From the Divi

(S.R., Z.S.); a

Canada (A.J.S

Reprint reque

600 Universit

mtsinai.on.ca)

0894-7317/$3

Copyright 201

http://dx.doi.o

Mitral Chordal-Leaflet-Myocardial Interactions inMitralValve Prolapse

Sherryn Rambihar, MD, FRCPC, Anthony J. Sanfilippo, MD, FRCPC, FASE, and Zion Sasson, MD, FRCPC,FASE, Toronto and Kingston, Ontario, Canada

Background: The submitral apparatus maintains annular-papillary continuity and myocardial geometry. Inmitral valve prolapse (MVP), elongated chords and redundant leaflets can interact at the region of myocardialattachment, leading to apparent discordant motion of the basal inferolateral wall. The aim of this study was totest the hypothesis that basal inferolateral wall inward motion would occur later in MVP and that this delay isassociated with MVP severity.

Methods: Thirty consecutive patients with MVP and matched controls underwent stress echocardiography.Time to peak transverse displacement (TPD) of the inferolateral wall compared with the anteroseptal wallwas measured using speckle-tracking echocardiography. The time difference was analyzed as raw data,normalized to the RR interval, and as a percentage of the time to maximal displacement of the anteroseptalsegment(s).

Results:Compared with controls, TPDwas delayed in patients with MVP both at rest and at peak stress, whenevaluating basal segments or basal-mid segments as a unit, both in real time and, more importantly, when cor-recting for anteroseptal TPD. In patients compared with controls, observed delay at rest and at peak stresswas 50 6 90 versus �30 6 90 msec (P = .006) and 70 6 80 versus �30 6 60 msec (P < .0001), respectively;relative to TPD of the anteroseptal segment, the observed delay at rest and at peak stress was 117 6 24%versus 97 6 22% (P = .007) and 144 6 68% versus 95 6 21% (P = .003), respectively. Similar significantfindings were observed in basal-mid segments. TPD results were not statistically significant when stratifiedby prolapse severity. Intraclass correlation coefficients were 0.88 and 0.93, and two-tailed t tests indicatedgood interobserver and intraobserver variability.

Conclusions: Inferolateral wall TPD is delayed in MVP. TPD is a novel method to characterize chordal-leaflet-myocardial interactions in patients with MVP. Prolapse severity does not predict TPD, likely because of thetiming of prolapse and dynamic loading conditions. Implications of this observation include attribution of aperceived wall motion abnormality in MVP during stress echocardiography to a physiologic state and newmechanistic insights into mitral valve physiology. (J Am Soc Echocardiogr 2014;-:---.)

Keywords: Mitral valve prolapse, Mechanics, Echocardiography, Speckle strain

The mitral chordal apparatus plays a key role in the maintenance ofventricular geometry and performance.1-12 Mitral valve prolapse(MVP) is a condition characterized, in part, by elongation of thechordal apparatus and delay in maximal stretch during systole.13,14

This delay may influence the timing of myocardial motion in thesegment(s) to which the elongated chordae attach and thereforecomplicate the assessment of left ventricular (LV) systolic function,both at rest and with the enhanced inotropy associated withexercise. A better understanding of this phenomenon may

sion of Cardiology, University of Toronto, Toronto, Ontario, Canada

nd Division of Cardiology, Queen’s University, Kingston, Ontario,

.).

sts: Zion Sasson, MD, FRCPC, FASE, Mount Sinai Hospital, 1601-

y Avenue, Toronto, ON M5X 1B2, Canada (E-mail: zsasson@

.

6.00

4 by the American Society of Echocardiography.

rg/10.1016/j.echo.2014.02.011

therefore benefit the echocardiographic interpretation of systolicfunction at rest and after exercise in patients with MVP.

