Full length article|Articles in Press

# Mitral Valve Coaptation Reserve Index; a Model to Localize Individual Resistance to Mitral Regurgitation caused by Annular Dilation

Open AccessPublished:November 13, 2022

## Abstract

### Objectives

The objective of this study was to develop a mathematical model for mitral annular dilatation simulation and to determine its effects on the individualized mitral valve coaptation reserve index (CRI).

### Design

Retrospective analysis of intraoperative transesophageal 3D-echocardiographic mitral valve (MV) datasets was performed. A mathematical model was created to assess the mitral CRI for each leaflet segment (A1-P1, A2-P2, A3-P3). Mitral CRI was defined as the ratio between the coaptation reserve (measured coaptation length along the closure line) and an individualized correction factor. Indexation was chosen to correct for MV sphericity and area of largest valve opening. Mathematical models were created to simulate progressive mitral annular dilatation and to predict the effect on the individual mitral CRI.

### Setting

Single Center Academic Hospital feasibility study.

### Participants

Twenty-five patients with normal functioning mitral valves undergoing cardiac surgery.

None.

### Measurements and Main Results

Direct measurement of leaflet coaptation along the closure line showed the lowest amount of coaptation (reserve) near the commissures (A1/P1 0.21+/-0.05 cm and A3/P3 0.22 +/- 0.06 cm) and the highest amount of coaptation (reserve) at region A2-P2 0.25 +/- 0.06 cm. After indexation, the A2-P2 region was the area with the lowest CRI in the majority of patients and also the area with the least resistance to mitral regurgitation (MR) occurrence after simulation of progressive annular dilation.

### Conclusions

Quantification and indexation of mitral coaptation reserve along the closure line is feasible. Indexation and mathematical simulation of progressive annular dilatation consistently showed that indexed coaptation reserve was lowest in the A2-P2 region. These results may explain why this area is prone to lose coaptation and is often affected in MR.

## Introduction

The mitral valve (MV) apparatus is a dynamic cardiac structure that aims to maintain left ventricular systolic competence and diastolic non-restriction. The coaptation zone of the valve is critical to valve competency and to preserving valve integrity. It is defined as the area of apposition between the anterior and posterior leaflets during left ventricular systole. This coaptation zone is the area that endures highest levels of mechanical stress during valve closure, where opposing forces from both leaflets interact
• Dellis SL
• Donnino R
• et al.
Analysis of the mitral coaptation zone in normal and functional regurgitant valves.
.
Identification and quantification of the amount of coaptation (reserve), the contact area between the anterior and posterior leaflet, may play an important role in determining a mitral valve's ability to sustain geometric distortions and its resistance to develop MR
• Frederick C
Cobey 1, Madhav Swaminathan 2, Barbara Phillips-Bute 2, Martin Hyca 3, Donald D Glower et al.
.

Nishimura RA, Otto CM, Bonow RO, et al.: 2017 AHA/ACC Focused Update of the 2014 AHA/ACC Guideline for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 135:e1159-e1195, 2017.

.
In this study we created a model to quantify the mitral valve coaptation zone along the MV leaflet closure line with correction for an individualized factor by indexation (the mitral valve coaptation reserve index or MV CRI). We hypothesized that our model could predict the location of MR and that the regions of lowest coaptation reserve would not consistently correspond to a low coaptation reserve index
In addition, we sought to develop a mathematical simulation model for annular dilatation and to determine its geometric effects on the individualized MV CRI.

## Methods

This study was designed as a single center retrospective feasibility study of prospectively collected data. Data analysis was performed as part of an institutional review-board-approved protocol of intraoperative data collection with a waiver of informed consent.

