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

The echocardiographic assessment of mitral valve regurgitation (MR) by characterizing specific morphological features and grading its severity is still challenging. Analysis of MR etiology is necessary to clarify the underlying pathological mechanism of the valvular defect. Severity of mitral regurgitation is often quantified based on semi-quantitative parameters. However, incongruent findings and/or interpretations of regurgitation severity are frequently observed. This proposal seeks to offer practical support to overcome these obstacles by offering a standardized workflow, an easy means to identify non-severe mitral regurgitation, and by focusing on the quantitative approach with calculation of the individual regurgitant fraction. This work also indicates main methodological problems of semi-quantitative parameters when evaluating MR severity and offers appropriateness criteria for their use. It addresses the diagnostic importance of left-ventricular wall thickness, left-ventricular and left atrial volumes in relation to disease progression, and disease-related complaints to improve interpretation of echocardiographic findings. Finally, it highlights the conditions influencing the MR dynamics during echocardiographic examination. These considerations allow a reproducible, verifiable, and transparent in-depth echocardiographic evaluation of MR patients ensuring consistent haemodynamic plausibility of echocardiographic results.

Introduction

The most frequently used tool for mitral regurgitation (MR) quantification in clinical practice is “eyeballing” of the colour flow jet area to differentiate between mild and severe MR [1]. This practice is primarily explained by its ease of use. However, it seems inadequate to solely use a qualitative diagnostic parameter to distinguish between mild, moderate, and severe MR [2-5]. As mentioned in recent recommendations, “eyeballing” of the MR jet area is misleading [3, 5, 6]. This is caused by its variations depending on ultrasound settings (Fig. 1), the different display of the jet area in respective sectional planes, and the haemodynamic variations influencing MR dynamics. In consequence, recent papers had eliminated this method in the respective tables [6, 7]. The key point statements—“The colour flow area of regurgitant jet is not recommended to quantify the severity of MR. The colour flow imaging should only be used for diagnosing MR. A more quantitative approach is required when more than a small central MR jet is observed” [3]—emphasize the necessity of a definite quantitative approach for grading MR severity.

Fig. 1

The methodological factors influencing color-coded flow phenomena (PISA, VC, jet area)—illustrated by optimal colour Doppler settings with 1.8 MHz Doppler frequency, increased Doppler sample volume, reduced low-velocity reject, increased frame rate, increased Doppler frequency with 3.1 and 3.6 MHz, increased colour pixel smoothing, reduced colour scale, reduced and increased 2D gain, reduced and increased colour gain, reduced and increased 2D priority, and reduced and increased zero line shift

At the same time, semi-quantitative and/or quantitative parameters, such as the 2D-PISA (proximal isovelocity surface area) method, are used by a minority of primary care physicians and cardiologists [1], whereas in clinical trials, it is the most frequently used method for MR quantification [8-11].

Although recent recommendations describe the numerous limitations of the 2D-PISA method, making its use difficult, one key point message remains, namely “When feasible, the PISA method is highly recommended to quantify the severity of MR” [3]. However, the exact way of measuring of the 2D-PISA radius is unclear, as illustrated in Fig. 2. In the recent guidelines [5] “The radius of PISA is measured from the point of color Doppler aliasing to the VC (vena contracta)”. However, the 2D-PISA radius is illustrated in this recommendation [5], in the first description of the method [12], and in several other references [13] from the proximal convergence area to the ostium of the regurgitant orifice. This discrepancy is not clearly analysed in the literature [5, 13–15]—especially using modern colour Doppler technologies. Also, the impact of MR jet orientation is being debated controversially when using the 2D-PISA method. While some recommendations advise the use of the 2D-PISA in both central and eccentric jets [3, 16], others advise caution [15].

Fig. 2

Limitations of the 2D-PISA method (PISA radius = r)—scheme of the proximal convergence areas and the proximal regurgitant flow phenomenon through the regurgitant orifice illustrating the importance of the accurate definition of the 2D-PISA radius. Example of regurgitant volume (MVRegVol) assessment using different 2D-PISA radii with equal velocity time integrals of regurgitant velocities (r = 8 mm, MVRegVol = 19 ml; r = 11 mm, MVRegVol = 39 ml; r = 15 mm, MVRegVol = 67 ml)

These two mainly used diagnostic features—the colour flow jet area and the 2D-PISA method—are complemented by additional semi-quantitative parameters, which eventuate in the “integrated approach” of MR quantification [3, 5, 6, 16]. However, all these semi-quantitative parameters have their limitations and can only be used in certain circumstances [13, 17–19]. A summary of the strengths and limitations of semi-quantitative parameters for grading of MR severity and the conditions when to apply or not to apply the respective parameters is particularized in Table 1. Considering the methodological challenges of the integrated approach, several concerns of semi-quantitative grading of MR severity should be considered to reduce the inter-observer variability to characterize more precisely and objectively MR severity. The dynamic nature of MR—especially with respect to loading conditions—cause a variability of MR quantification in clinical practice [20-22]. Thus, especially in secondary MR (SMR) recommendations favour the approach of MR assessment at compensated stage [16].

