Error bars indicate SDs; vertical lines are the median values. GDMT indicates guideline-directed medical therapy; HR, hazard ratio; and TMVr, transcatheter mitral valve repair.
GDMT indicates guideline-directed medical therapy; TMVr, transcatheter mitral valve repair.
Error bars indicate SDs; vertical lines are the median values.
eTable 1. Baseline Characteristics of the LVEDVi and EROA Subgroups
eTable 2. Comparison of Predictive Value of EROA/LVEDV for ACM+ HFH at 12 and 24 Months in COAPT
eFigure 1. Comparison of All-Cause Mortality or Heart Failure (HF) Hospitalizations Between TMVr plus GDMT Compared With GDMT Alone in the 6 Subgroups Derived From EROA and LVEDVi.
eFigure 2. Comparison of All-Cause Mortality or Heart Failure (HF) Hospitalizations Between TMVr plus GDMT Compared With GDMT Alone in the 6 Subgroups Derived From RV and LVEDVi
eFigure 3. MR Severity at 30 Days for EROA and LVEDVi Subgroups
Data Sharing Statement
Customize your JAMA Network experience by selecting one or more topics from the list below.
Lindenfeld J, Abraham WT, Grayburn PA, et al. Association of Effective Regurgitation Orifice Area to Left Ventricular End-Diastolic Volume Ratio With Transcatheter Mitral Valve Repair Outcomes: A Secondary Analysis of the COAPT Trial. JAMA Cardiol. 2021;6(4):427–436. doi:10.1001/jamacardio.2020.7200
Does the ratio of effective regurgitant orifice area (EROA) to left ventricular end-diastolic volume (LVEDV) support the proportionate-disproportionate hypothesis for the disparate results of the Multicenter Study of Percutaneous Mitral Valve Repair MitraClip Device in Patients With Severe Secondary Mitral Regurgitation and Cardiovascular Outcomes Assessment of the MitraClip Percutaneous Therapy for Heart Failure Patients With Functional Mitral Regurgitation in patients with secondary mitral regurgitation?
In this secondary analysis of a randomized clinical trial of EROA and regurgitant volume relative to LVEDV, which involved a total of 548 patients, follow-up at 24 months did not strongly support the proportionate-disproportionate hypothesis in determining patient outcome.
The ratio of EROA to LVEDV may not be the best factor associated with the benefits of transcatheter mitral valve repair for all-cause mortality or hospitalization for heart failure.
Transcatheter mitral valve repair (TMVr) plus maximally tolerated guideline-directed medical therapy (GDMT) reduced heart failure (HF) hospitalizations (HFHs) and all-cause mortality (ACM) in symptomatic patients with HF and secondary mitral regurgitation (SMR) compared with GDMT alone in the Cardiovascular Outcomes Assessment of the MitraClip Percutaneous Therapy for Heart Failure Patients With Functional Mitral Regurgitation (COAPT) trial but not in a similar trial, Multicenter Study of Percutaneous Mitral Valve Repair MitraClip Device in Patients With Severe Secondary Mitral Regurgitation (MITRA-FR), possibly because the degree of SMR relative to the left ventricular end-diastolic volume index (LVEDVi) was substantially lower.
To explore contributions of the degree of SMR using the effective regurgitation orifice area (EROA), regurgitant volume (RV), and LVEDVi to the benefit of TMVr in the COAPT trial.
Design, Setting, and Participants
This post hoc secondary analysis of the COAPT randomized clinical trial performed December 27, 2012, to June 23, 2017, evaluated a subgroup of COAPT patients (group 1) with characteristics consistent with patients enrolled in MITRA-FR (n = 56) (HF with grade 3+ to 4+ SMR, left ventricular ejection fraction of 20%-50%, and New York Heart Association function class II-IV) compared with remaining (group 2) COAPT patients (n = 492) using the end point of ACM or HFH at 24 months, components of the primary end point, and quality of life (QOL) (per the Kansas City Cardiomyopathy Questionnaire overall summary score) and 6-minute walk distance (6MWD). The same end points were evaluated in 6 subgroups of COAPT by combinations of EROA and LVEDVi and of RV relative to LVEDVi.
Interventions were TMVr plus GDMT vs GDMT alone.
A total of 548 participants (mean [SD] age, 71.9 [11.2] years; 351 [64%] male) were included. In group 1, no significant difference was found in the composite rate of ACM or HFH between TMVr plus GDMT vs GDMT alone at 24 months (27.8% vs 33.1%, P = .83) compared with a significant difference at 24 months (31.5% vs 50.2%, P < .001) in group 2. However, patients randomized to receive TMVr vs those treated with GDMT alone had significantly greater improvement in QOL at 12 months (mean [SD] Kansas City Cardiomyopathy Questionnaire summary scores: group 1: 18.36 [5.38] vs 0.43 [4.00] points; P = .01; group 2: 16.54 [1.57] vs 5.78 [1.82] points; P < .001). Group 1 TMVr-randomized patients vs those treated with GDMT alone also had significantly greater improvement in 6MWD at 12 months (mean [SD] paired improvement: 39.0 [28.6] vs −48.0 [18.6] m; P = .02). Group 2 TMVr-randomized patients vs those treated with GDMT alone tended to have greater improvement in 6MWD at 12 months, but the difference did not reach statistical significance (mean [SD] paired improvement: 35.0 [7.7] vs 16.0 [9.1] m; P = .11).