### Patients and Image Acquisition

Intraoperative real-time 3D transesophageal echocardiography (RT-3D TEE) and subsequent MV analysis was performed in 25 patients with normal functioning mitral valves (<mild) undergoing cardiac surgery for indications other than MV disease between July 2016 and April 2018 (Table 1 Demographics). Intraoperative transesophageal (TEE) real-time three-dimensional (RT-3D) was performed as part of routine intraoperative care, as described in our institution's perioperative imaging protocol for all patients undergoing cardiac surgery. 3D echocardiographic datasets were acquired after induction of general anesthesia and before sternotomy using a Vivid E9 ultrasound system and a GE 6VT-D, 4D TEE transducer (General Electric, Healthcare, Hoevelaken, NL). A standard mid-esophageal four chamber view, with a frame rate of 15 to 25 Hz, focus on the MV, and full inclusion of the MV was used to obtain the 3D echocardiographic datasets.
Table 1Demographics
 Male / Female (n) 21 / 4 Age (years) 64.6 ± 8.8 Weight (kg) 87 (83-94) Height (cm) 175.9 ± 9.3 Body Surface Area (m2) 2.05 ± 0.23 Body Mass Index (kg m−2) 28.6 ± 4.0 Type of Surgery (n) 20 CABG 2 AVR 1 Bentall 1 Lansac 1 Other Mitral insufficiency (n) 10 None 15 Trace Euroscore II 1.6 (1.1-1.9) Comorbidities (n) 10 Arterial hypertension 3 COPD 6 Diabetes 5 Cerebrovascular disease
AVR: aortic valve replacement; CABG: coronary artery bypass graft; COPD: chronic obstructive pulmonary disease.
Selection of data was narrowed to those patients in whom complete 3D echocardiographic datasets of the MV were available for off-line post processing and analysis, with no stitching artifacts (exclusion criterion). All 3D echocardiographic datasets were collected by the same anesthesiologist.

### Image Segmentation and Annular and Leaflet Modeling

Each full-volume 3D TEE data set was exported from EchoPac version 201 (General Electric, Healthcare, Hoevelaken, NL) to a separate computer workstation with TomTec Image Arena software and the semi-automated 4D MV assessment package (TomTec Imaging Systems GmbH, Munich, Germany) All analyses were performed at midsystole. MV parameters included in the analysis were MV anterior-posterior (AP) diameter, MV anterolateral-posteromedial (AL-PM) diameter, sphericity index of the MV saddle-shape, non-planarity angle, annular circumference, annular area, annular height, and commissural diameter
• Jeganathan J
• Knio Z
• et al.
Artificial intelligence in mitral valve analysis.
.

### Mitral Coaptation Analysis

The MV closure line was identified by the semi-automated 4D MV assessment package. To assess mitral coaptation we identified 7 coaptation points along the mitral closure line by placing 5 vertical lines through the MV closure line (Figure 1, Figure 2). Coaptation points were set individually by rotating the multiplanar reformatting lines (yellow, blue, and purple, Fig 1) while maintaining equal distances between the lines, starting with the first point right in the middle (A2-P2). The following 7 coaptation points were identified: the anterior commissure (k1), A1-P1 (k2), midpoint between A1-P1 and A2-P2 (k3), A2-P2 (k4), midpoint between A2-P2 and A3-P3 (k5), A3-P3 (k6), and the posterior commissure (k7). Finally, MV coaptation length at each of the 7 coaptation points was manually traced in the corresponding two-dimensional (2D) image (Fig. 1).

### Annular Dilatation Modeling and Coaptation Reserve Index

To determine the effect of progressive annular dilatation on the location of loss of coaptation along the closure line, we created a 2D model of the MV (Fig. 2) for simulation purposes. The model was customized for each patient, with the following individual parameters: AP diameter and AL-PM diameter, closure line length, and A2-P2 coaptation length.
In the model, the mitral annulus is described by the equation of an ellipse (Equation 1), where a is the major semi-axis (half of the AL-PM diameter), and b the minor semi-axis (half of the AP diameter).
$a=AL−PMdiameter2$

$b=APdiameter2$

$y=baa2−x2$
(1)

The closure line is described by the equation of a semi-ellipse (Equation 2), where e is the minor semiaxis and is calculated to obtain a closure line length matching the one that was measured in each patient (Equation 3). In the model, the closure line starts and ends slightly inside the ellipse perimeter representing the annulus (Fig. 2), this is obtained by introducing the arbitrary constant 1.33 in Equation 3. The constant was introduced to make the closure line start inside the perimeter of the ellipse. This is necessary to avoid a mathematical artifact: if the commissure points (k1-k7) were drawn on the perimeter of the ellipse, the model would not be able to simulate the displacement of these points that would remain in the same position.
$y2=eaa2−x2$
(2)

