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

The causes of AR and MR with their underlying mechanism are described according to the Carpentier’s classification of leaflet motion: Type I: normal leaflet motion, Type II: excessive motion, and Type III: restrictive motion [11]. Chronic isolated AR results in reverse transvalvular diastolic blood flow into the left ventricle due to primary (organic) damage of the cusps or secondary (functional) damage resulting from to dilatation of the aortic root complex causing a combined volume and pressure LV overload. Chronic AR usually evolves slowly and is well compensated in early stages, often documented by the presence of asymptomatic severe AR in physically fit patients [12]. Dyspnea upon exercise can usually be observed in early stages, whereas overt symptoms of heart failure like congestion, weakness, or arrhythmias occur in the later stages of the disease. The late appearance of symptoms in isolated AR due to effective compensatory mechanisms is explained mainly by the fact that both the left ventricle and the aortic root form part of the high-pressure system, separating the low-pressure system by an intact mitral valve (MV) protecting against damage to the left atrium, pulmonary vascular system, and right heart.. The volume overload in chronic AR results in progressive LV remodeling to normalize wall stress and maintain systolic function and is characterized by eccentric LV hypertrophy, LV dilatation, and LV spherification. Repetitive ischemic episodes caused by the ensuing increased LV end-diastolic pressure (LVEDP) are thought to promote myocardial fibrosis as the underlying mechanism for reduced LV compliance and diastolic dysfunction. LV dilatation and the reduction of LV ejection fraction (LVEF) as well as left atrial (LA) enlargement due to increased LV filling pressures in the later course of the disease [13‐15] are prognostically unfavorable factors in chronic AR.

Chronic isolated MR results in reverse transvalvular systolic blood flow into the left atrium due to primary (organic) structural abnormalities of the leaflets and the MV apparatus or secondary (functional) damage resulting from eccentric LV hypertrophy, LV dilatation, annulus dilatation, and an imbalance between tethering and closing forces caused by pathological enlargement or geometric changes of the left ventricle or the left atrium [16‐18]. Chronic MR usually develops slowly, but symptoms such as shortness of breath, edema, or palpitations appear earlier than in chronic AR.

The earlier onset of symptoms in isolated MR can be explained by the fact that the MV represents the boundary to the low-pressure system, and thus pressure elevation and signs of congestion in the pulmonary circulation are more likely to occur. In isolated MR, a considerable proportion of the total LV stroke volume (LVSV_tot) empties into the low-impedance left atrium as mitral regurgitant volume (RegVol_MR). To maintain the effective LV stroke volume (LVSV_eff), which corresponds to the forward LV stroke volume (LVSV_forward) in the absence of AR, both LV diastolic volume and LVSV_tot increase. During this early compensation, LVEF is usually in the normal-to-high range. This volume overload may lead to LV dilatation with eccentric LV hypertrophy and a change in the LV shape towards a more spherical LV cavity. Progressive LV remodeling may worsen or lead to secondary (functional) MR due to an imbalance between increased tethering forces and decreased closing forces during systole [19]. In the compensated stage, eccentric LV hypertrophy maintains a normal diastolic pressure with increase of wall stress due to predominant LV dilatation and with decrease of wall stress due to predominant LV wall thickening. LA enlargement is often a consequence of MR and may be associated with mitral annular enlargement inducing progression of secondary MR [20].

The combination of AR and MR may lead to LA and LV volume overload, which can result in atrial fibrillation, pulmonary hypertension, right ventricular (RV) enlargement, RV dysfunction and secondary tricuspid regurgitation in the low-pressure system. Importantly, RV dysfunction is a prognostic factor for postoperative mortality in patients with combined AR and MR [21, 22]. Currently, these coexistent pathophysiological consequences are given little consideration in the current guidelines [20]. The interdependency of AR and MR is based on the physical properties of incompressible fluid within the cardiac cavities. The cardiac chambers, therefore, serve both as a reservoir during filling and as a propagation pump during muscular contraction. When both forms of regurgitation are present, the reservoir function becomes the Achilles’ heel of LV remodeling due to LV volume overload caused by both relevant AR and MR. AR progression increases the forward flow through the AV as determined by Doppler echocardiography (LVSV_forward), calculated as the sum of LVSV_eff and the transaortic regurgitant volume (RegVol_AR). In the presence of AR and MR, eccentric LV hypertrophy serves as a compensatory mechanism to maintain an effective cardiac output. However, with LV dilatation and an almost unchanged LV wall thickness, LV wall stress increases inducing LV dysfunction.