We therefore sought to characterize mitral chordal-leaflet-myocardial (CLM) interactions in MVP using novel applications ofspeckle-tracking echocardiography. We hypothesized that the delayin mitral chordal-leaflet maximal stretch in patients with MVP willresult in delayed maximal inward motion of the basal inferolateralwall compared with the opposing corresponding basal anteroseptalsegment and that this delay is associated with MVP severity.

METHODS

Study Population

The study population consisted of 30 consecutive patients withMVP meeting the inclusion criteria, who underwent stress echocar-diography. Prolapse was defined as maximal end-systolic displace-ment of the body of the posterior mitral leaflet >2 mm superiorto the line connecting the annular hinge points in the parasternallong-axis view.15-18 All individuals had posterior mitral leaflet

1

Figure 1 Speckle strain analysisautomated measurement of peak

Abbreviations

LV = Left ventricular

MVP = Mitral valve prolapse

TPD = Time to peak

transverse displacement

2 Rambihar et al Journal of the American Society of Echocardiography- 2014

involvement. Inclusion criteriawere mitral regurgitation ofless than moderate severity,normal LV systolic function,and no LV regional wallmotion abnormalities. Thirtyage-matched and sex-matchedcontrols were selected from a

population of contemporaneous patients referred for stress echocar-diography at the same institution.

Exercise Stress Echocardiography

All patients underwent maximal symptom-limited treadmill exerciseechocardiography according to the Bruce protocol, as previouslyoutlined by the American Society of Echocardiography.19 Dopplerechocardiography with color flow was performed using a commer-cially available system (Vivid 7; GE Vingmed Ultrasound AS,Horten, Norway). Images were acquired from standard parasternaland apical views and stored digitally in cine loop format for threeconsecutive analyzable beats. Offline measurements were performedby a single observer using EchoPAC version 104.3.0 (GE VingmedUltrasound AS). Quantification of the extent of MVP was performedby two individuals.20 To assess intraobserver variability and avoidanceof recall bias, the same observer reassessed the amount of prolapse 6months after initial assessment in 10 randomly selected patients. Toassess interobserver variability, a second observer, blinded to theresults of the first, performed 10 independent measurements on thesame patients for two different parameters on a different readingstation.

Speckle-Tracking Strain Analysis

Standard grayscale two-dimensional images were obtained in para-sternal and apical views. All images were recorded at frame rates of70 to 100 Hz and digitally saved in cine loop format. Offlinespeckle-tracking analysis was performed using software for echocar-diographic quantification (EchoPAC version 104.3.0). Endocardialborders of the left ventricle were manually traced within the end-systolic frame. The epicardial tracing was automatically generated

, with automatic tracking of the LVdisplacement, from which the pri

by the software algorithm and manually adjusted when necessary.Tracking was accepted only if both visual inspection and EchoPACsoftware analysis confirmed that it was adequate. Peak transversedisplacement in the basal inferolateral, middle inferolateral, basalanteroseptal, and middle anteroseptal walls was assessed in apicallong-axis views, and a time-displacement profile was exported toMicrosoft Excel 2010 (Microsoft Corporation, Redmond, WA) forquantitative analysis.

Time to peak transverse displacement (TPD) was defined as thetime to maximal inward motion of a given myocardial segment(Figure 1). Speckle-tracking consistently tracks pixels within myocar-dial segments, thus tracking the displacement of myocardial segmentsat rest and immediately after exercise. By measuring and comparingthis parameter in basal and mid inferolateral and anteroseptalsegments, the extent of dyssynchrony could be determined inpatients with normal mitral valve and ventricular structure and inthose with MVP (Figures 2–4, Videos 1–4; available at www.onlinejase.com). Because chordal attachments may be variable, weassessed both the relative differences between TPD in basal segmentsand the average of TPD in all anteroseptal versus inferolateralsegments. Both groups were assessed at rest and immediately afterexercise. All data were abstracted electronically, using custom-programmed software macros, and checked qualitatively to ensurethat computer-derived results were not spurious.