$e=2·(1.33·closurelinelengthπ)2−a2$
(3)

Again seven coaptation points were defined at equal distances along the closure line, corresponding to the 7 coaptation points measured in our cohort. Afterwards, a symmetrical dilatation of the mitral annulus was simulated by increasing the dimensions of the ellipse. First, for each patient, we calculated the hypothetical increase (k) in the AP diameter necessary to lose coaptation at A2-P2. As an example, an increase k = 1.1 corresponds to a 10% increase in the AP diameter.
$k=1+A2−P2coaptationlength·2APdiameter$

The semi-axes of the enlarged ellipse were calculated as follows and were used to draw the new ellipse (Equation 4, Fig. 2B).
$c=k·a$

$d=k·b$

$y3=dcc2−x2$
(4)

To simulate the displacement of the commissure points, the linear distances between the commissure line and the anterior and posterior annulus respectively were computed along the y axis in the first ellipse, and were kept constant when increasing the dimensions of the ellipse (e.g. b+e = d+f in Fig. 2). Finally, 7 points were drawn at equal distances along the free border of both mitral leaflets and the linear distance between each couple of points was computed (k1 to k7, Fig. 2B). By dividing each of these values by k4 (the largest distance, corresponding to the A2-P2 coaptation point), we obtained the relative proportions of each segment. We defined these values as correction factors (j1 to j7). By using the measured AP and AL-PM diameters, closure line length and A2-P2 coaptation length for each patient, the correction factors were individualized (Table 1).
The coaptation length is indexed against the anatomical characteristics of the mitral valve that are included in the mathematical model (i.e. the anterior-posterior and anterolateral-posteromedial diameters, the closure line length and the A2-P2 coaptation length).
We defined the CRI as the ratio between the coaptation reserve (measured coaptation length) and the respective (individualized) correction factor. Indexation was chosen to correct for MV sphericity and area of largest valve opening 45. Therefore, the CRI should be considered as an estimation of the coaptation reserve that takes into account both the measured coaptation length and the position of the coaptation points along the closure line. The individualized application of the model to each patient is shown in the Supplemental material (Fig. S1).
Total time from image acquisition to data analysis was on average 15 minutes per patient.

### Statistics

Continuous variables are expressed as mean ± standard deviation (SD) or median (interquartile range) depending on data distribution. Categoric variables are expressed as percentages. The normal distribution of continuous data was tested with the Shapiro-Wilk test. The coaptation reserve and the coaptation reserve index were compared among different coaptation points with a repeated-measures ANOVA, after checking normality and sphericity assumptions. Afterwards, pairwise comparisons were performed with the Student's t-test and corrected for multiple comparisons according to Bonferroni. To avoid an excessive increase in the chance of a type II error, only the A2-P2 coaptation point was compared to the others six coaptation points (6 multiple comparisons). The mitral annular area was indexed to the body surface area according to the Du Bois formula. All tests were performed two-tailed, and a P value <0.05 was considered statistically significant. R software (version 4.0) was used for statistical analysis [R Core Team (2018), Vienna, Austria].

## Results

### Baseline characteristics

Median age was 68 years (range 49-79), 4 patients were female and 21 male. Twenty patients underwent coronary artery bypass grafting, two patients underwent aortic valve replacement, one Bentall procedure, one Lansac procedure, and one patient with urothelial carcinoma with tumor growth into the vena cava. MR (D) was scored 0 (absent) in 10 patients and trace in 15 patients.