Table 1

Strengths, and limitations of the semi-quantitative parameters for grading MR severity focusing when to use or not to use the respective parameters

Semiquantitative parametera	Strengths	Limitations	When to use or not to use
Valve morphology [3, 5, 23]	Easy to detect by TTE or TEE	Possibility of misinterpretation due to high heart rates	The only entity to imply severe MR is the rupture of a complete papillary muscle
LA and LV size [3, 5, 6]	LA and LV Enlargement are sensitive for chronic relevant MR Normal LV size excludes chronic relevant MR	Reliable results depend on standardization of sectional planes—thus, 3D volume assessment is preferred high inter-observer variability depending on image quality	Only if delineation of endocardial contours is practically possible If necessary, contrast echocardiography is recommended Quantitative assessment of LA and LV size is not reliable performed in foreshortening views and, if limited image quality is present
Vena contracta size [17, 18])	Easy to use, relatively independent of hemodynamic factors	Dependent on ultrasound settings, e.g., smoothing, low-velocity reject, 2D and color gain, etc. error-prone for eccentric jets	Mostly applicable in central jet formations using the parasternal long axis view Not reliable in the presence of eccentric jets especially in primary MR and in the region of the medial or lateral commissure if oblique sectional planes of the vena contracta (not perpendicular to the defect) are documented
2D-PISA-EROA; 2D-PISA-RegVolMV size [3, 5, 6]	Possible to quantitatively assess EROA (lesion severity) and RegVolMV with respects to methodologic accuracy	Underestimation of EROA and RegVolMV by the elliptical shape of EROA Overestimation by improper labeling of the PISA radius PISA elongation by constrained flow field or eccentric jets, and by the dynamic nature of the MR Very limited if applied in eccentric jets—even using angle correction; limited by error-proneness of the PISA radius detection. Thus, high inter-observer variability	Only broadly applicable in central jet formation with flat PISAs (mostly to be observed in SMR Carpentier type 1 in patients with reduced LV function) Not applicable in eccentric PISAs (eccentric jet formation) and in elevated parabolic PISAs (constrained flow patterns) Not usable in multiple MR jets Non-applicable in late systolic dynamic MR (primary MR)
Shape of the EROA by 2D- and 3D-echocardiography size [24–26]	Applicable to detect individual changes of EROA using TEE Semilunar shape of EROA predictable for moderate or severe MR	Difficult to standardize the EROA in just one sectional plane due to its 3-dimensional shape—even using 3D techniques dependent on pixel size and ultrasound settings, not well validated in the literature	Only applicable in TEE; applicable to document individual changes of EROA in relation to hemodynamic factors Applicable to document acute treatment effects during intervention or surgery, diagnostic conclusiveness is limited by TTE color Doppler due to low spatial and temporal resolution Interpretation of EROA shape in TTE is very error-prone
Systolic flow reversal into the pulmonary veins size [27]	Simply to use and—if detectable—specific for severe MR	Dependent of flow direction of the regurgitant jet, on LA size and LA function, on LV contractility, and on hemodynamic factors as well as heart rhythm	Only well applicable, if sinus rhythm is present, if regurgitant jet enters the right pulmonary veins in TTE, and the left pulmonary veins in TEE, if LA size is normal or only mildly enlarged, and if LV contractility is normal or mildly reduced Thus, not applicable in severely enlarged LA, in severe LV dysfunction and during atrial fibrillation
Intensity of the regurgitant velocity signal using continuous wave (cw) Doppler and cw-jet profile size [3, 5, 6]	Easy to document and to interpret A triangular cw-jet profile indicates relevant MR severity	The regurgitant flow velocities should be recorded during the complete systole, that implies correct Doppler delineation during the complete heart cycle	The interpretation and jet profile can only be interpreted if acquisition of the cw-spectrum is methodically correct Thus, this method is only accepted as qualitative parameter due to its methodologic limitations [3, 5] Not applicable, if cw- alignment with blood flow is not verified, especially in eccentric regurgitant jets Proper Doppler alignment is almost always feasible by TTE
Peak mitral E-wave velocity, peak mitral A-wave velocity size [6]	Easy to document by transmitral pulsed wave (pw) Doppler Vmax of E wave < 1 m/s often indicates non-relevant MR Increased A-wave excludes relevant MR	The correct interpretation depends on correct position of pw-sample volume Atrial fibrillation	Applicable, if the orifice area of the mitral valve is normal, if mitral annulus diameter is normal (a.p.-diameter < 35 mm) Limited diagnostic value in atrial fibrillation and severe diastolic dysfunction Not applicable in mitral stenosis Not applicable in severe mitral annulus dilatation
The ratio of transmitral velocity time integral (VTIMV) and flow velocity time integral within the LV outflow tract (VTILVOT) (VTIMV/VTILVOT) size [3, 28]	Easy to determine using pw-Doppler spectra	Diagnostic value depends on the accuracy of the positions of the sample volumes, which should be located at the tip of the MV leaflets and in the LVOT considering optimal alignment of the ultrasound beam with blood flow	Only applicable, if mitral annulus is not severely dilated and normal mitral valve morphology, leak-tight aortic valve and/or and/or atrial fibrillation is present Thus, not applicable in severe mitral annulus dilatation, in mitral stenosis, and in aortic regurgitation Error-prone in atrial fibrillation