To assess intraobserver variability, measurements of TPD andextent of MVP were performed by the same observer in 10 randomlyselected subjects (five patients and five controls) on the same cardiaccycle at baseline and repeated 6 months later. To assess interobservervariability, a second observer blinded to the results of the firstperformed independent measurements of the same variables in thesame 10 subjects and the same cardiac cycle on a different readingstation.

Statistical Analysis

Statistical analyses were performed using SPSS versions 11.0 and 20.0(SAS Institute Inc, Cary, NC). Prespecified analyses were performedusing paired and unpaired t tests for discrete variables and compari-sons for continuous variables, within and between patients withMVP and controls, at rest and at peak stress. Results are expressed

endocardial borders in the apical long axis. Demonstration ofmary end point, TPD, was derived.

Figure 2 Apical long-axis images at rest (A) and post-stress (B) in a patient with MVP, illustrating the MVP-related wall motionabnormality in the basal inferolateral wall(s). (Additional content: Video 1, patient with MVP at rest; Video 2, patient with MVP post-stress).

Figure 3 Speckle strain analysis, with automatic tracking of the LV endocardial borders in the apical long axis in a control subject (A)and in a patient with MVP (B). The basal inferolateral (yellow), mid inferolateral (cyan), basal anteroseptal (red), and mid anteroseptal(blue) segments are shown. (Additional content: Video 3, control; Video 4, patient with MVP).

Journal of the American Society of EchocardiographyVolume - Number -

Rambihar et al 3

Figure 4 Transverse displacement (y axis, mm) versus time (x axis, sec) derived from speckle strain analysis in the apical long axis, ina control subject (A) and in a patient with MVP (B). The basal inferolateral (yellow), middle inferolateral (cyan), basal anteroseptal (red),and middle anteroseptal (blue) segments are shown.

Table 1 Baseline characteristics

Variable

Patients with MVP

(n = 30)

Controls

(n = 30) P

Women 87% (26) 87% (26) NS

Age (y) 61.4 6 17.1 61.4 6 17.1 NS

Height (cm)* 160.3 6 6.5 161.3 6 9.2 NS

Weight (kg)* 64 6 8.6 65.5 6 11.9 NS

Resting HR (beats/min) 81 6 13 82 6 16 NS

Peak HR (beats/min) 147 6 15 149 6 20 NS

Resting BP (mm Hg) 118 6 18/73 6 9 120 6 20/69 6 12 NS

Peak BP (mm Hg) 168 6 27/72 6 17 170 6 24/66 6 11 NS

Time on treadmill (min) 7.7 6 2.4 8.6 6 2.6 NS

Mets achieved 8.7 6 2.8 9.63 6 3.6 NS

Maximal ECG ST-

segment deviation

at rest (mm)

�0.3 6 0.3 �0.5 6 0.4 NS

Maximal ECG ST-

segment deviation

at peak stress (mm)

�1.0 6 0.7 �1.1 6 0.9 NS

BP, Blood pressure; ECG, electrocardiographic; HR, heart rate.

Data are expressed as percentage (number) or mean 6 SD.*Limited data on height and weight were available for patients in the

case arm; data were abstracted from 9 subjects.

4 Rambihar et al Journal of the American Society of Echocardiography- 2014

as means or percentages with 95% confidence intervals.Nonparametric comparisons between groups were based onWilcoxon’s rank-sum test. Parametric comparisons between the twogroups were based on a two-sample Student’s t test. P values < .05were considered statistically significant. For the assessment ofsubgroups, the cut point for defining milder versus more severeprolapse was the modal value for prolapse, 4 mm. Unpaired t testswere used to determine differences between the two groups.Paired-samples two-tailed t tests with Pearson’s correlation coeffi-cients were used to assess the intraobserver variability of the samemeasures taken at baseline and 6 months later, and intraclass coeffi-cients were used to assess interobserver variability.

