Leaflet-Chordal Relations in Patients with Primary and Secondary Mitral Regurgitation

Leaflet-Chordal Relations in Patients with Primary and Secondary Mitral Regurgitation Kikuko Obase, MD, Lynn Weinert, BS, Andrew Hollatz, MD, Farhan Farooqui, MD, Joseph D. Roberts, MD, Mohammed M. Minhaj, MD, Avery Tung, MD, Mark Chaney, MD, Takeyoshi Ota, MD, Husam H. Balkhy, MD, Valluvan Jeevanandam, MD, Ken Saito, MD, Kiyoshi Yoshida, MD, Victor Mor-Avi, PhD, and Roberto M. Lang, MD, Chicago, Illinois; and Kurashiki, Japan

Background: The strategy for mitral valve (MV) repair has recently focused on the restoration of the submitral apparatus. However, the relationship between geometric changes of the submitral apparatus and the mitral leaflets has not been systematically investigated. The aim of this study was to determine the relationships among chordal length (CL) and LV size and leaflet surface area (LSA) in normal subjects, patients with primary (degenerative) mitral regurgitation (PMR), and patients with functional (secondary) mitral regurgitation (FMR). Methods: A total of 72 patients who underwent three-dimensional transesophageal echocardiography, including: 27 with PMR with isolated P2 flail leaflet, 25 with FMR with greater than mild mitral regurgitation, and 20 with normal mitral valves. LSA was quantified at midsystole from full-volume midesophageal views. CL was calculated by averaging the lengths of eight primary chords from transgastric full-volume data sets using multiplanar reconstruction. Results: Both CL and LSA in the PMR group were significantly longer compared with the FMR and normal control groups. No difference in CL was noted between patients with FMR and normal subjects. In all three groups, CL and LSA did not correlate with LV systolic or diastolic dimensions. Although CL did not correlate with LSA in the FMR group, a moderate correlation (R = 0.62) was observed in the PMR group. Conclusions: In patients with FMR with greater than mild mitral regurgitation, the chords retain normal length, despite LSA and LV enlargement. In patients with PMR with flail P2 scallops, CL elongation of primary chords is associated with larger LSA but not with LV dimensions. This information may have implications for clinical strategies for mitral valve repair surgery, including the submitral approach and percutaneous procedures. (J Am Soc Echocardiogr 2015;28:1302-8.) Keywords: Functional mitral regurgitation, Degenerative mitral regurgitation, Chordae tendineae, 3D transesophageal echocardiography

In patients with primary (degenerative) mitral regurgitation (PMR), the superiority of mitral valve (MV) repair over replacement in terms of improved outcomes is well established.1,2 The strategy for MV repair has recently focused on the restoration of the submitral apparatus, using neochord implantation, frequently performed without leaflet resection.3 In patients with functional (secondary) mitral regurgitation (FMR) with severely tethered leaflets, it has From the Section of Cardiology, Department of Medicine (K.O., L.W., V.M.-A., R.M.L.), the Department of Anesthesia & Critical Care (A.H., F.F., J.D.R., M.M.M., A.T., M.C.), and Cardiovascular Surgery (T.O., H.H.B., V.J.), University of Chicago, Chicago, Illinois; and the Department of Cardiology, Kawasaki Medical School, Kurashiki, Japan (K.O., K.S., K.Y.). Thomas J. Ryan, MD, FASE, served as guest editor for this report. Reprint requests: Roberto M. Lang, MD, Noninvasive Cardiac Imaging Laboratory, Section of Cardiology, Department of Medicine, University of Chicago Medical Center, 5841 S Maryland Avenue, Chicago, IL 60637 (E-mail: rlang@medicine. bsd.uchicago.edu). 0894-7317/$36.00 Copyright 2015 by the American Society of Echocardiography. http://dx.doi.org/10.1016/j.echo.2015.08.009

1302

been reported that the insertion of an undersized annuloplasty ring frequently fails to control mitral regurgitation (MR),4 and accordingly, additional submitral surgery strategies have been proposed, including chordal cutting, papillary muscle approximation, and/or suspension.5-7 In both these clinical scenarios, preoperative quantification of the submitral apparatus would be of clinical benefit. However, data on echocardiographic imaging aimed at quantifying the submitral apparatus are scarce. Degenerative MV disease includes chordal abnormalities that are associated with a spectrum of leaflet lesions.8 In FMR, leaflet adaptation accompanied by secondary chordal elongation has been described as a compensatory mechanism.9,10 In both these scenarios, three-dimensional (3D) echocardiographic imaging has provided valuable volumetric measurements of the mitral leaflets, but little has been described with respect to chordal morphology. To date, it remains unclear whether leaflet elongation in FMR and PMR is associated with chordal elongation. Accordingly, the aim of this study was to measure chordal length (CL) in PMR and secondary MR and determine the relationship between CL and leaflet surface area (LSA) in patients with FMR and PMR and in normal subjects.

