Echocardiographic assessment of mitral regurgitation: discussion of practical and methodologic aspects of severity quantification to improve diagnostic conclusiveness

High-Throughput Deep Learning Detection of Mitral Regurgitation

BACKGROUND

Diagnosis of mitral regurgitation (MR) requires careful evaluation by echocardiography with Doppler imaging. This study presents the development and validation of a fully automated deep learning pipeline for identifying apical 4-chamber view videos with color Doppler echocardiography and detecting clinically significant (moderate or severe) MR from transthoracic echocardiograms.

METHODS

A total of 58 614 transthoracic echocardiograms (2 587 538 videos) from Cedars-Sinai Medical Center were used to develop and test an automated pipeline to identify apical 4-chamber view videos with color Doppler across the mitral valve and then assess MR severity. The model was tested internally on a test set of 1800 studies (80 833 videos) from Cedars-Sinai Medical Center and externally evaluated in a geographically distinct cohort of 915 studies (46 890 videos) from Stanford Healthcare.

RESULTS

In the held-out Cedars-Sinai Medical Center test set, the view classifier demonstrated an area under the curve (AUC) of 0.998 (0.998-0.999) and correctly identified 3452 of 3539 echocardiography videos as having color Doppler information across the mitral valve (sensitivity of 0.975 [0.968-0.982] and specificity of 0.999 [0.999-0.999] compared with manually curated videos). In the external test cohort from Stanford Healthcare, the view classifier correctly identified 1051 of 1055 manually curated videos with color Doppler information across the mitral valve (sensitivity of 0.996 [0.990-1.000] and specificity of 0.999 [0.999-0.999]). In the Cedars-Sinai Medical Center test cohort, MR moderate or greater in severity was detected with an AUC of 0.916 (0.899-0.932) and severe MR was detected with an AUC of 0.934 (0.913-0.953). In the Stanford Healthcare test cohort, the model detected MR moderate or greater in severity with an AUC of 0.951 (0.924-0.973) and severe MR with an AUC of 0.969 (0.946-0.987).

CONCLUSIONS

In this study, a novel automated pipeline for identifying clinically significant MR from full transthoracic echocardiography studies demonstrated excellent performance across large numbers of studies and across multiple institutions. Such an approach has the potential for automated screening and surveillance of MR.

The role of echocardiography in sports cardiology: An expert opinion statement of the Italian Society of Sports Cardiology (SIC sport)

Transthoracic echocardiography (TTE) is routinely required during pre-participation screening in the presence of symptoms, family history of sudden cardiac death or cardiomyopathies <40-year-old, murmurs, abnormal ECG findings or in the follow-up of athletes with a history of cardiovascular disease (CVD). TTE is a cost-effective first-line imaging modality to evaluate the cardiac remodeling due to long-term, intense training, previously known as the athlete's heart, and to rule out the presence of conditions at risk of sudden cardiac death, including cardiomyopathies, coronary artery anomalies, congenital, aortic and heart valve diseases. Moreover, TTE is useful for distinguishing physiological cardiac adaptations during intense exercise from pathological behavior due to an underlying CVD. In this expert opinion statement endorsed by the Italian Society of Sports Cardiology, we discussed common clinical scenarios where a TTE is required and conditions falling in the grey zone between the athlete's heart and underlying cardiomyopathies or other CVD. In addition, we propose a minimum dataset that should be included in the report for the most common indications of TTE in sports cardiology clinical practice.

Echocardiographic assessment of atrial, ventricular, and valvular function in patients with atrial fibrillation-an expert proposal by the german working group of cardiovascular ultrasound

Echocardiography in patients with atrial fibrillation is challenging due to the varying heart rate. Thus, the topic of this expert proposal focuses on an obvious gap in the current recommendations about diagnosis and treatment of atrial fibrillation (AF)-the peculiarities and difficulties of echocardiographic imaging. The assessment of systolic and diastolic function-especially in combination with valvular heart diseases-by echocardiography can basically be done by averaging the results of echocardiographic measurements of the respective parameters or by the index beat approach, which uses a representative cardiac cycle for measurement. Therefore, a distinction must be made between the functionally relevant status, which is characterized by the averaging method, and the best possible hemodynamic status, which is achieved with the most optimal left ventricular (LV) filling according to the index beat method with longer previous RR intervals. This proposal focuses on left atrial and left ventricular function and deliberately excludes problems of echocardiography when assessing left atrial appendage in terms of its complexity. Echocardiography of the left atrial appendage is therefore reserved for its own expert proposal.