RESULTS

Clinical, electrocardiographic, and echocardiographic characteristicsof patients with MVP and age-matched and sex-matched controlsare compared in Table 1. There were no significant differences

between groups, with identical ages and sex and similar weights,heights, rest and peak heart rates, rest and peak blood pressures,and maximal ST-segment deviation at rest and during stress.

Compared with controls, patients with MVP demonstrated signif-icant delay in TPD of the segment(s) associated with the prolapsingleaflet at rest and at peak stress (P < .003 to P < .0001; Tables 2and 3). Results were statistically significant for all prespecified ana-lyses: raw data, normalized to the RR interval, and relative to theTPD for anteroseptal segment(s).

For the assessment of subgroups, we divided patients into groups ofthosewith less severe andmore severe prolapse according to themodalvalue in the population,#4-mmor >4-mm deviation from the annularplane. Results are shown in Table 4. The delay in TPD was not signifi-cantly different when assessed by severity of prolapse; numerically,however, the values trended in the hypothesized direction, with themyocardial segment involved with the prolapsing leaflet demonstratingearlier TPD and ‘‘normalizing’’ the perceived delay.

Reproducibility and Interobserver Variability

Excellent interobserver and intraobserver variability was observed,using data from 10 randomly selected patients for the amount ofMVP and TPD of the inferolateral and anteroseptal segments.Two-tailed t tests of values between one individual at baseline and af-ter 6 months were not statistically significant, with a Pearson’scoefficient of 0.656 for amount of prolapse and P = .883 for TPD.The mean difference between observers varied from 0% to 14%.Intraclass coefficients demonstrated excellent agreement (r = 0.88for amount of prolapse, r = 0.931 for TPD).

DISCUSSION

In MVP, the inward motion of the inferolateral myocardial segments,papillary muscle, and adjacent structures is temporarily dissociatedfrom the remaining LV wall segments. This appears to be due to thedelayed application of tethering forces by the abnormal mitral leafletsand chordae on the inferolateral papillary muscle and adjacentstructures. We believe this is the first comprehensive evaluation ofCLM interactions in this population. We demonstrated a delay inTPD in MVP involving the segment(s) to which abnormal chordsare attached. Although chordal insertion is variable, the delay wasdemonstrated in both proximal and averaged subtended segments.Prolapse severity does not predict TPD, possibly because of issueswith timing and dynamic physiologic loading conditions.

Table 2 TPD: delay of basal inferolateral wall relative to basal anteroseptal wall

Variable

Patients with MVP

(n = 30)

Controls

(n = 30) P

RestReal time (msec) 50 6 90 �30 6 90 .006

Relative to TPD of anteroseptal segment (%) (anteroseptal segment = 100%) 117 6 24 97 6 22 .007Peak stress

Real time (msec) 70 6 80 �30 6 60 <.0001Relative to TPD of anteroseptal segment (%) (anteroseptal segment = 100%) 144 6 68 95 6 21 .003

Data are expressed as mean 6 SD.

Table 3 TPD: delay of basal and mid inferolateral walls relative to basal and mid anteroseptal walls

Variable

Patients with MVP

(n = 30)

Controls

(n = 30) P

RestReal time (msec) 40 6 70 �10 6 80 .039

Relative to TPDof basal andmid anteroseptal segments (%)(basal and mid anteroseptal segments = 100%)

113 6 20 96 6 20 .034

Peak stress

Real time (msec) 60 6 70 �10 6 50 .001

Relative to TPDof basal andmid anteroseptal segments (%)

(basal and mid anteroseptal segments = 100%)

136 6 54 99 6 20 .007

Data are expressed as mean 6 SD.

Table 4 TPD in patients withMVP: delay of basal inferolateralwall relative to basal anteroseptal wall stratified by MVPseverity

Variable

Milder MVP*

(n = 15)

More Severe MVP†

(n = 15) P

TPD at rest (msec) 80 6 110 60 6 90 .65TPD at peak stress (msec) 70 6 110 30 6 110 .37

Data are expressed as mean 6 SD.*Milder MVP was defined as leaflet closure #4 mm beyond annular

line, the modal value.†More severe MVP was defined as leaflet closure >4 mm beyond the

annular line.