### Annular Geometry

Annular geometric values were all in the normal range, as shown in Table 2.
Table 2Correction factors and coaptation reserve (index)
Correction factor
Coaptation pointMeanMedianMinMaxCoaptation reserve (cm)Coaptation reserve index (cm)
A2-P2j410.25 ± 0.06
A1/A2-P1/P2j30.9480.9510.9260.9630.25 ± 0.060.26 ± 0.07
A2/A3-P2/P3j50.26 ± 0.060.27 ± 0.07
A1-P1j20.7800.7900.6990.8430.27 ± 0.060.34 ± 0.07a
A3-P3j60.26 ± 0.070.34 ± 0.09a
A1-P1 commissurej10.4580.4660.3230.5940.21 ± 0.05a0.46 ± 0.13a
A3-P3 commissurej70.23 ± 0.060.52 ± 0.14a
Summary of the individualized correction factors, coaptation reserve and coaptation reserve index. a P < .001 vs A2-P2 (the statistical significance for multiple comparisons was set at P < .008).
Table 3Mitral valve data
Mean ± SDMedianMinMax
AP-diameter (cm)3.28 ± 0.413.252.204.20
AL-PM diameter (cm)3.57 ± 0.413.592.754.61
Commissural diameter (cm)3.52 ± 0.403.522.734.53
Sphericity index0.92 ± 0.060.930.801.01
Closure line length (cm)3.31 ± 0.403.342.504.15
Annular circumference (cm)11.5 ± 1.311.58.314.8
Annular area (cm
• Frederick C
Cobey 1, Madhav Swaminathan 2, Barbara Phillips-Bute 2, Martin Hyca 3, Donald D Glower et al.
)
9.4 ± 2.29.34.915.4
Annular area (cm
• Frederick C
Cobey 1, Madhav Swaminathan 2, Barbara Phillips-Bute 2, Martin Hyca 3, Donald D Glower et al.
m−2)
4.6 ± 1.04.52.97.7
Annular height (cm)0.85 ± 0.160.840.491.12
Non-planarity angle (degrees)141.5 ± 9.4140.1125.0161.3
AP: anterior-posterior; AL-PM: anterolateral-posteromedial

### Coaptation and Coaptation Reserve Index

The coaptation reserve and the coaptation reserve index at each coaptation point along the closure line were charted for each patient (Fig. 3). When looking at the raw data (grey area in Fig. 3), the mean coaptation reserve was lower at the A1-P1 commissure, compared to the A2-P2 region. In contrast, after indexing using an individualized correction factor (green area in Fig. 3), the average coaptation reserve index was significantly higher at both commissures and at A1-P1 and A3-P3 as compared to A2-P2 (Table 1).
Similarly, from the raw data (direct measurements of coaptation), it appears that the point with the lowest coaptation reserve was located at the commissures in 60% of patients, while it was never located at the A2-P2 region (Fig. 4A). After indexing the values, however, the point with the lowest coaptation reserve index turned out to be A2-P2 or a nearby segment in 44% and 48% of cases, respectively (Fig. 4B). In only two patients the A1-P1 or A3-P3 points corresponded to the lowest coaptation reserve index. Summarizing the data, in less than one third (28%) of the patients the point of lowest coaptation reserve index corresponded to the point of lowest measured coaptation reserve. The data for each patient are presented in the Supplemental Material (Fig. S2).

### Annular Dilatation Modeling and Coaptation Reserve Index

Since the point with the lowest coaptation reserve index is expected to be the first to lose coaptation when the mitral annulus dilates, based on the definition given previously, we used the model to simulate in each patient various fixed percentage increases of the mitral annular area. Three representative examples are shown in Fig. 5, which presents the expected reduction of the coaptation reserve after a simulated 10%, 30% and 50% increase in the mitral annular area. In the hypothetical average patient, the first point to lose coaptation will be A2-P2 (Fig. 5A). In patient 14 (Fig. 5B), the lowest coaptation reserve is located at the commissures, but A2-P2 is expected to lose coaptation first, having the lowest coaptation reserve index. Alternatively, in patient 22 (Fig. 5C) A3-P3 will be the first point to lose coaptation, having the lowest coaptation reserve even after indexing the value. The simulations for the other patients are shown in the Supplemental Material (Fig. S3).

### Echocardiographic Location of MR and Coaptation Reserve Index

Fifteen patients had mild to trace MR, pre-operative TTE data was analyzed and the location of MR was blindly scored. In ten out of fifteen patients the MR location matched with the calculated lowest CRI, in five patients the lowest calculated CRI was in the region between A1/P1 and A2/P2 but was scored at the A1/P1 region on echo images.