In both AR and MR as a singular valvular lesion, the LV volume load increases. In contrast to MR, AR additionally increases LV afterload and thus causes an additional LV pressure load. Historical data show that in the presence of severe MR, mild-to-moderate AR may be well tolerated, but when AR is severe, any degree of MR may substantially worsen LV dilatation and LV dysfunction [23]. The safeguarding mechanism of early MV closure restricting the quantity of backward flow into the left atrium and pulmonary circulation in severe AR is not present in patients with combined AR and MR. This plays an important role in clinical worsening in comparison with isolated AR [6, 24, 25]. Postoperative data showed that LV dysfunction is more likely to occur in combined AR and MR than in isolated AR [26]. Another retrospective single-center study of 756 patients with at least moderate AR showed moderate to severe MR in 45%. Presence of moderate to severe and severe MR was associated with a larger LV size, lower LVEF, atrial fibrillation, as well as older age, female sex, and further comorbidities. Survival was increased if MR was also treated at the time of aortic valve (AV) replacement, and best if MV repair was feasible [27]. However, patients with combined AR and MR had a worse postoperative survival compared to patients with single-valve disease [6, 24]. Data on the diagnosis of combined AR and MR are limited in the literature [8, 11]. LVEF is recognized as a suitable variable to monitor LV function in VHD and can be used in combination with biomarkers like NT-proBNP to monitor potential impairment. Since LVEF is highly dependent on LV loading conditions, it has a limited ability to characterize abnormalities of myocardial contractility at early stages of severe combined MR and AR. Thus, LV deformation—especially global longitudinal strain—seems to be a more sensitive indicator of incipient LV dysfunction than LVEF [28]. For any given level of LV end-systolic volume, LV dysfunction is discussed as a prognostic marker in MR and AR [21, 22]. Severe LV dilatation may occur even in the combination of moderate AR plus moderate MR [5]. Consequently, the coexistence of significant AR and MR intensifies the negative impact on LV function and is associated with a worse prognosis compared with a single valvular lesion [3, 5, 24]. Moreover, the combination of non-severe AR and MR may lead to a clinically significant severe hemodynamic burden [22, 29].

For combined AR and MR, it often needs to be determined whether both valvular lesions or only one lesion are responsible for the pathological LV and RV changes [3, 5, 6, 20]. In a large cohort of 1239 patients with at least moderate AR, the incidence of at least moderate functional MR was 9%, and of primary MR 5%. Functional MR was associated with larger LV volumes and lower LVEF. The long-term mortality of AR patients was increased by concomitant MR—more so by functional MR than by primary MR [6]. In patients with moderate or severe AR, at least moderate functional MR was documented in 23%. Lower LVEF and a larger LA, as well as more MV tenting and larger interpapillary muscle distances were more frequently associated with MR [30].

The clinical relevance of combined AR and MR is usually obvious if one or both defects are moderate to severe. However, decision-making is rendered difficult if both defects are rated as mild to moderate, and symptoms of heart failure exist with no other obvious cause. Since cardiomyopathy of other causes is possible in the presence of LV dilatation, the differentiation from other non-valvular causes of heart failure is important [31]. AR contributes to delayed MV opening causing a prolonged isovolumetric relaxation time with LV filling due to AR prior to diastolic forward flow through the MV. Thus, LV filling pressure rapidly increases as a result of simultaneous LV filling due to AR and through the MV. The diastolic LA pressure is the driving force of LV filling. The effective regurgitant orifice area (EROA) of the AR serves as flow resistance of the diastolic pressure at the level of the tubular ascending aorta, which attenuates but does not abolish the diastolic driving forces of forward LV filling. Consequently, shortening of the pulmonary acceleration time (< 100 ms) and an increase in systolic pulmonary artery (PA) pressure occur in early stages of MR. In addition, shortening of transmitral E-wave acceleration and deceleration, and velocity reduction of the A-wave are signs or ‘red flags’ of relevant combined AR and MR. The restrictive transmitral LV filling pattern is in accordance with indirect evidence of reduced LV compliance or atrial cardiomyopathy in LA and/or LV dilatation.