Journal of the American Society of EchocardiographyVolume - Number -

Rambihar et al 5

Much of our understanding of the physiologic interactionsbetween the mitral valve and the left ventricle is derived from cardiacsurgical literature assessing the role of chordal preservation in mitralvalve surgery.1-12,21-24 The ‘‘papillary muscle–annular continuity’’ iscrucial in mediating geometrically effective LV contraction duringsystole23 and moderating LV distension in diastole. The complexsaddle shape of the mitral valve also deforms through the cardiaccycle, with leaflet-annular relationships crucial in reducing leafletstress and enhancing systolic competence.25-27

Much of the literature to date characterizes the physiology andmechanics of the mitral apparatus by focusing on leaflet-annularrelationships using three-dimensional geometric modeling,two-dimensional echocardiography with radiopaque markers,and, more recently, real-time three-dimensional, full-volume acqui-sition with transesophageal echocardiography.25-30 TPD byspeckle-tracking echocardiography is a novel method to charac-terize CLM interactions in patients with MVP, which we believe

adds further insight to the understanding of mitral and LV dynamicsin MVP.

In MVP, increased leaflet redundancy, chordal elongation, andbillowing of the prolapsing segment into the left atrium in systoleapplies a delayed and rapidly developing force to the papillarymuscle.21 This is translated to the adjacent segment of the basalto mid inferolateral wall and visualized as a late and briskdeformation. Using speckle-tracking echocardiography, we wereable to demonstrate this as a delay in TPD of the basal inferolateralsegment relative to the corresponding opposite wall, a novelmethod for characterizing mitral CLM interactions in patientswith MVP. A recent cardiac magnetic resonance imaging studydemonstrated localized basal inferolateral hypertrophy in patientswith MVP, corresponding to the delayed segment of interest inour study.31 The association between these two observations isintriguing but requires further study.

A significant association between the severity of MVP and delayin TPD was not observed. This may be because severity, quantifiedas the absolute distance of leaflet prolapse, is not equivalent totiming, and factors governing timing may not affect severity perse. Additionally, as more severe MVP tends to occur earlier insystole, an earlier pull exerted on the papillary muscle results in amore rapid TPD, mimicking, in the extreme, normal synergy withonset of TPD in early systole. Superimposed on these scenariosare numerous physiologic factors that influence preload, afterload,and inotropy and may confound the timing of prolapse. As such,more complex models may be required to better characterize theserelationships.

The implications of our research extend beyond physiologiccharacterization to patient care. The association between MVP andsymptoms such as chest pain could possibly be associated with therapid development of tensile forces, which can be studied

6 Rambihar et al Journal of the American Society of Echocardiography- 2014

prospectively in the future. With stress echocardiography, the delayeddisplacement of the inferolateral wall due to physiologic mitral CLMinteractions may be identified as hypokinesis or tardokinesis andfalsely interpreted as ischemia. Attribution of a perceived wall motionabnormality in MVP during stress echocardiography to an MVP-related wall motion abnormality may thus be helpful.

Limitations

This study was a post hoc observational analysis, which attempted toexplore and characterize mechanistic aspects of mitral valve and LVmotion in patients with MVP. The small number of patients in theMVP cohort may explain the wide confidence intervals, but robuststatistical significance was achieved for all primary analyses. Controlpatients were referred for stress echocardiography for clinicalindications and may not represent a random population sample.However, with normal results on rest and stress echocardiography,we believed that this subset represented an appropriate controlgroup. Finally, the inability to delineate differences in TPD in milderversus more severe MVP likely relates to the role of timing in severityof prolapse and small sample size, or both.