Morphological parameters like papillary muscle rupture, LA and LV size, as well as the semi-quantitative parameters vena contracta, 2D-PISA-EROA and 2D-PISA-MVRegVol, the shape of the EROA, systolic flow reversal into the pulmonary veins, intensity of the regurgitant velocity signal using continuous wave Doppler and the jet profile, peak mitral E- and A-wave velocity, and the ratio of transmitral velocity time integral (VTIMV) and flow velocity time integral within the LV outflow tract (VTILVOT) (VTIMV/VTILVOT) are introduced

aThe color Doppler jet area is not listed in the table of semi-quantitative parameters, because this parameter is solely useful for the qualitative detection of a mitral regurgitation, but is not recommended for grading the MR severity (Lancellotti et al. 2010 and 2016). Specific types of jet morphology, e.g., Coanda phenomenon, explain the defect morphology and give cause for quantitative assessment of MR severity

The assessment of MR and the grading of its severity remain challenging today. It is the objective of this work to present tools for an in-depth analysis of the MR, taking practical, methodological, and pathophysiological aspects into consideration. To improve diagnostic conclusiveness the quantitative approach of MR assessment by determining left ventricular (LV) total and effective stroke volume (LVSVtot, LVSVeff), regurgitant volume at the mitral valve level (MVRegVol) and regurgitant fraction (RF) is highlighted.

A proposal for a standardized workflow of the echocardiographic MR assessment

A standardized workflow during the echocardiographic examination and the patient`s visit is necessary to ensure a reproducible and verifiable MR assessment as well as documentation of treatment effects in MR patients. Patient`s characteristics and clinical parameters must be considered for therapeutic decision-making. Indexing of several echocardiographic parameters is based on body height, body weight, and surface area. Systemic blood pressure enables estimation of LV afterload. Clinical symptoms and their progression as well as alterations of echocardiographic parameters with disease progression are important to decide the necessity of therapeutic interventions. At last, age and comorbidities are not influencing MR severity, but are important to estimate the individual patients’ risk. Multiple factors cause differences in MR severity in the same patient at different time points, e.g., cardioversion of atrial fibrillation (AF) into sinus rhythm, resynchronisation therapy in patients with left bundle branch block (LBBB), optimized medical treatment (OMT) in heart failure, or revascularization in myocardial ischemia. To ensure comparability of echocardiographic investigations MR assessment should be performed according to recent recommendations at compensated stage [16].

The echocardiographic examination should consider and interpret the clinical symptoms, and the individual patient`s factors in relation to the presumed valvular defect (Figs. 3, 4). After qualitative MR detection by Doppler techniques, the next diagnostic steps by echocardiography should be the assessment of mitral valve (MV) morphology, LV wall thickness, left atrial (LA), and LV volume as well as LV shape and remodelling, prior to grading MR severity (Figs. 3, 4). Thereafter, a semi-quantitative MR assessment is advised, which should be followed by a quantification of MR severity, if moderate or severe MR is being suspected, or if severity of MR remains unclear (Fig. 4). Every changes of MR severity documented by repetitive standardized echocardiography should be noted to enable reliable conclusions about the respective treatment effects. Figure 5 depicts a recommended timeline for performing echocardiographic examinations in patients with significant MR who are considered for interventional/surgical treatment of MR.

Fig. 3

Scheme to illustrate the diagnostic steps to assess MR by TTE: The first step includes the interpretation of clinical symptoms and the chronicity of the underlying disease in the context of MR with different severity. The second step is the qualitative detection of MR. The third step is the analysis of MV morphology and the differentiation between PMR and SMR. The fourth step is the assessment of LV wall, LA- and LV volumes as well as LV size and LV remodelling to get insights into MR etiology, MR chronicity, and LV geometry. The last step is the grading of MR severity. HOCM hypertrophic cardiomyopathy, LA left atrial, LV left ventricular, LVEDP LV end-diastolic pressure, LVRI LV remodelling index, MR mitral regurgitation, MV mitral valve, PMR primary MR, RWT relative wall thickness, SMR secondary MR, sPAP systolic pulmonary arterial pressure

Fig. 4

Scheme to illustrate the echocardiographic workflow to assess MR severity: After interpretation of symptomatology with respect to the causal relationship to the MR qualitative MR detection results in MR classification due to the MV morphology. Echocardiographic parameters of LA and LV size and LV wall thickness characterize loading conditions and enable to distinguish between pressure or volume overload and between compensated or decompensated conditions. The assessment of MR severity starts with the integrated approach and the analysis of semi-quantitative parameters. The final experts’ task of analysis of MR severity is the quantitative assessment of LVSVtot, LVSVeff, MVRegVol, and RF as a plausibility check. At every level of the assessment of MR severity expert consultation as well as the quantitative analysis of MR severity should be considered with respect to severe symptoms, signs of volume overload and heart failure as well as incongruent results by the grading of MR severity by the semi-quantitative approach. 2D two-dimensional, EROA effective regurgitant orifice area, LA left atrial, LV left ventricular, LVOT LV outflow tract, LVRI LV remodelling index, LVSV effective LV stroke volume, LVSV total LV stroke volume, MR mitral regurgitation, MV mitral valve, MV regurgitant MV volume, PISA proximal isovelocity surface area, PMR primary MR, RF regurgitant fraction, RWT relative wall thickness, SMR secondary MR, VTI velocity time integral