LV remodeling due to chronic AR is characterized by eccentric LV hypertrophy and LV dilatation attributable to chronic AV damage—for example, in case of cusp prolapse of a bicuspid AV—or by aortic annulus dilatation in pathologies of the aortic root complex [32, 33]. In patients with hemodynamically significant AR, functional MR due to LV remodeling is observed in approximately 7% and is considered a more advanced stage in the natural course of the disease. In AR, wall stress was found to be markedly elevated due to a markedly increased afterload, whereas in MR wall stress reached only near-normal levels [34]. For valve regurgitation of similar severities, AR results in greater LV dilatation to the point of irreversible myocardial dysfunction compared with MR [35]. Acute worsening of combined chronic AR and MR may occur as a result of reaching the compensation limits for regurgitant volume at both valves. Examples of additional acute components of valve destruction are acute valve infection (endocarditis) or acute ischemia (myocardial infarction—especially due to occlusion of the circumflex and marginal branches—causing partial or complete rupture of papillary muscles). Chronification of high-grade secondary MR is unlikely, since chronic symptoms are expected to occur in early stages of MR. Thus, MR developing secondary to chronic severe AR is a unique subtype of combined AR and MR. Its prevalence has been reported to be between 6 and 45% and its occurrence has been associated with chronic changes in the size, shape, and function of the LV [36]. However, despite significant increases in LV dimensions commonly assumed to be associated with secondary MR, such as LV volume, LV sphericity, tethering distance and mitral annular size, severe secondary MR may be rather rare in chronic severe AR [37]. This seems to be due mainly to the ability of the MV to increase its leaflet area and thickness, thereby counterbalancing the consequences of chronic AR [37]. This enlargement of leaflet area is thought to protect against MR and seems to be lacking or blunted in functional MR due to LV dilatation and LV dysfunction. These data were recently corroborated by a study in sheep where a serotonin inhibitor nearly abrogated the development of functional MR by intensifying mitral leaflet growth after induction of myocardial infarction [38].

Albeit the LV volume load increases due to MR, relevant AR secondary to severe MR seems unlikely. LV remodeling due to chronic MR can, in theory, cause AR due to aortic annulus dilatation. However, dilatation of the left ventricular outflow tract (LVOT) is rarely observed even in severely dilated left ventricles. Furthermore, the aortic annulus withstands tethering forces by the surrounding tissue and myocardium much better than the mitral annulus (MA). Assuming a linear progression and excluding patients with endocarditis and diseases of the aortic root complex, observational studies showed that, on average, chronic AR progresses within more than 25 years [39].

In conclusion, the most prevalent phenotype is the combination of AR with functional MR due to LV dilatation. MR per se does not lead to AR, and the simultaneous occurrence of (pure) primary AR and/or (pure) primary MR is rare but possible in the context of endocarditis.

Concerning the grading of single valvular lesions in AR and MR, current guidelines suggest the use of an integrative approach with respect to methodological limitations (Table 2). [7, 8, 11, 20]. Regarding the echocardiographic evaluation of the combination of AR and MR, no specific recommendations exist that would fundamentally differ from the evaluation of single-valve regurgitant lesions [3, 40]. However, assessment by echocardiography should attempt to identify the hemodynamic scenario in which the combination of AR and MR becomes relevant. Thus, echocardiography in multiple VHD—especially in patients with combined AR and MR—is challenging due to the interdependency of both forms of regurgitation [3, 5, 8, 24, 36].

The echocardiographic parameters of the integrative approach are influenced by the individual hemodynamic conditions, by anatomical specifics, and by methodological factors. Therefore, errors and misjudgments are possible. In addition, the existence of multiple true jets impedes the grading of AR and MR severity. However, in many cases, the finding of ‘multiple jets’ is the result of the echocardiographic cut-plane displaying segments of an elliptic, crescent-shaped, or non-circular geometric regurgitant orifice area (GROA). This is commonly seen in secondary MR, but also in AR, especially in patients with a bicuspid AV. In the presence of multiple jets, biplane assessment of the vena contracta (VC) can be used. However, there are no established cut-offs for this biplane assessment of combined MR and AR.