CONCLUSIONS

Using novel speckle-tracking echocardiographic techniques, wehave demonstrated that in patients with MVP, there is a delay inmaximal inward motion of the basal inferolateral wall comparedwith the opposite corresponding basal anteroseptal segment andthat this delay varies with exercise inotropy. This builds on datafrom the surgical literature regarding the physiologic importanceof chordal apparatus and provides new mechanistic insights thatinform our understanding of the mitral chordal-leaflet-myocardialinteractions.

SUPPLEMENTARY DATA

Supplementary data related to this article can be found at http://dx.doi.org/10.1016/j.echo.2014.02.011.

REFERENCES

1. Sarris GE, Fann JI, Niczyporuk MA, Derby GC, Handen CE, Miller DC.Global and regional left ventricular systolic performance in the in situ eject-ing canine heart: importance of the mitral valve apparatus. Circulation1989;80(suppl):24-42.

2. Gams E, Schad H, Heimisch W, Hagl S, Mendler N, Sebening F. Preser-vation versus severance of the subvalvular apparatus in mitral valvereplacement: an experimental study. Eur J Cardiothorac Surg 1990;4:250-6.

3. David TE, Uden DE, Strauss HD. The importance of the mitral apparatusin left ventricular function after correction of mitral regurgitation. Circula-tion 1983;68(suppl):II-76-82.

4. David TE, Burns RJ, Bacchus CM,DruckMN.Mitral valve replacement formitral regurgitation with and without preservation of chordae tendineae. JThorac Cardiovasc Surg 1984;88:718-25.

5. Bonchek LI, Olinger GN, Siegel R, Tresch DD, Keelan MH Jr. Left ventric-ular performance after mitral reconstruction for mitral regurgitation. JThorac Cardiovasc Surg 1984;88:122-7.

6. David TE, Ho WC. The effect of preservation of chordae tendineae onmitral valve replacement for postinfarction mitral regurgitation. J ThoracCardiovasc Surg 1986;74(suppl):I-116-20.

7. Goldman ME, Mora F, Guarino T, Fuster V, Mindich BP. Mitral valvulo-plasty is superior to valve replacement for preservation of left ventricularfunction: an intraoperative two-dimensional echocardiographic study. JAm Coll Cardiol 1987;10:568-75.

8. Angell WW, Oury JH, Shah P. A comparison of replacement and recon-struction in patients with mitral regurgitation. J Thorac Cardiovasc Surg1987;93:665-74.

9. Miki S, Kusuhara K, Ueda Y, Komeda M, Ohkita Y, Tahata T. Mitral valvereplacement with preservation of chordae tendineae and papillary mus-cles. Ann Thorac Surg 1988;45:28-34.

10. Pitarys CJ II, Forman MB, Panayiotou H, Hansen DE. Long-term effects ofexcision of themitral apparatus on global and regional ventricular functionin humans. J Am Coll Cardiol 1990;15:557-63.

11. Tischler MD, Cooper KA, Rowen M, LeWinter MM. Mitral valve replace-ment versus mitral valve repair. Circulation 1994;89:132-7.

12. Takayama Y, Holmes JW, LeGrice I, Covell JW. Enhanced regional defor-mation at the anterior papillary muscle insertion site after chordal transec-tion. Circulation 1996;93:585-93.

13. FreedL, LevyD,LevineRA,LarsonMG,Evans JC, FullerDL, et al. Prevalenceand clinical outcome of mitral valve prolapse. N Engl J Med 1999;341:1-7.

14. Adams D, Rosenhek R, Falk V. Degenerative mitral valve regurgitation:best practice revolution. Eur Heart J 2010;31:1958-67.

15. Nishimura RA, McGoon MD, Shub C, Miller FA Jr., Ilstrup DM, Tajik AJ.Echocardiographically documented mitral-valve prolapse: long-termfollow-up of 237 patients. N Engl J Med 1985;313:1305-9.

16. Marks AR, Choong CY, SanfilippoAJ, Ferr�eM,WeymanAE. Identificationof high-risk and low-risk subgroups of patients with mitral-valve prolapse.N Engl J Med 1989;320:1031-6.