## Discussion

In this study we calculated an individualized MV coaptation reserve index to asses areas of least coaptation between the anterior and posterior mitral valve, we also constructed a mathematical simulation model for annular dilatation.
Most studies analyzing mitral valve coaptation were performed in surgical patients after mitral valve plasty
• Lancellotti P
• Tribouilloy C
• Hagendorff A
• et al.
European Association of Echocardiography recommendations for the assessment of valvular regurgitation. Part 1: aortic and pulmonary regurgitation (native valve disease).
,
• Del Forno B
• Castiglioni A
• Sala A
• et al.
Mitral valve annuloplasty.
. Only one study did a quantitative assessment of the mitral valve coaptation zone and defined a coaptation index in functional mitral regurgitation patients, the authors demonstrated with analysis of variance that the coaptation index was associated with the severity of FMR and that there also was a correlation between 2D vena contracta and the coaptation index
• Frederick C
Cobey 1, Madhav Swaminathan 2, Barbara Phillips-Bute 2, Martin Hyca 3, Donald D Glower et al.
. In our study we used and indexation based on the following parameters, AP diameter, AL-PM diameter, closure line length, and A2-P2 coaptation length, derived from 3D datasets. Our population was different in the way that they had either no or trace MR. With regard to the fifteen patients with mild MR, ten patients had MR scored on TTE comparable with the region with lowest CRI. Five patients had MR scored at A1/P1 whereas the lowest CRI was between A1/P1-A2/P2. Reasons for this discrepancy could be that our measured coaptation area was done on more accurate 3D data, or the tendency to chose for a certain location (segment 1, 2 or 3) rather than guessing a in-between location. Although 2D-echocardiography is operator dependent and requires the finesse of probe manipulation to delineate MV pathology compared to 3D, we had similar results in 2/3 of our patients, and in the remaining 1/3 we had a slight difference of location. For future analysis the best option is to compare trace MR from 3D-color doppler data-sets, from which more accurately the location of the MR can be assessed, with lowest CRI, unfortunately in our patient population we did not had 3D-color doppler data available.
The strength of our model compared to more complex 3D analysis previously published are the following 1) It can be performed within 15 minutes compared to 3 hours
• Frederick C
Cobey 1, Madhav Swaminathan 2, Barbara Phillips-Bute 2, Martin Hyca 3, Donald D Glower et al.
2) 2D datasets can be used if accurately obtained instead of 3D acquired MV area, 3) we were able to construct a mathematical model mimicking MV annular dilation.
Direct measurement of leaflet coaptation along the closure line showed the lowest amount of coaptation (reserve) near the commissures in most patients of our cohort. In only a few patients the A2-P2 area showed the lowest amount of coaptation. This may be explained by the hypothesis that the coaptation reserve is not only related to the absolute value of the measured coaptation length, but also to the position of the coaptation point along the closure line. In addition, the coaptation area tapers towards the sides when it reaches the commissures, which explains the relatively smaller coaptation lengths towards the commissures (this study). Literature has shown when the mitral annulus dilates, the central coaptation points are likely to be displaced to a greater extent than the points that are closer to the commissures
• Apostolidou E
• Poppas A
Primary mitral valve regurgitation: Update and review.
• Sorajja P
• Mack M
• Vemulapalli S
• et al.
Initial Experience With Commercial Transcatheter Mitral Valve Repair in the United States.
• Glower D
• Argenziano M
• et al.
EVEREST II randomized clinical trial: predictors of mitral valve replacement in de novo surgery or after the MitraClip procedure.
. In order to account for this, we indexed the coaptation length at seven points along the MV closure line for each patient to calculate the MV coaptation reserve index (CRI). After indexation the A2-P2 region was the area with the lowest CRI in the majority of patients, and it also was the area with the least resistance to MR after simulation of progressive annular dilatation. These findings are physiologically plausible in that they demonstrate that the commissural regions are most likely to exhaust coaptation reserve with leaflet tethering and the central region geometrically suited to sustain annular dilation.
