Original Contribution

The MitraClip Procedure in Patients With Moderate Resting but Severe Exercise-Induced Mitral Regurgitation

Jonathan Curio1,a;  Wajahat Tarar1,a;  Haitham Saleh Ali Al-Hindwan1;  Richard Neumann2;  Christian Berger, MD3; Marc-Oscar Hoting, MD1;  Mario Kasner, MD1;  Ulf Landmesser, MD1,b;  Markus Reinthaler, MD1,2,b

Jonathan Curio1,a;  Wajahat Tarar1,a;  Haitham Saleh Ali Al-Hindwan1;  Richard Neumann2;  Christian Berger, MD3; Marc-Oscar Hoting, MD1;  Mario Kasner, MD1;  Ulf Landmesser, MD1,b;  Markus Reinthaler, MD1,2,b

Abstract: Background. Optimal timing for percutaneous mitral regurgitation (MR) treatment using MitraClip (Abbott Vascular) remains unclear. We evaluated the outcome after MitraClip in patients with moderate resting MR, progressing to severe exercise-induced MR (MR2+) compared to patients with severe resting MR (MR3). Methods. We retrospectively investigated 221 patients undergoing MitraClip. All-cause deaths and heart failure (HF) hospitalizations were assessed as the combined primary endpoint. Results. We identified 55 MR2+ and 166 MR3 patients. At baseline, MR3 patients showed higher STS scores (6.7 ± 7.3 vs 4.4 ± 5.5; P<.01), more HF hospitalizations in the 2 years prior to the procedure (51% vs 29%; P<.01), worse left ventricular ejection fraction (44.9 ± 16.5% vs 52.5 ± 14.3%; P<.01), larger left ventricular end-diastolic diameter (LVEDd; 57.0 ± 9.3 mm vs 51.7 ± 8.2 mm; P<.001), and larger left atrial volumes (118.3 ± 55.8 mL vs 98.6 ± 35.2 mL; P=.02). Long-term outcome according to the combined endpoint was significantly worse in MR3 patients (P=.01). HF hospitalizations significantly declined in both groups 2 years after MitraClip (P<.001 in MR3 patients, P=.03 in MR2+ patients). Multivariate Cox regression analysis revealed LVEDd (hazard ratio, 1.035; 95% confidence interval, 1.005-1.066; P=.02) and previous HF hospitalizations (hazard ratio, 1.813; 95% confidence interval, 1.016-3.234; P=.04) as strong outcome predictors. Conclusions. Symptomatic patients with moderate resting and severe exercise-induced MR during handgrip echocardiography may represent an MR cohort at an earlier disease stage with improved treatment response following MitraClip implantation compared to individuals with severe resting MR. Larger left ventricular diameters and preprocedural HF hospitalizations were identified as independent adverse outcome predictors.


Key words: MitraClip, mitral regurgitation, MR severity, percutaneous edge-to-edge mitral valve repair

Transcatheter mitral valve (MV) repair with the MitraClip device (Abbott Vascular) has emerged as a potential treatment option for patients with significant mitral regurgitation (MR) and high or prohibitive surgical risk.1,2 Despite a growing number of treated patients worldwide, there is still a paucity of data to define characteristics predicting technical success and clinical improvement following the procedure.

According to current guidelines, MV repair/replacement is indicated in the presence of severe MR associated with symptoms or the presence of cardiac alterations caused by persisting left ventricular (LV) and atrial volume load like LV dilation/impairment or atrial fibrillation.3 However, some surgical data indicate that earlier intervention in the course of MR may result in an improved long-term outcome.4,5

Increasing attention has been paid to load-dependent MR severity.6-11 This entity is of clinical relevance, since patients with moderate MR at rest may still experience significant exercise-related limitations due to increasing MR severity on load. Recent trials and registries on percutaneous MV repair have also included patients with only moderate to severe MR, but have not investigated the treatment response in this subset of patients.12,13 Furthermore, only limited data are available regarding the role of exercise testing in these individuals.10,11

Thus, in the present study we report our experience of MV repair using the MitraClip device in symptomatic patients with moderate MR at rest who developed severe MR during handgrip echocardiography.