17. Levine RA, Stathogiannis E, Newell JB, Harrigan P,Weyman AE. Reconsid-eration of echocardiographic standards for mitral valve prolapse: lack of as-sociation between leaflet displacement isolated to the apical four chamberview and independent echocardiographic evidence of abnormality. J AmColl Cardiol 1988;11:1010-9.

18. Shah PM. Echocardiographic diagnosis of mitral valve prolapse. J Am SocEchocardiogr 1994;7:286-93.

19. Pellikka PA, Nagueh SF, Elhendy AA, Kuehl CA, Sawada SG. AmericanSociety of Echocardiography recommendations for performance, inter-pretation and application of stress echocardiography. J Am Soc Echocar-diogr 2007;20:1021-41.

20. Zoghbi WA, Enriquez-Sarano M, Foster E, Grayburn PA, Kraft CD,Levine RA, et al. Recommendations for evaluation of the severity of nativevalvular regurgitation with two-dimensional and Doppler echocardiogra-phy. J Am Soc Echocardiogr 2003;16:777-802.

21. Sanfilippo AJ, Harrigan P, Popovic A, Weyman AE, Levine RA. Papillarymuscle traction in mitral valve prolapse: quantitation by two-dimensional echocardiography. J Am Coll Cardiol 1992;19:564-71.

22. Granier M, Jensen M, Honge J, Bel A, Menasch�e P, Nielsen SL, et al. Con-sequences of mitral valve prolapse on chordal tension: ex vivo and in vivostudies in large animal models. J Thorac Cardiovasc Surg 2010;142:1585-7.

23. David TE. Papillary muscle-annular continuity: is it important. J Card Surg1994;9(suppl):252-4.

24. Athanasiou T, Chow A, Rao C, Aziz O, Siannis F, Ali A, et al. Preservationof the mitral valve apparatus: evidence synthesis and critical reappraisal ofsurgical techniques. Eur J Cardiothorac Surg 2008;33:391-401.

25. Levine RA, Handschumacher MD, Sanfilippo AJ, Hagege AA, Harrigan P,Marshall JE, et al. Three-dimensional echocardiographic reconstruction ofthe mitral valve, with implications for the diagnosis of mitral valve pro-lapse. Circulation 1989;80:589-98.

26. Levine RA, Triulzi MO, Harrigan P, Hagege AA, Harrigan P, Marshall JE,et al. The relationship of mitral annular shape to the diagnosis of mitralvalve prolapse. Circulation 1987;75:756-67.

27. Salgo IS, Gorman JH, Gorman RC, Jackson BM, Bowen FW, Plappert T,et al. Effect of annular shape on leaflet curvature in reducing mitral leafletstress. Circulation 2002;106:711-7.

28. Ormiston J, Shah P, Tei C, Wong M. Size and motion of the mitral valveannulus in man, II: abnormalities in mitral valve prolapse. Circulation1982;65:713-9.

Journal of the American Society of EchocardiographyVolume - Number -

Rambihar et al 7

29. Grewal J, Suri R, Mankad S, Tanaka A, Mahoney DW, Schaff HV, et al.Mitral annular dynamics in myxomatous valve disease: new insightswith real-time 3-dimensional echocardiography. Circulation 2010;121:1423-31.

30. Biaggi P, Jedrzkiewicz S, Gruner C, Meineri M, Karski J, Vegas A, et al.Quantification of mitral valve anatomy by three-dimensional transesopha-

geal echocardiography in mitral valve prolapse predicts surgical anatomyand the complexity of mitral valve repair. J Am Soc Echocardiogr 2012;25:758-65.

31. Zia M, Valenti V, Cherston C, Criscito M, Uretsky S, Wolff S. Relation ofmitral valve prolapse to basal left ventricular hypertrophy as determinedby cardiac magnetic resonance imaging. Am J Cardiol 2012;109:1321-5.