Development of a mathematical model and additional future studies are required to eventually determine if certain patient subgroups at risk of development of MR (i.e. patients with a low coaptation reserve) may benefit from regular monitoring and appropriate timing of early intervention.
The A2-P2 was the area with the least reserve and prone to lose coaptation with annular dilation in most of our patients. Although the coaptation reserve differs per patient when directly measured (raw data), it is within the expected range when the indexed value is used. It is important to realize that although the CRI is lowest at A2-P2 this doesn't mean that regurgitation will always occur at this location. Long term outcome of MR under medical treatment and after surgery is different in structural and functional disease
• Liu XM
• Wu H
• Zhang WK
• et al.
Long-term results of surgical treatment of aortic and mitral regurgitation with enlarged left ventricle.
• Lancellotti P
• Gerard PL
• Pierard LA
Long-term outcome of patients with heart failure and dynamic functional mitral regurgitation.
• Antoine C
• Benfari G
• Michelena HI
• et al.
Clinical Outcome of Degenerative Mitral Regurgitation.
. The natural history of structural regurgitation has been poorly defined, due to its complex nature and because of limitations in the assessment of its severity in the past
• Enriquez-Sarano M
• Akins CW
• Vahanian A
Mitral regurgitation.
. A follow up with indexed reserve areas could be an option for a more precise follow-up
• Shim H
• Harloff M
• Percy E
• et al.
Prediction for residual regurgitation after MitraClip for functional mitral regurgitation using leaflet coaptation index.
.
Our data provides additional insights into the MV coaptation zone. This study has several limitations: First, our applied model is a bi-dimensional oversimplification of a complex structure. Second, for simplicity we simulated a symmetrical dilation of the mitral annulus. Therefore, the prediction of imminent MR will probably be most accurate for impending Carpentier type 1 MR
• Carpentier A
Cardiac valve surgery–the "French correction".
, mitral annular dilation, pulling the leaflets away from each other. In other cases (i.e. eccentric dilations of the annulus and Carpentier type II (prolapse) or III (restriction) MR), the value of repeated pre-regurgitation 3D TEE is more important, due to the fact that progression of MR (due to structural degenerative or functional restrictive valvular disease) is much more difficult to predict. Third, our model considers mitral leaflets as fixed structures and does not take into account leaflet remodeling
• Kagiyama N
• Mondillo S
• Yoshida K
• et al.
Subtypes of Atrial Functional Mitral Regurgitation: Imaging Insights Into Their Mechanisms and Therapeutic Implications.
, this anatomical adaptation process of the MV leaflets occurs in response to stresses on the MV structure and might change coaptation reserve over time
• Deferm S
• Bertrand PB
• Verbrugge FH
• et al.
Atrial Functional Mitral Regurgitation: JACC Review Topic of the Week.
,
• Deferm S
• Bertrand PB
• Verhaert D
• et al.
Mitral Annular Dynamics in AF Versus Sinus Rhythm: Novel Insights Into the Mechanism of AFMR.
. Fourth, coaptation reserve studies are limited in that no study can account for the stretchability of the leaflets. Finally, although the analyzed data comes from 3D-data sets, we extracted the perfect orthogonal 2D images/data on which the analysis were done. The reason we did this was partly because we were limited by our software package, and because we wanted to use easily obtainable parameters, rather than 3D MV area which is more difficult to obtain. Performing new 3D-data rendering and analysis would mean writing and validating new software, which would be time consuming and not available to everyone. We used the TomTec software because it is 1) vendor independent, and 2) commercially available. Our calculations can be repeated by anyone (supplemental file).
In conclusion, quantification and indexation of mitral coaptation reserve along the closure line is feasible using 2D datasets derived from 3D echocardiographic data. Indexation and mathematical simulation of progressive annular dilatation, which was specific to our model, consistently can be used to reveal the hypothetical area of least coaptation. This study provides a mathematical model using an individualized MV coaptation reserve index, for quantifying the area of contact between the anterior and posterior mitral leaflets. Additional studies are required to determine the potential ‘’clinical’’ value of this mathematical model.