Fig. 5

Proposal for standards of echocardiographic timing in MR patients. The scheme illustrates a potential timeline of echocardiographic investigations during MR treatment. The upper red box presents the therapeutic aspects and strategies, the mid blue box presents the proposed time points of echocardiographic investigations—especially focusing on secondary mitral regurgitation (SMR)-, the bottom green box illustrates the diagnostic targets of the respective echocardiographic investigations. LV left ventricular, MR mitral regurgitation, OMT optimized medical treatment, TOE transoesopageal echocardiography, TTE transthoracic echocardiography

The rationale for the stepwise workflow to assess MR severity to implement the causal relationships between clinical complaints, disease progression, and echocardiographic characteristics into the “integrated approach”

Identifying a causal relationship between clinical symptoms and MR might facilitate the interpretation of echocardiographic results in MR patients. However, symptoms as well as echocardiographic findings depend on chronicity of the disease progress. Acute MR is normally linked to severe symptoms, smaller LA and LV cavities, and severe PH, whereas chronic MR is linked to mild symptoms, larger LA and LV cavities, and different secondary PH severity. Due to this pathophysiological complexity, all possible morphologic variations of LA and LV size can be observed in clinically relevant MR.

If MR is qualitatively detected by Doppler techniques—e.g., colour flow Doppler—MV morphology should help differentiating between primary MR (PMR) and secondary MR (SMR) [29]. This classification focusses on morphological defects of the MV apparatus (PMR) and on secondary MV alterations induced by underlying LV diseases. Thus, structural involvement of the MV apparatus characterizes PMR and LV dilatation and/or LV dysfunction SMR. Pathologies of the leaflets or alterations of the intricate anatomy of the MV apparatus are causes of PMR, failure of MV leaflet coaptation due to MV annulus dilatation, increased leaflet tethering, and/or papillary muscle (PM) restriction are causes of SMR [3–6, 30, 31]. Furthermore, Carpentier’s classification scheme according to leaflets mobility [32] considers functional aspects of the MV leaflets.

The pathophysiological understanding of cardiac alterations in MR requires a morphological characterization of the cardiac cavities [3–6, 16]. Both, PMR and SMR, impose a volume load on the left ventricle and the left atrium. LV dilatation increases MV tethering forces, while LV dysfunction reduces MV closing forces, both driving factors of SMR [33]. SMR resulting from predominant mitral annular dilatation is increasingly being recognized as SMR induced by atrial remodelling [34]. The volume load in chronic PMR and SMR further aggravates LV dilatation to accommodate for the MVRegVol and to maintain LVSVeff. LV function is preserved in the compensated state in PMR, but declines in a decompensated condition. In the decompensated state, MVRegVol itself is a pathophysiological driver that contributes to the disease progress with concomitant increase of LV end-diastolic pressure (LVEDP) and secondary pulmonary hypertension (PH) [35, 36]. LV ejection fraction (LVEF) overestimates LV function in MR. Forward LVEF = LVSVeff/LV end-diastolic volume (LVEDV) seems to represent more reliably LV function than global LVEF in MR [37, 38]. Hence, MR severity relative to LV remodeling has been proposed [16, 39, 40]. In consequence, LV function, LV remodelling, and global haemodynamics often differ between PMR and SMR. Thus, LV wall thickness, LV mass, LV mass index, LV diameter, LV volume, LVEF, as well as LA volume and LA volume should be measured by echocardiography to characterize LV geometry, e.g., concentric remodelling, and concentric and eccentric LV hypertrophy [41]. Relative wall thickness (RWT) and LV mass should be measured using the posterior wall. Considering clinical symptoms and chronicity of the underlying diseases in relation to the specific echocardiographic findings, an extended MR classification is proposed for PMR and SMR patients. Five subtypes can be differentiated in PMR (Table 2). Furthermore, seven subtypes can be differentiated in SMR (Table 3) with respect to symmetric LV remodelling, asynchrony of LV contraction, regional myocardial injury, asymmetric LV hypertrophy, LV stiffening, and LA remodelling [3, 5].

Table 2

Proposal to classify primary mitral regurgitation (PMR) more in detail with respect to specific echocardiographic findings, the chronicity of the underlying diseases, and the clinical complaints of the patients

PMR subtypes are characterized by description of left-ventricular (LV) size, LV wall thickness, left atrial (LA) size, the course of the disease, and one respective echocardiographic example

MV mitral valve, sPAP systolic pulmonary arterial pressure

Table 3

Proposal to classify secondary mitral regurgitation (SMR) more in detail with respect to specific echocardiographic findings, the chronicity of the underlying diseases, and the clinical complaints of the patients

PMR subtypes are characterized by description of left ventricular (LV) size, LV wall thickness, left atrial (LA) size, the course of the disease and one respective echocardiographic example

AF atrial fibrillation, AM acute myocarditis, AS aortic valve stenosis; CHF chronic heart failure, DCM dilated cardiomyopathy, HCM hypertrophic cardiomyopathy, HHD hypertensive heart disease, ICM inflammatory cardiomyopathy, IHD ischemic heart disease, LBBB left bundle branch block, LVH left-ventricular hypertrophy, LVOT left-ventricular outflow tract, MV mitral valve, NCCM non-compaction cardiomyopathy, SAM “systolic anterior movement”, TAI-CM tachyarrhythmia-induced cardiomyopathy