EasyPISA: Automatic Integrated PISA Measurements of Mitral Regurgitation From 2-D Color-Doppler Using Deep Learning

OBJECTIVE

The proximal isovelocity surface area (PISA) method is a well-established approach for mitral regurgitation (MR) quantification. However, it exhibits high inter-observer variability and inaccuracies in cases of non-hemispherical flow convergence and non-holosystolic MR. To address this, we present EasyPISA, a framework for automated integrated PISA measurements taken directly from 2-D color-Doppler sequences.

METHODS

We trained convolutional neural networks (UNet/Attention UNet) on 1171 images from 196 recordings (54 patients) to detect and segment flow convergence zones in 2-D color-Doppler images. Different preprocessing schemes and model architectures were compared. Flow convergence surface areas were estimated, accounting for non-hemispherical convergence, and regurgitant volume (RVol) was computed by integrating the flow rate over time. EasyPISA was retrospectively applied to 26 MR patient examinations, comparing results with reference PISA RVol measurements, severity grades, and cMRI RVol measurements for 13 patients.

RESULTS

The UNet trained on duplex images achieved the best results (precision: 0.63, recall: 0.95, dice: 0.58, flow rate error: 10.4 ml/s). Mitigation of false-positive segmentation on the atrial side of the mitral valve was achieved through integration with a mitral valve segmentation network. The intraclass correlation coefficient was 0.83 between EasyPISA and PISA, and 0.66 between EasyPISA and cMRI. Relative standard deviations were 46% and 53%, respectively. Receiver operator characteristics demonstrated a mean area under the curve between 0.90 and 0.97 for EasyPISA RVol estimates and reference severity grades.

CONCLUSION

EasyPISA demonstrates promising results for fully automated integrated PISA measurements in MR, offering potential benefits in workload reduction and mitigating inter-observer variability in MR assessment.

Mitral Valve Segmentation and Tracking from Transthoracic Echocardiography Using Deep Learning

OBJECTIVE

Valvular heart diseases (VHDs) pose a significant public health burden, and deciding the best treatment strategy necessitates accurate assessment of heart valve function. Transthoracic echocardiography (TTE) is the key modality to evaluate VHDs, but the lack of standardized quantitative measurements leads to subjective and time-consuming assessments. We aimed to use deep learning to automate the extraction of mitral valve (MV) leaflets and annular hinge points from echocardiograms of the MV, improving standardization and reducing workload in quantitative assessment of MV disease.

METHODS

We annotated the MV leaflets and annulus points in 2931 images from 127 patients. We propose an approach for segmenting the annotated features using Attention UNet with deep supervision and weight scheduling of the attention coefficients to enforce saliency surrounding the MV. The derived segmentation masks were used to extract quantitative biomarkers for specific MV leaflet scallops throughout the heart cycle.

RESULTS

Evaluation performance was summarized using a Dice score of 0.63 ± 0.14, annulus error of 3.64 ± 2.53 and leaflet angle error of 8.7 ± 8.3°. Leveraging Attention UNet with deep supervision robustness of clinically relevant metrics was improved compared with UNet, reducing standard deviations by 2.7° (angle error) and 0.73 mm (annulus error). We correctly identified cases of MV prolapse, cases of stenosis and healthy references from a clinical material using the derived biomarkers.

CONCLUSION

Robust deep learning segmentation and tracking of MV morphology and motion is possible by leveraging attention gates and deep supervision, and holds promise for enhancing VHD diagnosis and treatment monitoring.

Expert proposal to analyze the combination of aortic and mitral regurgitation in multiple valvular heart disease by comprehensive echocardiography

The assessment of valvular pathologies in multiple valvular heart disease by echocardiography remains challenging. Data on echocardiographic assessment-especially in patients with combined aortic and mitral regurgitation-are rare in the literature. The proposed integrative approach using semi-quantitative parameters to grade the severity of regurgitation often yields inconsistent findings and results in misinterpretation. Therefore, this proposal aims to focus on a practical systematic echocardiographic analysis to understand the pathophysiology and hemodynamics in patients with combined aortic and mitral regurgitation. The quantitative approach of grading the regurgitant severity of each compound might be helpful in elucidating the scenario in combined aortic and mitral regurgitation. To this end, both the individual regurgitant fraction of each valve and the total regurgitant fraction of both valves must be determined. This work also outlines the methodological issues and limitations of the quantitative approach by echocardiography. Finally, we present a proposal that enables verifiable assessment of regurgitant fractions. The overall interpretation of echocardiographic results includes the symptomatology of patients with combined aortic and mitral regurgitation and the individual treatment options with respect to their individual risk. In summary, a reproducible, verifiable, and transparent in-depth echocardiographic investigation might ensure consistent hemodynamic plausibility of the quantitative results in patients with combined aortic and mitral regurgitation.