Journal of the American Society of Echocardiography Volume 28 Number 11

Obase et al 1303

METHODS

Analysis of Transgastric Images. Full-volume data sets obtained from the transgastric approach were exported to dedicated analysis software (QLAB version 9.0 with the 3DQ plug-in; Philips Medical Systems) for multiplanar reconstruction. The detailed primary chordal measurement method using 3D TEE has been recently described by our group.12 Briefly, after the short-axis plane (Figure 1, left) was extracted to identify the root of the chords and their distribution, the long-axis plane (Figure 2, middle row) was visualized to separate primary chords from secondary or tertiary chords. Fine adjustment of the reference line in this long-axis plane allows visualization of the entire CL of the primary chord in the corresponding mitral bicommissural plane (Figure 2, right row). In each patient, the entire length of the primary chords was visualized in all eight segments: A2 lateral, A1, P1, P2 lateral, A2 medial, A3, P3, and P2 medial (Figure 1, right). The dominant chord in each segment was selected for measurement. Accordingly, eight CLs were measured in each patient, and the lengths of the eight chords were averaged to calculate CL. LVDd and LV endsystolic dimension were measured in the extracted LV short-axis multiplanar reconstructed view at the papillary muscle level. Analysis of Midesophageal Images. The full-volume data sets acquired from the midesophageal approach were exported to different analysis software (Image Arena, 4D-MV Assessment 2.3; TomTec Imaging Systems, Unterschleissheim, Germany). A midsystolic volume data set was selected, and the leaflet surface was automatically identified. Rotational cross-sections of the mitral annular plane centered at the midanterior annulus as well as the center point of the mitral annulus were used to verify the correct positioning of the data set. Manual edits were performed in cases of inadequate automated tracing. Cine loops were used for precise identification of leaflet configuration at midsystole. In addition to LSA, annular height and annular perimeter were measured.

Abbreviations

CL = Chordal length

Patients

A total of 72 patients, including 27 with severe PMR (mean age, FMR = Functional 60 6 14 years; 22 men) and 25 (secondary) mitral with FMR with more than mild regurgitation MR due to dilated left ventricles with reduced left ventricular LSA = Leaflet surface area (LV) function (mean age, LV = Left ventricular 62 6 13 years; 20 men), as well as 20 patients with normal MV LVDd = Left ventricular enddiastolic dimension morphology (the normal control group; mean age, 56 6 13 years LVEDV = Left ventricular endold; 16 men) were studied with diastolic volume 3D transesophageal echocardiogLVEF = Left ventricular raphy (TEE). Normal controls ejection fraction were selected from patients who underwent 3D TEE for the LVESV = Left ventricular endassessment of a cardioembolic systolic volume source of stroke. In this group, paMR = Mitral regurgitation tients were excluded if they had MV = Mitral valve more than mild MR, reduced LV ejection fraction (LVEF) PMR = Primary (<52% in men, <54% in (degenerative) mitral women) or increased LVEF regurgitation (>72% in men, >74% in TEE = Transesophageal women), or dilated left ventricles, echocardiography according to current guidelines.11 The normal range of LV size was 3D = Three-dimensional defined as a linear LV end2D = Two-dimensional diastolic dimension (LVDd) between 42.0 and 58.4 mm in men and between 37.8 and 52.2 mm in women.11 The FMR group included 11 patients with ischemic and 14 with nonischemic etiologies. This group consisted of patients with LVDd > 60 mm and LVEF < 30%, who were studied at the time of coronary artery bypass graft surgery with (n = 3) or without (n = 1) MV surgery, MV surgery alone (n = 4), and LV assist device implantation (n = 17). Patients with structurally abnormal MVs were excluded. The PMR group consisted of patients who underwent MV repair surgery for severe MR with isolated P2 ruptured chord(s) without evidence for significant leaflet thickening or excess tissue in other leaflet scallops. In both the FMR and PMR groups, 3D TEE was performed in the operating room after the induction of anesthesia and endotracheal intubation and before cardiopulmonary bypass. FED = Fibroelastic deficiency