Study population and patient selection. This retrospective study included 221 patients undergoing a MitraClip procedure at our institution between November 2010 and July 2016. All patients were discussed by an interdisciplinary heart team and judged inoperable or at high surgical risk. Surgical risk was estimated using the Society of Thoracic Surgeons (STS) risk score.14 All patients suffered from persistent heart failure (HF) symptoms despite optimized medication; HF was thus most likely attributable to significant MR. Clinical relevance of MR was assumed in the presence of severe or moderate resting MR, the latter when combined with exercise-induced MR progression (handgrip echocardiography) and pulmonary hypertension (systolic pulmonary artery pressure [PAP] >30 mm Hg + central venous pressure). This retrospective study was approved by our local ethics committee (EA4/079/17).

Periprocedural management. Anticoagulation was discontinued at least 12 hours before the procedure. In patients on warfarin, an international normalized ratio of <2 was targeted. Prophylactic antibiotics were administered before the intervention in all patients.

All procedures were performed under general anesthesia with transesophageal echocardiographic (TEE) and fluoroscopic guidance. An activated clotting time between 250 and 350 seconds was achieved by heparin titration (initial bolus of 100 IU/kg body weight) and was then maintained throughout the procedure. The procedures were conducted as described elsewhere.1,15

Patients on oral anticoagulation were continued on these medications after the procedure. All other patients received 100 mg aspirin/day for at least 6 months in combination with 75 mg clopidogrel/day for 1 month.

Procedures were deemed successful when an MR reduction ≥1 grade was achievable while the mean transmitral gradient (MVG) was <5 mm Hg.

Postprocedural follow-up and endpoint definitions. After the procedure, MR severity, MVG, and PAP were assessed at the time of discharge using transthoracic echocardiography (TTE). To assess the patient outcomes, the primary endpoint was defined as the combined all-cause mortality and HF hospitalizations. These two endpoints also were analyzed separately. Data were screened up to 2 years following the procedure.

Echocardiographic assessment. In all patients, preprocedural TTE and TEE were performed by experienced sonographers to assess baseline parameters as well as MR severity and pathophysiology. In accordance with the recommendations of the European Association of Echocardiography (EAE), MR severity was graded on a scale ranging from 0 (none) to 3 (severe), and was determined using an integrative approach. Measurements of vena contracta and effective regurgitation orifice area (EROA) were considered, as well as regurgitation jet size in relation to left atrial (LA) size, eccentricity of the MR jet, and systolic flow reversal in the pulmonary veins.16 Patients with moderate MR but assumed clinical significance of MV dysfunction underwent MR evaluation under load by handgrip echocardiography, as published previously.17 Blood pressure measurements were taken before resting echocardiograms were performed. Commercially available equipment was used (Philips iE33 and EPIQ 7 [Philips Healthcare]; GE Vivid 7 [GE Healthcare]).

Statistical analysis. All continuous variables are depicted as mean ± standard deviation. All categorical variables are stated as absolute numbers of individuals (n) and percentages. Groups were compared using t-test, Mann-Whitney U-test, Chi-squared test, and McNemar test, wherever appropriate. Kaplan-Meier analyses were performed regarding the combined primary endpoint as well as all-cause mortality. Univariate and multivariate Cox regressions were calculated to prove event predictors for the combined endpoint and to define hazard ratios (HRs) where appropriate. Due to the low event rate, three multivariate regression models, each of them including 3 parameters, were calculated. In order to account for differences in our baseline characteristics, the STS score, which included most of the relevant baseline parameters, was incorporated in every model. A receiver operating characteristic (ROC) curve analysis was performed to define a cut-off value regarding the prognostic relevance of left ventricular end-diastolic diameter (LVEDd) measurements. P-values <.05 were deemed significant. All statistical analyses were executed using SPSS v. 24 for Mac OS (IBM SPSS Statistics).