Severe discordances between echocardiography and cardiac magnetic resonance imaging (MRI) for grading MR severity were recently found [9, 10, 41]. In addition, the debatable data were reported in trials of interventional MR therapy for LVSV_tot and regurgitant volume through the MV (RegVol_MR), documenting low flow conditions which are not compatible with live conditions [42‐44], reveal the weakness of the echocardiographic integrative approach, if plausibility of hemodynamics is not considered. Therefore, a severe underestimation of LV end-diastolic volume (LVEDV) and overestimation of RegVol_MR can be assumed in these trials [45‐48]. Similar errors in grading AR and MR severity are probable in routine settings—especially because qualitative or semi-quantitative grading of AR and MR severity by jet area and the two-dimensional proximal isovelocity surface area (2D-PISA) method are still commonly used [49].

In this paper, we discuss the echocardiographic integrative approach in order to find evidence to support a quantitative approach for grading AR and MR severity [18, 50, 51]. There is still skepticism as to whether cardiac volumes can be accurately determined by echocardiography because several studies have reported differing cardiac volumes when measured by echocardiography or cardiac MRI [52, 53]. This is surprising because methodological studies using phantoms have shown comparable volumes between the two methods [54]. Contour delineation of the inner edge due to blurring underestimates volumes in the range of 5–10%, even in phantoms [52]. The pronounced differences in LV volume determination when using different methods are therefore incomprehensive, inconclusive, and contradictory [55, 56]. Plausible explanations for lower cardiac volumes by native 2D echocardiography in comparison with 2D contrast and native 3D echocardiography as well as with cardiac MRI in the clinical setting [55, 57, 58] are methodological errors due to foreshortening or differing contour delineations of the endocardium and limitations due to spatial resolution. In contrast to these previous studies, recent communications and trials using modern ultrasound technologies showed that comprehensive echocardiography can provide reliable and verifiable cardiac volume measurements by planimetry/volumetry as well as by Doppler echocardiography to correctly characterize cardiac hemodynamics [18, 59‐62]. In conclusion, using a definitive quantitative approach to grade the severity of valvular regurgitation includes the acceptance of the requirements to properly and plausibly determine LV volumes by echocardiography. Based on this assumption normal values, cut-off values of LVEDV, LV end-systolic volume (LVESV), regurgitant volumes and regurgitant fraction (RF) are provided in recommendations and guidelines for the echocardiographic assessment of valvular regurgitation [3, 7, 8, 11, 17, 20].

Concerning the quantitative assessment in isolated valvular AR or MR, similar cut-offs for regurgitant volume (≥ 60 mL) as well as RF (≥ 50%) have been defined for severe regurgitations.

Considering the methodological limitations of Doppler echocardiography and low-flow conditions in heart failure patients with secondary MR, a rigid cut-off of 60 mL for severe regurgitation might prove impractical in individual patients—especially when dealing with interdependent valve lesions [3, 7, 8, 11, 17, 20]. Therefore, a cut-off value of ≥ 45 mL for severe secondary MR has been proposed in low-flow conditions [20]. Consequently, when assessing the hemodynamic impact and relevance of combined AR and MR in a setting where each valve lesion seems only moderate, quantitative echocardiographic assessment should focus on estimating the total as well as each individual RF. Understanding the definition of the respective LV volumes—particularly LV filling volume, LVSV_tot and LVSV_eff—and their echocardiographic assessment is a prerequisite for reliably assessing hemodynamics in patients with combined AR and MR (see Fig. 1). Thus, the RegVol_AR and RegVol_MR can be estimated using different approaches (see Fig. 2).

In general, cardiac volumes—filling volumes, stroke volumes and regurgitant volumes—can be determined by different echocardiographic methods. Both Doppler techniques and volumetric measurements have methodological limitations. For example, when using pulsed-wave (PW) Doppler, it is essential to align the cursor with the position of the sample volume; when using planimetry or volumetry, labeling the mitral annulus and making a delineation between compacted and non-compacted myocardium make an accurate assessment challenging.

Both AR and MR tend to increase RV and PA pressures. Both lesions increase LV size, which in turn increases functional MR but almost never AR. Increased LVEF is recognized as a suitable due to AR contributes to LA and mitral annulus dilatation and thus to functional MR. LV dilatation caused by severe AR [7, 8, 20] must be critically verified in the setting of combined moderate AR and MR. An increased LV preload owing to the additional MR causes more severe and possibly earlier LV dilatation than LV dilatation caused by AR alone [37]. However, in the context of secondary MR, LV dilatation may also be caused by entirely other diseases like dilated or ischemic cardiomyopathy or myocarditis. In these scenarios, AR may only be a bystander. One possible way to discriminate between these two pathophysiological entities is to examine the ratio of leaflet area to annulus area, which is higher in patients with both significant AR and MR [37, 38]. Since both AR and MR increase LV preload and AR increases LVEDP while decreasing LV compliance, exercise echocardiography might be useful in assessing combined AR and MR. While also exposing typical symptoms, exercise testing can induce an increase in PA pressure with values ≥ 60 mmHg indicating significant hemodynamic relevance in combined AR and MR [63].