Three-Dimensional Imaging and Analysis Images were acquired using an iE33 imaging system equipped with a fully sampled 3D matrix-array TEE transducer (model X72t; Philips Medical Systems, Andover, MA). Electrocardiographically gated fullvolume data sets of the MV were acquired from the midesophageal and transgastric approach over four consecutive cardiac cycles. When imaging from the midesophageal approach, attention was paid to ensure that the central ultrasound beam was kept parallel to the LV long axis. This allowed the ultrasound beam to traverse the mitral annular plane and optimize MV leaflet visualization. When imaging from the transgastric approach, care was taken to have the ultrasound beam perpendicular to the LV long axis to optimize chordal visualization. From both approaches, sector width and imaging depth were minimized to maximize spatial and temporal resolution.

Two-Dimensional Imaging and Analysis Standard two-dimensional (2D) transthoracic echocardiography was also performed before surgery on the iE33 (Philips Medical Systems). LV end-diastolic volume (LVEDV) and end-systolic volume (LVESV) were measured using the biplane disk summation method, and LVEF was calculated as [(LVEDV LVESV)/LVEDV]  100. The MR grade in the FMR group was quantified during intraoperative 2D TEE, using vena contracta width, which was measured as the narrowest portion of the regurgitant jet as it enters the receiving chamber in the long-axis view perpendicular to the coaptation line at midsystole. More than mild MR in the FMR group was defined as vena contracta width $ 0.3 cm.13 Reproducibility Inter- and intraobserver variability for the CL measurements was assessed in 10 randomly selected patients by repeated measurements. To determine interobserver variability, two observers independently measured 60 chords in these patients, while being blinded to all prior measurements. To determine intraobserver variability, one observer repeated the measurements in 80 chords $1 month later to minimize recall bias. Statistical Analysis All measurements are expressed as mean 6 SD. Comparisons between groups were performed using two-tailed t tests. P values < .05 were considered significant. The relationships among the measurements of CL, LSA, LV dimensions, annular height, and perimeter

1304 Obase et al

Journal of the American Society of Echocardiography November 2015

Figure 1 An example of the alignment of the root of the chords is shown in an extracted short-axis plane (left). The roots, which include the primary chords, align like a smiley mouth (middle, blue dots). The red dots indicate the root of the secondary chords of the anterior leaflet, while the green dots show secondary or tertiary chord of the posterior leaflet. The right panel shows the segmentation of the primary chords. When multiple chords were seen in one segment, the dominant chord was selected for measurement. In this case, the P2 lateral chords are absent due to chordal rupture. were calculated using linear regression analysis with Pearson correlation coefficients. Inter- and intraobserver variability was evaluated by calculating interclass correlation coefficients and absolute differences between the corresponding repeated measurements as percentages of their mean values.

RESULTS There were no intergroup differences in age and body surface area. LVDd was largest in the FMR group, followed by the PMR group. LV end-systolic dimension was largest and LVEF was lowest in the FMR group, while no difference in LV end-systolic dimension and LVEF were noted between the normal and PMR groups. Threedimensional measurements of annular perimeter and LSA were largest in the PMR group, followed by FMR, and smallest in the normal group. Annular height was lowest in the FMR group, whereas no differences were noted between the normal and PMR groups (Table 1). Examples of representative reconstruction images of the 3D measurements of annulus and leaflets in a normal subject and in patients with FMR and PMR are shown in Figure 3. Measurement of CL In all normal subjects and in 23 of 25 patients in the FMR group, all eight chords of every patient were successfully measured. In the remaining two patients in the FMR group, images were inadequate for CL measurements. Nevertheless, in each of these two patients, five chords in one patient and seven chords in another could be measured. In the PMR group, all eight chords were measured in 11 patients. In the remaining 16 patients, the lateral P2 chord was not measured in 12 patients, medial P2 was not measured in two patients, and both lateral and medial chords were not measured in two patients because of the presence of chordal rupture. Inter- and intraobserver variability of the measurements were 7.0 6 6.5% and 4.7 6 4.7%, with interclass correlation coefficients of 0.93 and 0.94, respectively. CL was longest in the PMR group (2.13 6 0.34, ranging from 1.40 to 2.97 cm), while no differences were observed between the normal (1.65 6 0.29, ranging from 1.23 to 2.24 cm) and FMR

(1.62 6 0.27, ranging from 1.10 to 2.10 cm) groups. Examples of A2 chords originating from the posterior papillary muscle in a normal subject and representative patients with FMR and PMR are shown in Figure 4. Correlations between CL and LV Dimension and LSA In the PMR group, CL correlated moderately with both annular perimeter (R = 0.57) and LSA (R = 0.62). These correlations were not observed in the normal and FMR groups. No correlations were observed between CL and LV dimensions or annular height in all groups (Figure 5).