Demographic data at baseline. A detailed description of the baseline patients characteristics is given in Table 1. In our cohort, a total of 55 patients (25%) suffered from moderate resting MR (grade 2+) with progression to severe MR under handgrip. Severe MR (grade 3) at rest was found in 166 patients (75%). A total of 75% of the patients in the MR 3 group and 84% in the MR 2+ group suffered from functional MR (P=.11). Baseline creatinine levels (1.6 ± 0.9 mg/dL in the MR 3 group vs 1.2 ± 0.4 mg/dL in the MR 2+ group; P<.01) and the incidence of renal insufficiency (54% in the MR 3 group vs 35% in the MR 2+ group; P=.01) were significantly increased in patients with MR 3. Furthermore, STS scores (6.7 ± 7.3 in the MR 3 group vs 4.4 ± 5.5 in the MR 2+ group; P<.01) and the number of preprocedural HF hospitalizations in the 2 years prior to the procedure (85 hospitalizations in MR 3 patients [51%] vs 16 hospitalizations in MR 2+ patients [29%]; P<.01) were significantly higher in MR 3 patients.

Echocardiographic parameters showed a significantly lower left ventricular ejection fraction in the MR 3 group (44.9 ± 16.5% vs 52.5 ± 14.3% in the MR 2+ group; P<.01) as well as larger LVEDd measurements (57.0 ± 9.3 mm vs 51.7 ± 8.2 mm in the MR 2+ group; P<.001) and LA volumes (118.3 ± 55.8 mL vs 98.6 ± 35.2 mL in the MR 2+ group; P=.02).

Procedure characteristics and technical success. Procedural characteristics are provided in Table 2. By our definition, procedural success was obtained in 177 patients (80%). Thirty-six patients (16%) were discharged with MVG >5 mm Hg and 11 patients (5%) showed MR reduction of <1 grade.

The total amount of MR improvement evaluated at the time of discharge was significantly higher in our MR 3 cohort (-1.8 ± 0.6 grades in the MR 3 group vs -1.6 ± 0.5 grades in the MR 2+ group; P<.01) (Figure 1). On the other hand, residual MR still was significantly worse in these patients (1.2 ± 0.6 in the MR 3 group vs 0.9 ± 0.5 in the MR 2+ group; P<.01). Significant PAP reductions were achievable in both groups when comparing discharge with baseline values (P<.001 each). No differences between the two groups concerning the extent of this reduction were found (-7.9 ± 13.2 mm Hg in the MR 3 group vs -12.1 ± 15.4 mm Hg in the MR 2+ group; P=.18). The incidence of patients with > grade 1 MR at discharge was higher in MR 3 patients (32% vs 7% in MR 2+ patients; P<.001) (Figure 1). Significantly more clips were implanted in MR 3 patients (1.6 ± 0.7 clips vs 1.3 ± 0.5 clips in MR 2+ patients; P<.01), which did not result in higher MVG at discharge (3.7 ± 1.5 mm Hg in MR 3 patients vs 3.6 ± 1.7 mm Hg in MR 2+ patients; P=.72).

Outcome regarding the primary endpoint. Median follow-up time was 374 days (interquartile range, 618 days). The Kaplan-Meier curves for the investigated groups are shown in Figures 2-4. Significantly more patients with baseline MR 3 suffered the combined endpoint after 2 years (log-rank P=.01). The difference concerning the combined endpoint was driven by the incidence of decompensations (log-rank P=.02 vs P=.16 for death). At 2 years, a total of 38 patients (23%) in the MR 3 group and 8 patients (15%) in the MR 2+ group were hospitalized for HF (log-rank P=.02). In both groups, a significant reduction of HF hospitalizations in the 2 years after the procedure compared with the 2 years prior to the procedure was achievable (P<.001 in MR 3 patients; P=.03 in MR 2+ patients)(Figure 5). The mortality rates after 2 years were 4% in the MR 2+ group and 8% in the MR 3 group (log-rank P=.16).