Thus, one target of paramount importance is to characterize cardiac remodelling due to MR effects, which implies the specific assessment of LV [43-45] and LA geometry by echocardiography [46]—especially in SMR patients [3, 5, 6]. Despite recent technical improvements in echocardiography and automated features to analyze LA and LV volumes and function, conventional 2D echocardiography remains the current standard and enables the assessment of relevant cardiac parameters as illustrated in Table 4. Linear internal 2D measurements of LV diameters and LV wall thickness as well as LV volume measurements by 2D planimetry are still used in clinical practice [5, 41, 47]—especially for calculation of LV mass [41]. 3D approaches for LV mass determination are preferably recommended [48]. The sphericity ratio and sphericity index, interpapillary muscle distance, the anterior–posterior and medial–lateral PM displacement, and the length between the PM bulges and the respective contralateral MV annulus should be considered for characterization of LV remodelling [3, 49–51]. Furthermore, LV remodelling with disease progression or reverse LV remodeling during treatment can be assessed by monitoring LV geometry [3, 51]. LA volume measurement by 2D planimetry of the maximum LA area in the 2- and 4-chamber view (2-ChV, 4-ChV) or using 3D echocardiography is preferred [48]. The progression of LA and LV volumes and reduction of LVEF during follow-up examinations are helpful to determine haemodynamically significant deterioration even in MR patients classified as clinically not severe.

Table 4

Echocardiographic parameters characterizing left-ventricular (LV) remodelling in MR patients using conventional 2D echocardiography or 3D TTE

In the first column, the echocardiographic target parameters are listed including the normal ranges: LV wall thickness and relative wall thickness (RWT), LV diameter, LV mass (LVM), sphericity ratio, sphericity index, interpapillary muscle distance (IPMD), anterior–posterior and medial–lateral displacement of the papillary muscles (PM), as well as the length between the bulges of the posterolateral or anterolateral PM and the respective contralateral MV annulus (MA). The parameters recommended as mandatory [3, 5] are marked with ●. The respective images illustrate the assessment of the respective parameters in 2D sectional planes or within 3D data sets

The second target is the analysis of MV morphology by echocardiography. Due to the complexity of the MV apparatus, 3D image acquisition has become an indispensable tool of echocardiographic MV assessment [48, 52–55]. However, conventional 2D echocardiography enables the measurement of specific parameters characterizing pathologies of MV morphology. MV degeneration can be identified by the presence of intensified echo-densities due to thickening and calcification of the MV annulus [42]. MV prolapse is characterized by systolic displacement of a leaflet by ≥ 2 mm overriding the annular plane into the LA [3, 52]. Rupture of the primary chordae, or ultimately of a PM, causes flail of the leaflet into the LA and is usually associated with severe MR. Analysis of MV involvement in endocarditis should include size of vegetations, presence of abscesses, aneurysms, or perforations [56]. Congenital MV defects, e.g., clefts, can be uncovered in the 3D TOE, favouring definitively this modern technology [55]. MV deformation due to LV remodelling in SMR should be assessed by measurement of MV annulus, coaptation distance/gap, coaptation length/height, as well as coaptation depth and tenting height (Table 5) [3, 5, 57]. The tenting area (area between the MA and the leaflets during systole) of ≥ 2.1 cm2 is a pathologic finding due to tethering in SMR [3-5]. MV analysis in SMR should be completed by the assessment of the anterior/medial and posterior/posterolateral tethering angle (Table 5) [3, 5]. As pathophysiology of MR is a constant and complex interplay between initial pathology and further propagation of the disease by volume overload, coexistence between PMR and SMR can be observed and should be labeled as mixed origin.

Table 5

Echocardiographic parameters characterizing mitral valve (MV) deformation in SMR patients using conventional 2D echocardiography or 3D TTE

In the first column, the echocardiographic target parameters are listed including the normal ranges: anterior–posterior and medial–lateral MV annulus diameter (intercommissural diameter) at early diastole, coaptation distance or gap, coaptation length or height, effective height, tenting height or tenting distance, the “seagull” sign, MV annulus diameter at end-systole, posterolateral and medial tethering angle, and mitral valve orifice area. The parameters recommended as mandatory [3, 5] are marked with ●. The respective images illustrate the assessment of the respective parameters in 2D sectional planes or within 3D data sets

The rationale to implement a quantitative MR assessment to characterize MR severity

The echocardiographic workflow of grading MR severity (Figs. 3, 4) starts with a semi-quantitative MR assessment and serves two goals. First, all non-severe MR should be detected, preventing unnecessary and time-consuming further evaluation. For example, when sinus rhythm is present, an a-wave dominant inflow pattern into the LV using Doppler interrogation above the MV excludes severe MR. Also, a dominant inflow during systole from the pulmonary veins into the LA cannot be observed in severe MR. Finally, a normal LA volume is not found in chronic severe MR. These and other semi-quantitative parameters, along with their strengths, limitations, and appropriateness are listed in Table 1. In-depth quantitative evaluation should be initiated in cases if MR classification remains unclear.

The quantitative approach is based on the determination of the individual RF. This parameter is included in all current recommendations [3, 5, 6]. RF relies on the determination of LVSVtot and LVSVeff. The absolute value of MVRegVol should always be interpreted with respect to LVEDV. It is obvious that the amount of MVRegVol is much more important in small hearts than in larger hearts, which can be impressively illustrated by interspecies comparisons (Fig. 6). In consequence, haemodynamic conditions can be characterized by plausible LVEDV, LVEF, and LV forward stroke volume (= LVSVeff). Determination of MVRegVol by the 2D-PISA method alone was associated with significant overestimation of MRRegVol as documented in recent transcatheter MV repair (TMVR) trials [10, 58, 59] and further MR outcome trials [60].