DISCUSSION In this study, average CL was compared in patients with FMR with more than mild MR, patients with PMR with isolated P2 flail scallop, and normal control subjects. We found that (1) there was no correlation between CL and LV dimension in all three groups; (2) CL was not elongated in the FMR group, despite the presence of LV enlargement; and (3) in the PMR group, CL was elongated and positively correlated with both mitral annular perimeter and LSA. The anatomy of papillary muscles is highly diverse. However, the tips of the papillary muscles, where the chordae tendineae originate, appear to align systematically according to the leaflet distribution. Three-dimensional echocardiography allows clear visualization of this alignment, enabling CL measurements.12 To the best of our knowledge, no previous study has described the relationship between LV size and CL. Interestingly, in normal subjects, it has been reported that the distance between the papillary muscle tips and mitral annulus correlates with body surface area,14 which in turn appears to correlate with LV size. However, in the present study, no correlation was noted between CL and LV dimension in all three groups. Variability in papillary muscle length and insertion location of the papillary muscles is probably responsible for the lack of correlation between CL and LV dimensions. In a different study, the variability in LV papillary muscle location was

Obase et al 1305

Journal of the American Society of Echocardiography Volume 28 Number 11

Figure 2 Long-axis plane allows detailed identification of the primary chords (middle row). Fine adjustment of the reference line (green line) in the long-axis plane visualizes the entire CL in the mitral bicommissural plane (right row). Yellow arrows indicate the leaflet margin and the papillary tip. The distance between these structures was measured as the CL.

Table 1 Measurements in patients with normal MV, FMR, and P2 flail leaflet (PMR) Variable

Age (y)

NL group (n = 20)

FMR group (n = 25)

56 6 13

62 6 13

Body surface area (m2) 2.04 6 0.28 1.94 6 0.26

PMR group (n = 27)

60 6 14 1.92 6 0.23

LVDd (cm)

4.8 6 0.5

7.2 6 1.0*

5.3 6 0.6†‡

LV end-systolic dimension (cm)

3.2 6 0.4

6.5 6 1.0*

3.4 6 0.5‡

LVEF (%)

62.9 6 6.3

20.6 6 5.8*

61.4 6 6.3‡

Annular height (cm)

0.56 6 0.14 0.45 6 0.15* 0.66 6 0.21‡

Annular perimeter (cm) 10.7 6 1.0

11.9 6 1.3*

13.3 6 1.8†‡

9.3 6 1.7

13.1 6 3.0*

15.2 6 4.3†‡

2

LSA (cm ) Averaged CL (cm)

1.65 6 0.29 1.62 6 0.27

2.13 6 0.34†‡

*P < .05, normal versus FMR. † P < .05, normal versus PMR. ‡ P < .05, FMR versus PMR.

highlighted.15 Those investigators reported that the anterolateral papillary muscle was located in the basal third, middle third, and lower third in 19%, 79.5%, and 1.5% of patients, respectively.15 This was further corroborated by a recent study reporting that papillary muscles are predominantly positioned at the middle third of the LV wall (95%).16

In a previous study performed during autopsy of 50 normal hearts, the investigators reported direct measurements of rough zone CL to be 1.75 6 0.23 cm in the anterior leaflet chords and 1.40 6 0.08 cm in the posterior leaflets,17 values that are similar to our echocardiographic measurements. Contrary to our results of CL measurements, in an experimental sheep model of FMR, it was hypothesized that the mechanical stress imposed by chord and leaflet tethering elongates the secondary chords, also affecting the matrix with augmented thickness and cellular changes, suggesting an active MV adaptation process.18 However, in this sheep model, only secondary chords were studied, and the severity of MR was not considered. In the present study, the primary CL in patients with more than mild MR was studied. It is also reported that the development of significant MR in patients with dilated ventricles may not be solely associated with ventricular dilatation but also be due to insufficient MV adaptation.10 Therefore, our results in patients with FMR with more than mild MR may also reflect insufficient adaptation of the primary chords. From the surgical point of view, degenerative MV disease is divided into Barlow’s disease and fibroelastic deficiency (FED), mainly because these two entities require different surgical techniques and skills.19,20 Macroscopically, Barlow’s disease is characterized by diffusely redundant and thickened leaflets, elongated chords, and severe annular dilation, while FED is characterized by thin leaflets without excess tissue and thin and slightly elongated chords, with only a mildly dilated annulus. It has also been reported that these two entities have different