As depicted in Figure 3, individuals with preprocedural hospitalizations for HF showed a worse outcome (log-rank P<.01). The same was true for patients with increased LVEDd (Figure 4). In a ROC curve analysis, an LVEDd cut-off for the increased incidence of endpoint events was found at a level of 53.5 mm (area under the curve, 0.696; P<.001; sensitivity, 86%, specificity, 54%; log-rank P<.001 for cut-off) (Figures 4A and 4B).

STS score remained a significant predictor of worse outcome in all three multivariate Cox regression models: HR, 1.060; 95% CI, 1.023-1.098 (P<.01); HR, 1.046; 95% CI, 1.003-1.090 (P=.03); and HR, 1.060; 95% CI, 1.019-1.102 (P<.01), respectively (Table 3). Preprocedural HF admissions (HR, 1.813; 95% CI, 1.016-3.234; P=.04) and increased LVEDd (HR, 1.035; 95% CI, 1.005-1.066; P=.02) were also related to worse outcome in multivariate testing (Table 3).

Patients who were discharged with residual MR >1 (more common in the MR 3 group), showed a worse 2-year outcome with regard to the combined endpoint in a Kaplan-Meier analysis (log-rank P=.02).


This study was intended to determine the outcome of patients with moderate resting MR (MR 2+), progressing to severe MR on exercise, undergoing percutaneous repair with the MitraClip device. Our main findings were: (1) compared to patients with severe resting MR (MR 3), our MR 2+ cohort presented with significantly smaller LV and LA cavities as well as a more preserved LV function at baseline, indicating an earlier stage of the disease; (2) the combined primary endpoint of all-cause mortality and HF hospitalizations was significantly less frequently observed in MR 2+ patients; (3) the incidence of > grade 1 residual MR after repair, which was related to worse long-term outcomes, was observed less frequently in these individuals; and (4) HF hospitalizations preceding the intervention and LV dilation (both signs of advanced HF) independently predicted worse outcome.

The timing of MV interventions in patients with high or prohibitive surgical risk suffering from MR is one of the most challenging decisions of the interdisciplinary heart team, as several individual factors need to be considered. The severity of MR is highly pre- and after-load dependent, and may be misleading in terms of decision making. It is therefore important to consider the level of blood pressure control, fluid balance, and echocardiographic as well as clinical signs of HF.18-20

In our patient cohort, we found a substantial number of individuals with moderate resting MR combined with clinical signs of HF, who showed increasing MR severity during handgrip echocardiography. Compared with our MR 3 patients, this group was characterized by smaller LV and LA cavities as well as superior LV function. Since worsening of these parameters may be interpreted as the cause of MR (functional MR) or the consequence of MR (degenerative MR), it is important to mention that the distribution of MR etiology was comparable in both groups.

In the present study, individuals with MR 3 were also more likely to suffer from HF symptoms necessitating hospitalization during the preprocedural period.

Taken together, the baseline characteristics indicate that our MR 2+ individuals most likely represent a cohort at an earlier stage of the disease.21-24 Although current guideline recommendations for MV intervention do not refer to these individuals, it may still be important to identify these patients, as recent investigations in surgical patients found a favorable outcome when MR correction was performed at an earlier stage.5,25,26

Two recent studies have also evaluated the outcome after MitraClip in patients with functional MR and found diverging results. While the COAPT trial reported a significant benefit in terms of HF hospitalizations and all-cause mortality following MitraClip, the MITRA-FR study found outcomes in patients managed medically and those undergoing transcatheter intervention to be similar.27,28 However, a subgroup of patients with severely dilated LV and less severe MR in the COAPT trial, which corresponds to the overall MITRA-FR cohort, also showed a lack of treatment-response after percutaneous mitral repair. On the other hand, the respective COAPT subgroup with such less-severe MR (but at the same time smaller LV cavities) experienced a benefit comparable to the overall cohort of the study.29

These results highlight the relevance of different disease stages in the patient selection for transcatheter MV repair. As indicated by our MR 2+ cohort, which according to LV dimension and MR severity resembles the promising trial subgroup described above, exercise testing may help to define an additional group of individuals benefiting from MV repair.