Chronic mild to moderate mitral regurgitation will not have an impact on left atrial strain parameters in the pediatric population

BACKGROUND

Left atrial strain (LAS) analysis represents a newer non-invasive, sensitive and specific technique for assessing left atrial (LA) function and early detection of its deformation and dysfunction. However, its applicability in mitral regurgitation (MR) in pediatric population remains unexplored, raising pertinent questions regarding its potential role in evaluating the severity and progression of the disease.

OBJECTIVE

To investigate the impact of chronic MR in children and adolescents on LA remodeling and function.

METHODS

The study included 100 participants. Patients with primary and secondary chronic MR lasting at least 5 years fit our inclusion criteria. The exclusion criteria from the study were: patients with functional mitral regurgitation due to primary cardiomyopathies, patients with artificial mitral valve, patients with MR who had previously undergone surgery due to obstructive lesions of the left heart (aortic stenosis, coarctation of the aorta), patients with significant atrial rhythm disorders (atrial fibrillation, atrial flutter). The echocardiographic recordings were conducted by two different cardiologists. Outcome data was reported as mean and standard deviation (SD) or median and interquartile range (Q1-Q3).

RESULTS

The study included 100 participants, of whom 50 had MR and the remaining 50 were without MR. The average age of all participants was 15.8 ± 1.2 years, with a gender distribution of 37 males and 63 females. There was a significant difference in the values of LA volume index (LAVI), which were higher in patients with MR (p= 0.0001), S/D ratio (and parameters S and D; p= 0.001, p= 0.0001, p= 0.013), mitral annulus radius (p= 0.0001), E/A ratio (p= 0.0001), as well as septal e' (m/s), lateral e' (m/s), and average E/e' ratio, along with the values of TV peak gradient and LV global longitudinal strain (%). There was no significant difference in LA strain parameters, nor in LA stiffness index (LASI).

CONCLUSION

Our findings revealed significant differences in several echocardiographic parameters in pediatric patients with MR relative to those without MR, providing insight into the multifaceted cardiac structural and functional effects of MR in this vulnerable population.

Timing of Surgery for Asymptomatic Primary Mitral Regurgitation: Possible Value of Early, Serial Measurements of Left Ventricular Sphericity

Asymptomatic primary mitral regurgitation due to myxomatous degeneration of the mitral valve leaflets may remain so for long periods, even as left ventricular function progresses to a decompensated stage. During the early compensated stage, the ventricle's initial response to the volume overload is an asymmetric increase in the diastolic short axis dimension, accomplished by a diastolic shift of the interventricular septum into the right ventricular cavity, creating a more spherical left ventricular diastolic shape, increasing diastolic filling and stroke volume. Early valve repair is recommended to reduce postoperative left ventricular dysfunction. Early serial measurements of left ventricular sphericity index [LV-Si]. during the compensated stage of mitral regurgitation might identify subtle changes in left ventricular shape and assist in determining the optimal earliest timing for surgical intervention.

Challenges and pitfalls in classification of disproportionate mitral regurgitation

The concept of disproportionate mitral regurgitation (dispropMR) has been introduced to identify patients with functional mitral regurgitation (MR) who benefit from percutaneous treatment. We aimed to examine echocardiographic characteristics behind this entity. We retrospectively included 172 consecutive patients with reduced left ventricular ejection fraction (LVEF), and more than mild MR referred to clinically indicated echocardiography. According to the proportionality ratio (effective regurgitant orifice area (EROA)/left ventricular end-diastolic volume (LVEDV)) patients were divided into dispropMR and proportionate MR (propMR) group. Potential factors which might affect proportionality definition were analyzed. 55 patients (32%) had dispropMR. Discrepant grading of MR severity was observed when using regurgitant volume (RegVol) by proximal isovelocity surface area (PISA) method or volumetric method, with significant discordance only in dispropMR (p < 0.001). Patients with dispropMR had more frequently left ventricular foreshortened images for LVEDV calculation than patients with propMR (p = 0.003), resulting in smaller LVEDV in dispropMR group. DispropMR group had more substantial dynamic variation of regurgitant flow compared to propMR. Accordingly, EROA was consistently overestimated by standard single-point PISA method compared to serial PISA method. This was more pronounced in dispropMR (bias:10.5 ± 28.3 mm2) compared to propMR group (bias:6.4 ± 12.8 mm2). DispropMR may be found in roughly one third of clinically indicated echocardiographic studies in patients with reduced LVEF and more than mild MR. EROA overestimation due to dynamic variation of regurgitant flow and LVEDV underestimation due to LV foreshortening were more frequently found in dispropMR. Our results indicate that methodological limitations of echocardiographic MR grading could not be neglected in classifying the proportionality of MR.

