether inclusion of the prolapsed volume can cause clinically relevant discrepancies in the clinical assessment of mitral regurgitation and LV function. The work “Effects of Mitral Valve Prolapse on Quantification of Mitral Regurgitation and Ejection Fraction Using Cardiac MRI” by Otemuyiwa et al (5) in this issue of Radiology: Cardiothoracic Imaging provides new data illustrating that the inclusion versus exclusion of the prolapsed volume can result in significant changes in MRV, mitral regurgitant fraction (MRF), and LV EF. The authors also evaluated the accuracy of the inclusion versus exclusion approach by acquiring a reference standard independent of LV volumes: 2D phase-contrast MRI measurements of mitral inflow and aortic flow to quantify mitral regurgitation.

In their study with 19 patients, the authors showed that MRV and MRF were in good agreement with the reference standard when the prolapsed volume was included in the calculation of the LV end-systolic volume, but significantly larger in magnitude (by 13.8 mL and 8.6% on average for MRV and MRF, respectively) when excluding the prolapsed volume (ie, the standard approach). Importantly, the study also showed that including the prolapsed volume when calculating MRV can cause discordant mitral regurgitation severity assessments. Severe mitral regurgitation had to be reclassified to nonsevere in 16% of the patients when MRF greater than 50% was considered severe. The validation of MRV and MRF against the reference standard independent of LV volumetry provides important new information as prior studies focused on the changes in mitral regurgitation quantities caused by prolapsed volume (6,7). A study by Vincenti et al (6) validated LV SV by using the right ventricular SV as a reference in patients with mitral valve prolapse but without mitral regurgitation. They showed that including the prolapsed volume produced more consistent SV measurements between the left and right ventricle. These findings indicate that including the prolapsed volume for MRV quantification may provide anatomically more appropriate LV volumes for improved accuracy of MRV assessment.

Of note, the study also showed that including the prolapsed volume may significantly lower LV EF compared with the standard approach. In the study cohort of 19 patients, 37% had to be reclassified from the normal to abnormal LV systolic function category. However, it is debatable whether the lower LV EF truly reflects the reduced LV function. The authors discuss this matter in favor of excluding the prolapsed volume when assessing LV EF for physiologic and prognostic reasons. This is supported by a prior study by Wolff and Uretsky (7). They showed that LV EF calculated with the LV end-systolic volume using the standard approach (ie, LV base at the mitral annulus) yielded a better correlation with global myocardial strain measured with the feature-tracking technique. In addition, LV EF derived by including the prolapsed volume may have limited prognostic value due to a lack of outcome data, as most available clinical data are based on the LV volumes that define the base at the mitral annulus. Of note, echocardiography, which has a larger outcome database, quantifies LV volumes without considering the prolapsed volume. Therefore, even though it may be anatomically more correct to consider the prolapsed volume as a part of the LV end-systolic volume, the authors suggest the use of the standard approach for cardiac function assessment while including the prolapsed volume for MRV and MRF quantification.

The study had a number of limitations that should be considered in the context of the main study findings. A major challenge regarding the assessment of mitral regurgitation is the lack of a robust reference standard technique for quantification. Currently, echocardiography serves as the primary imaging modality for mitral regurgitation in clinical settings. It allows for a comprehensive evaluation of the valves’ and surrounding structures’ function. However, inherent weaknesses (2D, one-directional velocity, operator variability) and the complex nature of mitral regurgitation flow (turbulent flow jet, central or eccentric direction with dynamically changing regurgitant orifice area) limit the reproducibility and precision of echocardiographic quantification of mitral regurgitation. In the current study (5), the authors used flow measurements with 2D phase-contrast MRI as the reference standard. MRV and MRF were derived by subtracting aortic forward flow from mitral inflow volume. However, this indirect quantification approach, which also is the case of the volumetric approach, is prone to amplification of errors inherent in the measurement of each component. This amplified error may become large compared with the magnitude of MRV as the two volumes are generally larger than MRV. Further, the use of 2D static imaging planes to measure mitral inflow is prone to additional errors due to substantial through-plane mitral valve movement. Mitral regurgitation quantification using four-dimensional (4D) flow MRI may be a promising new technique to overcome the limitations of indirect techniques. 4D flow MRI provides three-directional intracardiac blood velocity vectors with whole-heart coverage over the cardiac cycle, capturing the mitral regurgitation flow jet and its movement in three dimensions (3D) (8). With recent developments of valve and jet flow tracking techniques, mitral regurgitation flow can be directly measured by placing a plane across the 3D mitral regurgitation flow jet that follows the jet movement and adapts to the changing jet direction for improved MRV quantification (9,10). However, challenges still exist for clinical implementation of 4D flow MRI, such as long data acquisition times and cumbersome postprocessing.

In summary, this study (5) presents important new data focused on improving the quantification of mitral regurgitation in patients with mitral valve prolapse using the indirect volumetric approach, which is probably the most widely used cardiac MRI method for mitral regurgitation assessment. Further validation studies in larger patient cohorts with variable mitral regurgitation severity and appropriate reference standards are warranted to validate the study findings and confirm the diagnostic value of including the prolapsed volume for mitral regurgitation severity assessment.