Importantly, a clinical benefit was obvious in both groups of our study, as indicated by a significant decline in HF hospitalizations following the procedure. Furthermore, MR improvement and regression of PAP following the procedure were significant and comparable in both cohorts. However, the degree of residual MR was worse in patients with MR 3, which may have caused the increased incidence of HF admissions in these patients at 2-year follow-up.30,31 Mortality rates were low in both groups, without a significant difference (4% in MR 2+ vs 8% in MR 3 patients).

Our findings raise the question of whether MR severity itself or LV remodeling caused by MR is responsible for the prognostic divergence of the analyzed groups. As indicated in our multivariate analysis, STS scores (but not MR severity at baseline) determined the outcome regarding the combined primary endpoint. However, parameters like left ventricular ejection fraction, pulmonary hypertension, and kidney function are incorporated in the STS score, which is therefore also likely to represent the stage of the disease. Furthermore, the presence of preprocedural HF admissions and an LVEDd of >53.5 mm were independent predictors for worse outcome in the present study, which again indicates the prognostic significance of late-stage MR, corresponding to advanced HF.

These results indicate that some patients with moderate resting MR and optimized medical treatment may still benefit from mitral repair. Clinical and echocardiographic signs of HF, as well as load dependency of MR, may help to identify patients who benefit from a more aggressive approach than proposed by current guidelines.

Study limitations. Some limitations need to be noted. First, this study is limited by the retrospective observational design and the rather small number of patients, especially in the MR 2+ group. This may have affected the validity of the regression analysis. Second, due to the limited number of patients, we were not able to consider the etiology of MR in our analysis, as the cut-off levels of LV dimensions as well as the optimal timing of repair are likely to differ between patients with functional and degenerative MR. Third, data of patients managed conservatively are not available, which is why we cannot definitely conclude that earlier MV repair results in an improved outcome in this complex inoperable or high surgical risk patient setting. However, HF admissions significantly declined after the procedures in both groups, and the outcome regarding the combined endpoint was in favor of MR 2+ patients. Overall, our study represents a well-characterized, contemporary, real-world patient population undergoing a MitraClip procedure in a high-volume center in Germany.   


High or prohibitive surgical risk patients with moderate MR, clinical signs of HF, and MR progression during handgrip echocardiography may benefit from percutaneous MV repair with the MitraClip device. When compared with patients with severe resting MR, these patients showed an improved outcome and technical result regarding residual MR.

Irrespective of baseline MR severity, the presence of a preprocedural HF admission or an LVEDd >53.5 mm at baseline independently predicted worse outcome in our study, indicating the prognostic significance of late-stage MR and advanced HF. Handgrip echocardiography, however, may help to identify patients in whom moderate resting MR already is clinically relevant. According to our results, these individuals may show improved outcome after a MitraClip procedure. Larger studies, including a comparison of interventional versus medical treatment, are needed to answer the question of whether more aggressive management of MR is indicated in this subset of patients.

Acknowledgment. We would like to thank Professor Andreas Lendlein, HZG Teltow, for consultation during drafting of the manuscript.

aJoint first authors; bJoint senior authors.

From the 1Department of Cardiology, Charité Berlin, Campus Benjamin Franklin, Berlin, Germany; 2Institute of Biomaterial Science, Helmholtz-Zentrum Geesthacht, Teltow, Germany; and 3Department of Anesthesiology, Charité Berlin, Campus Benjamin Franklin, Berlin, Germany.