Contemporary Echocardiographic Evaluation of Mitral Regurgitation and Guidance for Percutaneous Mitral Valve Repair

Mitral valve regurgitation (MR) is a multifaceted valvular heart disease. Echocardiography plays a central role in etiology assessment, severity quantification, treatment candidacy, outcome evaluation, and patient follow-up. In this review, we describe the comprehensive echocardiographic assessment of MR, including transthoracic (TTE) and transesophageal (TEE) approaches, 2D and 3D modalities, strain imaging, stress echocardiography, and artificial intelligence (AI) applications. Transcatheter edge-to-edge mitral valve repair (TEER) has been established as a key therapy for patients with severe, symptomatic MR and high surgical risk. TEER is performed under TEE guidance. We outline a practical overview of echocardiographic guidance on TEER.

Mitral regurgitation in the critically ill: the devil is in the detail

Mitral regurgitation (MR) is common in the critically unwell and encompasses a heterogenous group of conditions with diverging therapeutic strategies. MR may present acutely with haemodynamic instability or more insidiously with failure to wean from mechanical ventilation. Critical illness is associated with marked physiological stress and haemodynamic changes that dynamically influence the severity and implication of MR. The expanding role of critical care echocardiography uniquely positions the intensivist to apply advanced bedside valvular assessment to recognise haemodynanically significant MR, manipulate and optimise cardiopulmonary physiology and identify patients requiring urgent cardiology and surgical referral. This review will consider common clinical scenarios, therapeutic strategies and the pearls and pitfalls of echocardiographic assessment and quantification in the critically unwell.

Valid and Reproducible Quantitative Assessment of Cardiac Volumes by Echocardiography in Patients with Valvular Heart Diseases-Possible or Wishful Thinking?

The analysis of left ventricular function is predominantly based on left ventricular volume assessment. Especially in valvular heart diseases, the quantitative assessment of total and effective stroke volumes as well as regurgitant volumes is necessary for a quantitative approach to determine regurgitant volumes and regurgitant fraction. In the literature, there is an ongoing discussion about differences between cardiac volumes estimated by echocardiography and cardiac magnetic resonance tomography. This viewpoint focuses on the feasibility to assess comparable cardiac volumes with both modalities. The former underestimation of cardiac volumes determined by 2D and 3D echocardiography is presumably explained by methodological and technical limitations. Thus, this viewpoint aims to stimulate an urgent and critical rethinking of the echocardiographic assessment of patients with valvular heart diseases, especially valvular regurgitations, because the actual integrative approach might be too error prone to be continued in this form. It should be replaced or supplemented by a definitive quantitative approach. Valid quantitative assessment by echocardiography is feasible once echocardiography and data analysis are performed with methodological and technical considerations in mind. Unfortunately, implementation of this approach cannot generally be considered for real-world conditions.

Quantification of primary mitral regurgitation by echocardiography: A practical appraisal

The accurate quantification of primary mitral regurgitation (MR) and its consequences on cardiac remodeling is of paramount importance to determine the best timing for surgery in these patients. The recommended echocardiographic grading of primary MR severity relies on an integrated multiparametric approach. It is expected that the large number of echocardiographic parameters collected would offer the possibility to check the measured values regarding their congruence in order to conclude reliably on MR severity. However, the use of multiple parameters to grade MR can result in potential discrepancies between one or more of them. Importantly, many factors beyond MR severity impact the values obtained for these parameters including technical settings, anatomic and hemodynamic considerations, patient's characteristics and echocardiographer' skills. Hence, clinicians involved in valvular diseases should be well aware of the respective strengths and pitfalls of each of MR grading methods by echocardiography. Recent literature highlighted the need for a reappraisal of the severity of primary MR from a hemodynamic perspective. The estimation of MR regurgitation fraction by indirect quantitative methods, whenever possible, should be central when grading the severity of these patients. The assessment of the MR effective regurgitant orifice area by the proximal flow convergence method should be used in a semi-quantitative manner. Furthermore, it is crucial to acknowledge specific clinical situations in MR at risk of misevaluation when grading severity such as late-systolic MR, bi-leaflet prolapse with multiple jets or extensive leak, wall-constrained eccentric jet or in older patients with complex MR mechanism. Finally, it is debatable whether the 4-grades classification of MR severity would be still relevant nowadays, since the indication for mitral valve (MV) surgery is discussed in clinical practice for patients with 3+ and 4+ primary MR based on symptoms, specific markers of adverse outcome and MV repair probability. Primary MR grading should be seen as a continuum integrating both quantification of MR and its consequences, even for patients with presumed "moderate" MR.