Fig. 6

Illustration of the interspecies differences of regurgitant volume in relation to total stroke volume (LVSVtot). The normal LVSVtot of a rat heart is about 0.5 ml [61] resulting in a regurgitant fraction (RF) of 50% if regurgitant volume at the mitral valve (MVRegVol) is about 0.25 ml. The normal LVSVtot of an elephant heart is about 20 l [62] resulting in a RF of about zero, if MVRegVol is about 0.25 ml. An RF of about 50% needs an MVRegVol of about 5 l

Calculation of RF is based on the measurement of LVEDV and LV endsystolic volume (LVESV) as well as LVSVeff and MVRegVol to estimate cardiac output (CO) and cardiac index (CI) by echocardiography. Practical tips to avoid pitfalls when determining cardiac volumes—especially LVSVtot, LVSVeff, and right-ventricular (RV) stroke volume (RVSVeff), are listed in Table 6. The practical approach to check Doppler measurements of RVSVeff by a plausibility cross-check is illustrated in Fig. 7. However, this concept is still not validated by prognostic data [2, 3, 5]. Compared to cardiac magnetic resonance (CMR) tomography, a significant underestimation of LV volumes by echocardiography has been reported [63]. Furthermore, over- and underestimation of LV volumes in humans [64] and phantoms [65, 66] have been described comparing different imaging methods, e.g., native 2D- and 3D echocardiography, contrast echocardiography, CMR, and computed tomography. Recently, conclusive LV volume assessment by 2D echocardiography was illustrated if image quality is adequate [67-70]. The differences in LV volumes between 2D echocardiography and CMR can be minimized by triplane, 3D-, and contrast echocardiography [71, 72]. A Doppler echocardiographic approach to calculate LVSVtot by the LV filling volume has been proposed in recent recommendations using MV diameter in the 4-chamber view and the transmitral velocity time integral (VTI) at the level of the mitral annulus [2, 3, 5]. However, this approach seems to be error-prone due to the non-circular shape of the MV annulus.

Table 6

Target parameters of left-ventricular (LV) volumes and mitral regurgitant volume (MVRegVol), the different methods for assessment, the methodological limitations, and the conditions when to use or not to use the respective method