A general problem when assessing regurgitant volumes and regurgitant fractions is the reference size of the single valvular regurgitation. Whereas the ratio of RegVol_AR to LVSV_forward determines RF_AR, the ratio of RegVol_MR to LVSV_tot characterizes RF_MR. Thus, RF_AR is comprehensibly analyzed with varying regurgitant volumes through the AV, because an increase in RegVol_AR will cause an increase in LVSV_forward. In contrast, the relevance of RF_MR cannot be properly assessed by RegVol_MR/LVSV_tot, because RF_MR remains stable with an increasing amount of RegVol_AR. Theoretically, despite a decrease in LVSV_eff due to an increase of RegVol_AR (assuming a constant LV size), RF_MR can be stable. An increase in RegVol_MR (assuming an increasing LV size) RF_MR can be stable, too (see Figs. 3, 4, 5). Consequently, the contribution of RegVol_MR and RF_MR to the individual hemodynamic scenario in combined MR and AR can be surprisingly misinterpreted. Therefore, the determination of RegVol_tot to characterize the impact of both MR and AR might be more meaningful. In addition, it might be expedient to relate the individual regurgitant volumes of both MR and AR to LVSV_eff for a better characterization of their individual impact on hemodynamics (see Fig. 6). However, the ranges of the values for indexed RF_AR and indexed RF_MR as well as indexed RF_tot differ to the conventional values of RF_AR, RF_MR and RF_tot. The indexed RF values do not differ with respect to comparable amounts of regurgitant volumes at the respective valves. In addition, use of these indexed RF values is not yet introduced and implemented in current recommendations.

Echocardiographic assessment can be used to define the predominant lesion and underlying mechanism of the lesion as well as estimate the potential treatment implication for the predominant lesion. It puts the focus on anticipating whether or not the treatment of one singular lesion can improve the symptoms and/or the functional state of the other lesion.

In moderate and severe AR, at least moderate MR is present in 5–45% [6, 27, 30, 37]. In the study of Lim et al., 35% of patients undergoing isolated AV replacement due to severe AR had concomitant moderate functional MR [64]. In 88% of these patients, MR improved to mild functional MR associated with postoperative LV remodeling [64]. However, the current evidence is conflicting [65]. Another study showed lower survival rates for surgical AV replacement alone compared to the combined treatment of significant AR and at least moderate primary or secondary MR—especially when MV repair was feasible [27]. Furthermore, in patients with relevant combined AR, MR will most likely be secondary. However, primary MR is present in up to 5% of combined moderate or severe AR and MR, favoring concomitant surgical therapy [27]. Considering that significant MR has less or even little impact on AR severity [7, 8, 17, 20] simultaneous treatment of both AR and MR is understandable.

The current database of the Society of Thoracic Surgeons (STS) reveals an increasing number of concomitant aortic and mitral valve surgeries over the last years [66]. This may be due to advances in the surgical technique and growing experience in the perioperative setting. However, the morbidity and mortality of concomitant valvular surgery must still be taken into account [67, 68]. Thus, the decision to undergo surgical or interventional therapy remains challenging. The following factors should be considered:

In summary, the decision how to treat combined AR and MR is complex. Thus, it should be made by a team including cardiac surgeons, interventional cardiologists, and cardiovascular imaging specialists.

Significant individual variations in the LV volume despite a good correlation have been described for 2D planimetry/3D volumetry and Doppler echocardiography in the literature [56]. However, recent studies using Doppler echocardiography show no differences in cardiac output in comparison to thermodilution [60]. In addition, using modern techniques—particularly real-time 3D echocardiography—no significant or only minor differences in cardiac volumes are described in comparison with MRI [75, 76]. Thus, implementing new echocardiographic technologies in routine settings will presumably enable quantitative cardiac volume assessment in future, provided that echocardiography is performed correctly.