1306 Obase et al

Journal of the American Society of Echocardiography November 2015

Figure 3 Reconstructed 3D images for the measurement of annulus and leaflets in a patient with normal MV (NL), FMR, and flail P2 scallop (PMR) are shown.

Figure 4 Visualization of primary chords in a patient with normal MV (NL), FMR, and flail P2 leaflet. The arrows show the A2 chord originating from the posterior papillary muscle (PM), indicating the connecting point of the chord with the papillary muscle tip/leaflet margin. histologies.21 However, even in FED, localized myxomatous degeneration with thickening can frequently be observed in flail scallop(s). It therefore remains unclear whether these two entities are genetically different or represent a different spectrum of the same disease. Histologically, myxomatous degeneration in Barlow’s disease is characterized by expansion of the middle spongiosa layer by myxoid infiltration and structural collagen alterations. In contrast, FED is characterized by impaired connective tissue production with collagen deficiency. Fornes et al.21 noted that in Barlow’s prolapse, there is more myxoid infiltration in the leaflets compared with the chords. In contrast, in FED the altered scallops are poor in collagen alteration, elastic fiber lesions, and myxoid infiltration. Our results show that CL was positively correlated with LSA. This may indicate that the prevalence of degenerative involvement in leaflets and chords is similar, also suggesting a spectrum of valvular apparatus elongation. However, to determine the precise prevalence of the degenerative lesion in patients with PMR, not only LSA or CL but also the thicknesses of leaflets and chords need to be evaluated.

Limitations The true LSA is larger than LSA. LSA was measured in midsystole, and this measurement does not include the coaptation area. Therefore, enlargement of the leaflets in FMR, which has been reported in previous studies,10,22 was not accounted for in the present study. Of note, in the PMR group, only patients with flail P2 prolapse were studied, with the understanding that even in these patients, there is a spectrum of histologic and mechanical changes. To determine mild MR severity in the FMR group, only 2D TEE– derived vena contracta width was used. This is because the 2D TEE–derived flow convergence method is known to underestimate effective regurgitant orifice area because of the crescent shape of the regurgitant orifice. Only primary chords were measured in this study. Although technically, it is not difficult to measure the secondary chords either, the number of secondary chords and their anatomy vary widely among patients. Thus, further standardization of the measurement methodology is needed for the secondary chords.

Journal of the American Society of Echocardiography Volume 28 Number 11

Obase et al 1307

Figure 5 Correlations of averaged CL with LVDd and LSA in patients with normal MV (NL), FMR and P2 flail scallops (PMR). CONCLUSIONS In patients with FMR with more than mild MR, CL remains normal, despite LV enlargement. In patinets with PMR with flail P2 scallops, CL elongation of primary chords is associated with larger LSA but not with LV dimension. This information may have clinical implications for optimizing strategies for mitral repair surgery, including the submitral approach and percutaneous procedures.

REFERENCES 1. Perier P, Deloche A, Chauvaud S, Fabiani JN, Rossant P, Bessou JP, et al. Comparative evaluation of mitral valve repair and replacement with Starr, Bjork, and porcine valve prostheses. Circulation 1984;70: I187-92. 2. Mohty D, Orszulak TA, Schaff HV, Avierinos JF, Tajik JA, EnriquezSarano M. Very long-term survival and durability of mitral valve repair for mitral valve prolapse. Circulation 2001;104:I1-7. 3. Perier P, Hohenberger W, Lakew F, Batz G, Urbanski P, Zacher M, et al. Toward a new paradigm for the reconstruction of posterior leaflet prolapse: midterm results of the ‘‘respect rather than resect’’ approach. Ann Thorac Surg 2008;86:718-25. discussion 718. 4. Calafiore AM, Gallina S, Di Mauro M, Gaeta F, Iaco AL, D’Alessandro S, et al. Mitral valve procedure in dilated cardiomyopathy: repair or replacement? Ann Thorac Surg 2001;71:1146-52. discussion 1152.