Funding: This work was supported by the Helmholtz-Association through program-oriented funding.

Disclosure: The authors have completed and returned the ICMJE Form for Disclosure of Potential Conflicts of Interest. Prof Landmesser reports personal fees from Abbott Vascular, outside the submitted work. The remaining authors report no conflicts of interest regarding the content herein.

Manuscript submitted June 20, 2019 and accepted July 2, 2019.

Address for correspondence: Dr Markus Reinthaler, Department of Cardiology, Charité Berlin, Campus Benjamin Franklin, Hindenburgdamm 30, 12203 Berlin. Email: markus.reinthaler@gmx.at

  1. Feldman T, Foster E, Glower DD, et al. Percutaneous repair or surgery for mitral regurgitation. N Engl J Med. 2011;364:1395-1406.
  2. Feldman T, Kar S, Elmariah S, et al. Randomized comparison of percutaneous repair and surgery for mitral regurgitation: 5-year results of EVEREST II. J Am Coll Cardiol. 2015;66:2844-2854.
  3. Baumgartner H, Falk V, Bax JJ, et al. 2017 ESC/EACTS guidelines for the management of valvular heart disease. Eur Heart J. 2017;38:2739-2791.
  4. Kang DH, Kim JH, Rim JH, et al. Comparison of early surgery versus conventional treatment in asymptomatic severe mitral regurgitation. Circulation. 2009;119:797-804.
  5. Kang DH, Park SJ, Sun BJ, et al. Early surgery versus conventional treatment for asymptomatic severe mitral regurgitation: a propensity analysis. J Am Coll Cardiol. 2014;63:2398-2407.
  6. Keren G, Katz S, Strom J, Sonnenblick EH, LeJemtel TH. Dynamic mitral regurgitation. An important determinant of the hemodynamic response to load alterations and inotropic therapy in severe heart failure. Circulation. 1989;80:306-313.
  7. Kizilbash AM, Willett DL, Brickner ME, Heinle SK, Grayburn PA. Effects of afterload reduction on vena contracta width in mitral regurgitation. J Am Coll Cardiol. 1998;32:427-431.
  8. Lancellotti P, Troisfontaines P, Toussaint AC, Pierard LA. Prognostic importance of exercise-induced changes in mitral regurgitation in patients with chronic ischemic left ventricular dysfunction. Circulation. 2003;108:1713-1717.
  9. Lancellotti P, Piérard LA. Chronic ischaemic mitral regurgitation: exercise testing reveals its dynamic component. Eur Heart J. 2005;26:1816-1817.
  10. Magne J, Lancellotti P, Piérard LA. Exercise-induced changes in degenerative mitral regurgitation. J Am Coll Cardiol. 2010;56:300-309.
  11. Bakkestrøm R, Banke A, Christensen NL, et al. Hemodynamic characteristics in significant symptomatic and asymptomatic primary mitral valve regurgitation at rest and during exercise. Circ Cardiovasc Imaging. 2018;11:e007171.
  12. Maisano F, Franzen O, Baldus S, et al. Percutaneous mitral valve interventions in the real world: early and 1-year results from the ACCESS-EU, a prospective, multicenter, nonrandomized post-approval study of the MitraClip therapy in Europe. J Am Coll Cardiol. 2013;62:1052-1061.
  13. Vakil K, Roukoz H, Sarraf M, et al. Safety and efficacy of the MitraClip® system for severe mitral regurgitation: a systematic review. Catheter Cardiovasc Interv. 2014;84:129-136.
  14. O’Brien SM, Shahian DM, Filardo G, et al. The Society of Thoracic Surgeons 2008 cardiac surgery risk models: part 2 — isolated valve surgery. Ann Thorac Surg. 2009;88:S23-S42.
  15. Mauri L, Garg P, Massaro JM, et al. The EVEREST II trial: design and rationale for a randomized study of the evalve MitraClip system compared with mitral valve surgery for mitral regurgitation. Am Heart J. 2010;160:23-29.