Target parameter	Methods	Limitations	When to use or not to use
LVSVtot	LV planimetry (2D) Monoplane long axis view (LAX) Biplane 2- and 4- chamber view (2- and 4-ChV) Triplane LV volumetry (3D) Mitral inflow (Doppler)	LV planimetry (2D) not-sufficient standardization of the views not-sufficient imaging conditions of endocardial contours foreshortening views regional wall motion abnormalities LV volumetry (3D) not-sufficient image quality, especially spatial resolution Mitral inflow (Doppler) Mitral annulus is not circular Transmitral pw-Doppler spectrum must be acquired at mitral annulus level Position of the sample volume cannot be standardized due to the movement of the mitral annulus	LV planimetry (2D)— in general, only to use if endocardial contours can be adequately delineated. If not, try to use LV opacification with contrast echocardiography. Delineation of all trabecula as endocardium causes underestimation, delineation of the midmyocardial contour between longitudinal and circumferential fibers causes overestimation of LV volumes. Carefully labeling of the apex of the cavity, the mitral annulus and the LVOT—especially wrong labeling of the basal regions produces significant underestimation of LV volumes Monoplane LV planimetry is only applicable if no wall motion abnormalities are present. Monoplane LAX planimetry results mostly in larger LV volumes in comparison to 2- and 4-ChV. Monoplane LV planimetry is misleading in patients with regional wall motion abnormalities Biplane 2- and 4-ChV is not allowed in foreshortening and not-standardized views. Thus, it is only applicable if maximum LV length is accurately documented. Monoplane 2-ChV planimetry results mostly in the lowest LV volumes, monoplane 4-ChV planimetry results mostly in the underestimated LV volumes due to foreshortening. Biplane LV planimetry is misleading in patients with regional wall motion abnormalities Triplane is the best approach to document standardized views. Triplane LV planimetry is an acceptable approach to assess reliable LV volumes in patients with regional wall motion abnormalities. Triplane LV planimetry is superior to LV volumetry (3D) in patients with not optimal image quality LV volumetry (3D)—This approach is the best one—especially in patients with regional wall motion abnormalities. However, it can only be used in patients with excellent image quality and sufficient temporal resolution (volume rates > 20/s). If volume stitching is needed, image acquisition requires regular heart rate and cooperation of the patient during breath hold Mitral inflow (Doppler)—in clinical practice this method is too error-prone to be recommended because diameter of the mitral annulus is not exactly determined in the 4-ChV and cannot be corrected with respect to the dynamic alterations during diastole. Transmitral pw-Doppler spectrum at the level of the mitral annulus must be aligned to the inflow velocities. This approach is generally obsolete in patients with mitral valve stenosis or pathologically increased transmitral velocities
LVSVeff	Doppler calculation using LVOT diameter (DLVOT) and LVOT velocity time integral (VTILVOT): LVSVeff = 0.785 × DLVOT2 x VTILVOT	Oblique labeling of DLVOT mostly causing underestimation of DLVOT and LVSVeff Wrong position of the position of the sample volume. If it is located too far into the left ventricle, LVSVeff is underestimated	LVSVeff assessment by Doppler echocardiography is well applicable in patients with normal morphology of aortic valve and LVOT LVSVeff assessment by Doppler echocardiography is not applicable in patients with relevant aortic stenosis (overestimation of LVSVeff due to increased VTILVOT because of flow increase proximal to the aortic valve stenosis) and/or relevant aortic valve regurgitation (overestimation of LVSVeff due to increased VTILVOT which represent the addition of LVSVeff and regurgitant volume at the aortic valve) If DLVOT cannot be accurately measured in TTE, DLVOT or cross-sectional LVOT area can be determined by 2D- or 3D-TOE imaging
RVSVeff	Doppler calculation using RVOT diameter (DRVOT) and RVOT velocity time integral (VTIRVOT): RVSVeff = 0.785 × DRVOT2 x VTIRVOT	Wrong labeling of DRVOT mostly caused by lung shadowing causing underestimation of DRVOT and RVSVeff Wrong labeling of DRVOT too far into the right ventricle causing severe overestimation of DRVOT and RVSVeff Wrong position of the position of the sample volume in relation to the labeling of DRVOT. Causing both over- or underestimation of RVSVeff	RVSVeff assessment by Doppler echocardiography is well applicable in patients with normal morphology of pulmonary valve and RVOT. Plausibility control assessment is recommended comparing measurements at different levels at the RVOT, the pulmonary valve and the pulmonary trunk (see Fig. 4) RVSVeff assessment by Doppler echocardiography is not applicable in patients with relevant pulmonic stenosis or regurgitation In patients with aortic valve disease, RVSVeff assessment by Doppler echocardiography (if pulmonary valve is normal and no or mild regurgitation is present) enables the estimation of LVSVeff because during these conditions RVSVeff is equal to LVSVeff If DRVOT cannot be accurately measured in TTE, DRVOT or cross-sectional RVOT area can be determined by 2D- or 3D-TOE imaging
2D-PISA-MRRegVol	2D-PISA-method	underestimation of RegVolMV by the elliptical shape of EROA overestimation by improper labeling of the PISA radius, PISA elongation by constrained flow field or eccentric jets, and by the dynamic nature of the MR; very limited, if applied in eccentric jets—even using angle correction; limited by error-proneness of the PISA radius detection	MRRegVol by the 2D-PISA method is only applicable in patients with mitral regurgitation if regurgitant jet formation is not eccentric and proximal convergence areas are flat, e.g. in patients with mitral valve regurgitation type Carpentier I with reduced LV function Highly error-prone in primary MR with eccentric jet formation Not applicable in the presence of relevant mitral valve stenosis Not applicable in the presence of concomitant aortic valve diseases
Calculated MRRegVol	Calculation using LVSVtot assessment by planimetry or volumetry and LVSVeff by Doppler echocardiography: MRRegVol = LVSVtot—LVSVeff	In principle, error-prone due to the assessment of multiple parameters for both, LVSVtot and LVSVeff determination The validity of this approach is highly dependent on image quality, standardization, technical skill, and expertise	LVSVtot assessment can only be performed in native 2D echocardiography if image quality is adequate. Otherwise contrast echocardiography is recommended The choice of method for LVSVtot assessment depends on alterations of LV geometry due to regional wall motion abnormalities. If image quality is adequate, 3D volumetry is superior to triplane. Triplane LV planimetry is superior to biplane. Biplane LV planimetry is superior to monoplane LVSVeff assessment requires the correct position of the sample volumes of pw Doppler and the correct allocation of the respective diameters of LVOT and RVOT to the positions of the sample volume. Alternatively, diameters and cross-sectional areas can be determined by 2D- and 3D-TOE data sets

LVSV total LV stroke volume, LVSV effective forward LV stroke volume, RVSV effective forward RV stroke volume, LVOT—LV outflow tract, RVOT right-ventricular outflow tract, TOE transoesophageal echocardiography, TTE transthoracic echocardiography

Fig. 7

Illustration of practical aspects of LVSVeff or RVSVeff determination. Labeling of the DLVOT and correct positioning of the pw-sample volume documented by the cusp artefact in the pw-Doppler spectrum with the respective results (a); three-point labeling of diameters at the level of the pulmonic valve (1) documented by the origin of the pulmonary regurgitation, at the level of the proximal pulmonic trunk (2) and at the level of the distal RVOT (3) for the respective position of the pw-Doppler sample volume (b); labeling of the DRVOT and the corresponding pw-Doppler spectrum at the RVOT (c), at the pulmonic valve (d), and at the proximal pulmonic trunk (e) with the respective results. All determined forward stroke volumes are within similar ranges, hence documenting plausible results

LVSVeff in “pure” MR can be determined by Doppler calculations using cross-sectional area (CSA) or diameters of the LV outflow tract (LVOT) and the corresponding pulsed wave (pw) Doppler velocity time integral (VTI) [2, 3, 5]. In patients with combined aortic valve disease, LVSVeff assessment is more complex, because Doppler calculations of LVSVeff should be performed using the CSA or diameter of the RV outflow tract (RVOT) and the respective pw-Doppler VTI to assess RVSVeff, which corresponds to LVSVeff, if no or only mild pulmonary regurgitation is present. However, RVSVeff measurement is challenging due to the variable anatomy of the RVOT and the additional time needed for precise measurements.