## Uncited References

[
• Sasaki H
• Mahara K
• et al.
Short Coaptation Length is a Predictor of Recurrent Mitral Regurgitation After Mitral Valve Plasty.
]

## Disclosures

TWLS received research grants and honoraria from Edwards Lifesciences (Irvine, CA, USA) and Masimo Inc. (Irvine, CA, USA) for consulting and lecturing (all payments made to institution).
MM is a consultant (theoretical and practical training activities and developement of new technologies) for Artivion, Atricure, CorCym, Medtronic. Grants from Abbott, Atricure, Getinge, Edwards.

None

## References

• Dellis SL
• Donnino R
• et al.
Analysis of the mitral coaptation zone in normal and functional regurgitant valves.
Ann Thorac Surg. 2010; 89: 1158-1161
• Frederick C
Cobey 1, Madhav Swaminathan 2, Barbara Phillips-Bute 2, Martin Hyca 3, Donald D Glower et al.
Ann Thorac Surg. 2014 Jun; 97: 1998-2004
1. Nishimura RA, Otto CM, Bonow RO, et al.: 2017 AHA/ACC Focused Update of the 2014 AHA/ACC Guideline for the Management of Patients With Valvular Heart Disease: A Report of the American College of Cardiology/American Heart Association Task Force on Clinical Practice Guidelines. Circulation. 135:e1159-e1195, 2017.

• Jeganathan J
• Knio Z
• et al.
Artificial intelligence in mitral valve analysis.
Ann Card Anaesth. 2017; 20: 129-134
• Lancellotti P
• Tribouilloy C
• Hagendorff A
• et al.
European Association of Echocardiography recommendations for the assessment of valvular regurgitation. Part 1: aortic and pulmonary regurgitation (native valve disease).
Eur J Echocardiogr. 2010; 11: 223-244
• Del Forno B
• Castiglioni A
• Sala A
• et al.
Mitral valve annuloplasty.
Multimed Man Cardiothorac Surg. 2017; : 2017
• Sasaki H
• Mahara K
• et al.
Short Coaptation Length is a Predictor of Recurrent Mitral Regurgitation After Mitral Valve Plasty.
Heart Lung Circ. 2021; 30: 1414-1421
• Apostolidou E
• Poppas A
Primary mitral valve regurgitation: Update and review.
Glob Cardiol Sci Pract. 2017; 2017e201703
• Sorajja P
• Mack M
• Vemulapalli S
• et al.
Initial Experience With Commercial Transcatheter Mitral Valve Repair in the United States.
J Am Coll Cardiol. 2016; 67: 1129-1140
• Glower D
• Argenziano M
• et al.
EVEREST II randomized clinical trial: predictors of mitral valve replacement in de novo surgery or after the MitraClip procedure.
J Thorac Cardiovasc Surg. 2012; 143: S60-S63
• Liu XM
• Wu H
• Zhang WK
• et al.
Long-term results of surgical treatment of aortic and mitral regurgitation with enlarged left ventricle.
International journal of clinical and experimental medicine. 2014; 7: 709-713
• Lancellotti P
• Gerard PL
• Pierard LA
Long-term outcome of patients with heart failure and dynamic functional mitral regurgitation.
Eur Heart J. 2005; 26: 1528-1532
• Antoine C
• Benfari G
• Michelena HI
• et al.
Clinical Outcome of Degenerative Mitral Regurgitation.
Circulation. 2018; 138: 1317-1326
• Enriquez-Sarano M
• Akins CW
• Vahanian A
Mitral regurgitation.
Lancet. 2009; 373: 1382-1394
• Shim H
• Harloff M
• Percy E
• et al.
Prediction for residual regurgitation after MitraClip for functional mitral regurgitation using leaflet coaptation index.
J Card Surg. 2020; 35: 3555-3559
• Carpentier A
Cardiac valve surgery–the "French correction".
J Thorac Cardiovasc Surg. 1983; 86: 323-337
• Kagiyama N
• Mondillo S
• Yoshida K
• et al.
Subtypes of Atrial Functional Mitral Regurgitation: Imaging Insights Into Their Mechanisms and Therapeutic Implications.
JACC Cardiovasc Imaging. 2020; 13: 820-835
• Deferm S
• Bertrand PB
• Verbrugge FH
• et al.
Atrial Functional Mitral Regurgitation: JACC Review Topic of the Week.
J Am Coll Cardiol. 2019; 73: 2465-2476
• Deferm S
• Bertrand PB
• Verhaert D
• et al.
Mitral Annular Dynamics in AF Versus Sinus Rhythm: Novel Insights Into the Mechanism of AFMR.
JACC Cardiovasc Imaging. 2022; 15: 1-13