  16. Lancellotti P, Moura L, Pierard LA, et al. European Association of Echocardiography recommendations for the assessment of valvular regurgitation. Part 2: mitral and tricuspid regurgitation (native valve disease). Eur J Echocardiogr. 2010;11:307-332.
  17. Spain MG, Smith MD, Kwan OL, DeMaria AN. Effect of isometric exercise on mitral and aortic regurgitation as assessed by color Doppler flow imaging. Am J Cardiol. 1990;65:78-83.
  18. Stevenson LW, Bellil D, Grover-McKay M, et al. Effects of afterload reduction (diuretics and vasodilators) on left ventricular volume and mitral regurgitation in severe congestive heart failure secondary to ischemic or idiopathic dilated cardiomyopathy. Am J Cardiol. 1987;60:654-658.
  19. Pozzoli M, Traversi E, Cioffi G, Stenner R, Sanarico M, Tavazzi L. Loading manipulations improve the prognostic value of Doppler evaluation of mitral flow in patients with chronic heart failure. Circulation. 1997;95:1222-1230.
  20. Grayburn PA. How to measure severity of mitral regurgitation: valvular heart disease. Heart. 2008;94:376-383.
  21. Rossi A, Cicoira M, Zanolla L, et al. Determinants and prognostic value of left atrial volume in patients with dilated cardiomyopathy. J Am Coll Cardiol. 2002;40:1425.
  22. Trichon BH, Felker GM, Shaw LK, Cabell CH, O’Connor CM. Relation of frequency and severity of mitral regurgitation to survival among patients with left ventricular systolic dysfunction and heart failure. Am J Cardiol. 2003;91:538-543.
  23. Gaasch WH, Meyer TE. Left ventricular response to mitral regurgitation: implications for management. Circulation. 2008;118:2298-2303.
  24. Messika-Zeitoun D, Bellamy M, Avierinos JF, et al. Left atrial remodelling in mitral regurgitation — methodologic approach, physiological determinants, and outcome implications: a prospective quantitative Doppler-echocardiographic and electron beam-computed tomographic study. Eur Heart J. 2007;28:1773-1781.
  25. Gillinov AM, Mihaljevic T, Blackstone EH, et al. Should patients with severe degenerative mitral regurgitation delay surgery until symptoms develop? Ann Thorac Surg. 2010;90:481-488.
  26. Enriquez-Sarano M, Sundt TM. Early surgery is recommended for mitral regurgitation. Circulation. 2010;121:804-811; discussion 812.
  27. Stone GW, Lindenfeld J, Abraham WT, et al. Transcatheter mitral-valve repair in patients with heart failure. N Engl J Med. 2018;379:2307-2318.
  28. Obadia JF, Messika-Zeitoun D, Leurent G, et al. Percutaneous repair or medical treatment for secondary mitral regurgitation. N Engl J Med. 2018;379:2297-2306.
  29. Stone GW, Mack M, Abraham W, Lindenfeld J. COAPT: a randomized trial of transcatheter mitral valve leaflet approximation in patients with heart failure and secondary mitral regurgitation. Paper presented at: Transcatheter Cardiovascular Therapeutics. September 23, 2018.
  30. Toggweiler S, Zuber M, Sürder D, et al. Two-year outcomes after percutaneous mitral valve repair with the MitraClip system: durability of the procedure and predictors of outcome. Open Heart. 2014;1:e000056.
  31. Neuss M, Schau T, Isotani A, Pilz M, Schöpp M, Butter C. Elevated mitral valve pressure gradient after MitraClip implantation deteriorates long-term outcome in patients with severe mitral regurgitation and severe heart failure. JACC Cardiovasc Interv. 2017;10:931-939.