5. Messas E, Guerrero JL, Handschumacher MD, Conrad C, Chow CM, Sullivan S, et al. Chordal cutting: a new therapeutic approach for ischemic mitral regurgitation. Circulation 2001;104:1958-63. 6. Rama A, Praschker L, Barreda E, Gandjbakhch I. Papillary muscle approximation for functional ischemic mitral regurgitation. Ann Thorac Surg 2007;84:2130-1. 7. Kron IL, Green GR, Cope JT. Surgical relocation of the posterior papillary muscle in chronic ischemic mitral regurgitation. Ann Thorac Surg 2002; 74:600-1. 8. Adams DH, Rosenhek R, Falk V. Degenerative mitral valve regurgitation: best practice revolution. Eur Heart J 2010;31:1958-66. 9. Grande-Allen KJ, Barber JE, Klatka KM, Houghtaling PL, Vesely I, Moravec CS, et al. Mitral valve stiffening in end-stage heart failure: evidence of an organic contribution to functional mitral regurgitation. J Thorac Cardiovasc Surg 2005;130:783-90. 10. Chaput M, Handschumacher MD, Tournoux F, Hua L, Guerrero JL, Vlahakes GJ, et al. Mitral leaflet adaptation to ventricular remodeling: occurrence and adequacy in patients with functional mitral regurgitation. Circulation 2008;118:845-52. 11. Lang RM, Badano LP, Mor-Avi V, Afilalo J, Armstrong A, Ernande L, et al. Recommendations for cardiac chamber quantification by echocardiography in adults: an update from the American Society of Echocardiography and the European Association of Cardiovascular Imaging. J Am Soc Echocardiogr 2015;28:1-39. 12. Obase K, Jeevanandam V, Saito K, Kesner K, Barry A, Hollatz A, et al. Visualization and measurement of mitral valve chordae tendineae using threedimensional transesophageal echocardiography from the transgastric approach. J Am Soc Echocardiogr 2015;28:449-54.

1308 Obase et al

13. Nishimura RA, Otto CM, Bonow RO, Carabello BA, Erwin JP 3rd, Guyton RA, et al. 2014 AHA/ACC guideline for the management of patients with valvular heart disease: executive summary: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines. Circulation 2014;129:2440-92. 14. Sonne C, Sugeng L, Watanabe N, Weinert L, Saito K, Tsukiji M, et al. Age and body surface area dependency of mitral valve and papillary apparatus parameters: assessment by real-time three-dimensional echocardiography. Eur J Echocardiogr 2009;10:287-94. 15. Victor S, Nayak VM. Variations in the papillary muscles of the normal mitral valve and their surgical relevance. J Card Surg 1995; 10:597-607. 16. Gunnal SA, Wabale RN, Farooqui MS. Morphological variations of papillary muscles in the mitral valve complex in human cadaveric hearts. Singapore Med J 2013;54:44-8. 17. Lam JH, Ranganathan N, Wigle ED, Silver MD. Morphology of the human mitral valve. I. Chordae tendineae: a new classification. Circulation 1970; 41:449-58.

Journal of the American Society of Echocardiography November 2015

18. Dal-Bianco JP, Aikawa E, Bischoff J, Guerrero JL, Handschumacher MD, Sullivan S, et al. Active adaptation of the tethered mitral valve: insights into a compensatory mechanism for functional mitral regurgitation. Circulation 2009;120:334-42. 19. Barlow JB, Pocock WA. Billowing, floppy, prolapsed or flail mitral valves? Am J Cardiol 1985;55:501-2. 20. Anyanwu AC, Adams DH. Etiologic classification of degenerative mitral valve disease: Barlow’s disease and fibroelastic deficiency. Semin Thorac Cardiovasc Surg 2007;19:90-6. 21. Fornes P, Heudes D, Fuzellier JF, Tixier D, Bruneval P, Carpentier A. Correlation between clinical and histologic patterns of degenerative mitral valve insufficiency: a histomorphometric study of 130 excised segments. Cardiovasc Pathol 1999;8:81-92. 22. Beaudoin J, Thai WE, Wai B, Handschumacher MD, Levine RA, Truong QA. Assessment of mitral valve adaptation with gated cardiac computed tomography: validation with three-dimensional echocardiography and mechanistic insight to functional mitral regurgitation. Circ Cardiovasc Imaging 2013;6:784-9.