The problem of incongruent haemodynamic measurements in MR patients is highlighted by the recently introduced terms “proportionate” and “disproportionate” MR [40, 73–75]. The concept of proportionality between blood flow and orifice areas can be illustrated by the continuity equation determining effective orifice area in patients with aortic valve stenosis (AS) [36, 76]. The same principle of proportionality can only theoretically be applied to the calculation of the MVRegVol (Fig. 8), because MVRegVol cannot be practically measured by pw-Doppler techniques due to methodological limitations. However, a plausibility cross-check of LVSVtot, LVSVeff, MVRegVol,CO, and CI can be performed independently of the method used for determination of these parameters, because proportionality is a prerequisite between EROA and MVRegVol. The usage of the continuity equation for MVRegVol determination is impossible due to the high transmitral velocities of regurgitant flow at the level of the mitral annulus, the EROA changes of the valve during the systolic time interval, and the deceleration of flow velocities between EROA and the mitral annulus level. Because of the rheological need of proportionality between EROA and retrograde volume flow or flow velocities, the term “disproportionateness” [40, 73–75] can only be interpreted as a characterization of SMR severity in relation to the impaired LV function. However, the potential therapeutical benefit of MR treatment in relation to heart failure cannot be described by the disproportionality between LVEDV and EROA, because these parameters are proportionally interrelated at a defined LVEF (Fig. 9).

Fig. 8

Illustration of the proportionality of forward blood flow volume or effective left-ventricular stroke volume (LVSVeff) and of transmitral regurgitant volume (MVRegVol) between the respective cross-section areas (CSAs) and blood flow velocities in a system of communicating tubes. Considering the volume flow during one heart cycle total left-ventricular stroke volume (LVSVtot) is the summation of LVSVeff and MVRegVol. LVSVeff at the level of the left-ventricular outflow tract (LVOT) is equal to the level of the aortic valve (AV) orifice according to the continuity equation. By analogy MVRegVol at the level of the effective regurgitant orifice area (EROA) is equal to MVRegVol at the level of mitral valve (MV) annulus. Thus, both LVSVeff and MVRegVol exhibit proportionality between respective cross-section areas (CSA) and velocity time integrals (VTI). CSA CSA of the AV orifice, CSA CSA of the MV regurgitant orifice, CSA CSA of the LVOT, CSA CSA at the level of the MV annulus, D diameter of the AV orifice, D diameter of the MV regurgitant orifice, D diameter of the LVOT, D  diameter at the level of the MV annulus, VTI VTI of the systolic forward blood flow through the AV orifice, VTI VTI of the diastolic backward blood flow through the MV regurgitant orifice, VTI VTI of the systolic forward blood flow through the LVOT, VTI VTI of the diastolic forward mitral flow at the level of the MV annulus, VTI-MV VTI of the systolic regurgitant transmitral blood flow at the level of the MV annulus

Fig. 9

The relation between LVSVtot, which is equal to LVSVeff in the absence of mitral regurgitation (MR) and aortic regurgitation (AR), and left-ventricular end-diastolic volume (LVEDV) with respect to left ejection fraction (LVEF) If LVEDV of 200 ml in the presence of LVEF of 30% is assumed at stable haemodynamic conditions labeled by the blue area ( LVSVtot = LVSVeff, = 60 ml indicating a cardiac index > 2.2 l/min m2 at a normal heart rate of 65/min), LVSVtot must be equal to LVSVeff, indicating the absence of MR and AR to provide the necessary cardiac output or cardiac index. The red arrows display the necessary increase of LVEDV or LVEF assuming severe MR with a regurgitant fraction of 50%. Thus, to provide LVSVeff of 60 ml and MVRegVol of 60 ml, LVSVtot of 120 ml is necessary. Consequently, LVEDV must be 400 ml if LVEF is 30%, and LVEF must be 60% if LVEDV is 200 ml

MR severity can be assessed as mild or moderate in heart failure patients at rest during compensated stage with OMT. However, this MR characterization at rest might not describe the individual risk of re-decompensation. Thus, in these cases, haemodynamic impairment should predominantly be documented by increase in SMR severity during mild-to-moderate dynamic stress testing to support this hypothesis [77, 78]. Early treatment of SMR is comprehensible during these conditions because of the potential for reverse LV remodeling and prevention of further deterioration of LV function, which should be documented by prospective trials.

Summary and conclusion

The analysis of MR severity has become more and more important with respect to therapeutic options for MR treatment. The grading of MR severity by “eyeballing” and the 2D-PISA method is common in clinical practice, but it often leads to incongruent results with a high inter-observer variability. In addition, the dynamics of MR due to volume conditions, heart rhythm, and respective medical treatment require a high level of standardization in echocardiography. However, echocardiography allows for reliable assessment of LVSVtot, LVSVeff, MVRegVol, and RF in MR patients. Prerequisites are verifiable documentations, respective technical skill, and plausible measurements. The present proposal provides a “new” haemodynamically oriented workflow, which integrates a detailed MR classification scheme, considering the clinical complaints, the chronicity of the disease process, the MV morphology, and the echocardiographic parameters characterizing LA and LV remodelling. The essential point to note is the integration of a quantitative assessment of MR severity into the recent “integrated approach” to provide haemodynamic plausibility and to avoid inconsistencies of echocardiographic measurements.