[Functional Diagnostics by Rational Echocardiography]

Functional diagnostics by rational echocardiography are the base of a verifiable and reliable analysis of ventricular and valvular function. The most important functional parameters in echocardiography are cardiac volumes - especially total and effective left ventricular stroke volume -, global longitudinal strain, effective orifica areas in valvular stenoses and effective regurgitant volumes, regurgitant fraction and regurgitant orifice areas in valvular regurgitations. Standardized documentation and correct measurements are the prerequisites for accurate echocardiographic estimations. In addition, optimization of image quality supports the correctness of measurements to ensure plausibility of hemodynamics in each individual patient. However, measurements of cardiac volumes by planimetry are error-prone. In addition, calculation of the effective orifice areas using the continuity equation needs methodological accuracy and standardization. Finally, assessment of regurgitant volumes, regurgitant fraction, and regurgitant orifice is often inconsistent with stable hemodynamic conditions - especially by functional assessment the 2D-PISA method. Thus, functional diagnostics by echocardiography in valvular heart diseases should focus on a plausible hemodynamic assessment.

Plausible Functional Diagnostics by Rational Echocardiography in the Assessment of Valvular Heart Disease - Role of Quantitative Echocardiography in the Assessment of Mitral Regurgitation

The echocardiographic assessment of valvular heart diseases is the basic analysis of valvular defects next to clinical investigation and stethoscopy. Severity of mitral regurgitation (MR) is usually estimated by an integrated approach using semi quantitative parameters and is still one of the biggest challenges of echocardiography. Quantitative echocardiographic analysis of MR severity often fails to describe comprehensible hemodynamic conditions. However, comprehensive echocardiography based on standardized image acquisition and proper image quality is required to properly assess hemodynamic parameter comparable to cardiac magnetic resonance tomography. This review focuses on the uncertainty of MR severity assessed by echocardiography in recent trials of interventional MR treatment. In addition, the necessity to provide plausible echocardiographic data for individual decision making is highlighted. In conclusion, plausible functional diagnostics by rational echocardiography is a prerequisite in patients with valvular heart diseases.

Manual zur Indikation und Durchführung spezieller echokardiographischer Anwendungen

Das zweite Manual zur Indikation und Durchführung der Echokardiographie bezieht sich auf spezifische Anwendungen der Echokardiographie und besondere Fragestellungen bei speziellen Patientengruppen. Dabei stehen v. a. praktische Aspekte im Vordergrund. Methodisch etabliert sind die transösophageale Echokardiographie, die Stressechokardiographie und die Kontrastechokardiographie. Bei nahezu allen echokardiographischen Untersuchungen spielen aktuell 3‑D-Echokardiographie und Deformationsbildgebung eine Rolle. Das gesamte Spektrum der echokardiographischen Möglichkeiten wird derzeit in Notfall- und Intensivmedizin, bei der Überwachung und Führung von Katheterinterventionen, bei strukturellen Herzerkrankungen, bei herzchirurgischen Operationen, bei der Nachsorge von kardialen Unterstützungssystemen, bei kongenitalen Vitien im Erwachsenenalter und bei der Versorgung von hochinfektiösen Patienten in Pandemiezeiten angewandt. Die diagnostischen Fortschritte der konventionellen und modernen echokardiographischen Anwendungen stehen im Fokus dieses Manuals. Die 3‑D-Echokardiographie zur Charakterisierung der kardialen Morphologie und die Deformationsbildgebung zur Objektivierung der kardialen Funktion sind bei vielen Indikationen im klinischen Alltag etabliert. Die Stressechokardiographie zur Ischämie‑, Vitalitäts- und Vitiendiagnostik, die Bestimmung der koronaren Flussreserve und die Kontrastechokardiographie bei der linksventrikulären Wandbewegungsanalyse und kardialen Tumordetektion finden zunehmend klinische Anwendung. Wie für die konventionelle Echokardiographie im ersten Manual der Echokardiographie 2009 beschrieben, erfordert der Einsatz moderner echokardiographischer Verfahren die standardisierte Dokumentation und Akquisition bestimmter Bildsequenzen bei optimierter Geräteeinstellung, da korrekte und reproduzierbare Auswertungen nur bei guter Bildqualität möglich sind.