Conclusions and Relevance
A small subgroup of COAPT-resembling patients enrolled in MITRA-FR did not achieve improvement in ACM or HFH at 24 months but had a significant benefit on patient-centered outcomes (eg, QOL and 6MWD). Further subgroup analyses with 24-month follow-up suggest that the benefit of TMVr is not fully supported by the proportionate-disproportionate hypothesis.
ClinicalTrials.gov Identifier: NCT01626079
In patients with heart failure (HF), the presence of secondary mitral regurgitation (SMR) is associated with increased mortality.1,2 However, whether correction of SMR has a benefit effect on mortality, HF hospitalizations (HFHs), or symptoms is unknown. Two randomized clinical trials, the Multicenter Study of Percutaneous Mitral Valve Repair MitraClip Device in Patients With Severe Secondary Mitral Regurgitation (MITRA-FR) and Cardiovascular Outcomes Assessment of the MitraClip Percutaneous Therapy for Heart Failure Patients With Functional Mitral Regurgitation (COAPT), recently compared the effect of transcatheter mitral valve repair (TMVr) with the MitraClip (Abbott) and medical therapy with medical therapy alone for symptomatic patients with HF, reduced ejection fraction, and SMR, reporting notably different results.3,4
In MITRA-FR, the primary efficacy end point was a composite of all-cause mortality (ACM) or HFH at 12 months. Neither the primary outcome nor ACM alone was different between the groups. The results of the COAPT trial were markedly different. The primary effectiveness end point of the annualized rate of HFH within 24 months of randomization was reduced with TMVr plus maximally tolerated guideline-directed medical therapy (GDMT) compared with GDMT alone (35.8% vs 67.9% per patient year, P < .001) as was ACM (29.1% vs 46.1%, P < .001). Speculation about the reasons for these different results has been intense.5-12 One important hypothesis is that patients with less severe SMR and more left ventricular (LV) enlargement and dysfunction (proportionate MR) have less benefit than patients with more severe MR and less LV remodeling (disproportionate MR)7,12 In MITRA-FR, the mean (SD) effective regurgitant orifice area (EROA) was 31 (10) mm2 and the mean (SD) LV end-diastolic volume index (LVEDVi) was 135 (37) mE/m2. In COAPT, the mean (SD) EROA was 41 (15) mm2 and the mean (SD) LVEDVi was 101 (34) mL/m2.
The objectives of this secondary analysis of the COAPT trial were to understand the reasons for the disparate outcomes of these trials by providing additional data about quality of life (QOL) and functional capacity in patients from COAPT who resemble patients enrolled in MITRA-FR and to extend previous analyses of the COAPT trial that address the proportionate-disproportionate hypothesis.12
This post hoc secondary analysis of the COAPT randomized clinical trial performed from December 27, 2012, to June 23, 2017, evaluated a subgroup of COAPT patients (group 1) with characteristics consistent with patients enrolled in MITRA-FR (n = 56) compared with the remainder (group 2) of the COAPT patients (n = 492) using the end point of ACM or HFH at 24 months, the components of the primary end point, as well as QOL and 6-minute walk distance (6MWD). Institutional review board approval was obtained at each site, and all patients provided written informed consent before trial participation. Data were deidentified. This study followed the Consolidated Standards of Reporting Trials (CONSORT) reporting guideline.
The study design for COAPT has been described previously.13 In brief, COAPT was a randomized, parallel-controlled, open-label, multicenter trial that evaluated TMVr with the MitraClip in symptomatic patients with HF with reduced ejection fraction and moderate to severe or severe SMR. Eligible patients had ischemic or nonischemic cardiomyopathy with an LV ejection fraction of 20% to 50%, had grade 3+ or 4+ SMRs confirmed by an echocardiographic core laboratory before enrollment, and remained symptomatic (New York Heart Association functional class II-IVa [ambulatory]) despite a stable maximally tolerated GDMT regimen and cardiac resynchronization therapy if appropriate per societal guidelines.4
We hypothesized that a subgroup of patients within COAPT who resemble MITRA-FR patients (ie, with relatively smaller EROAs and larger ventricles) may experience a lesser degree of benefit from TMVr. To examine this hypothesis, we dichotomized the EROA based on the mean EROA in MITRA-FR (<0.31 cm2) and the LVEDVi based on the median value observed in COAPT (96 mL/m2). Sixty-six patients with missing baseline EROA or LVEDVi data were excluded. Group 1 consisted of 56 MITRA-FR–like patients: those with an EROA of 0.30 cm2 or smaller and an LVEDVi greater than 96 mL/m2 of whom 22 were in the TMVr arm and 34 in the control GDMT-only arm. Group 2 included all 492 other patients: those with an EROA greater than 0.30 cm2 and/or an LVEDVi of 96 mL/m2 or less.
We examined the composite of ACM or HFH at 24 months for the 2 groups as well as each component measure between the TMVr plus GDMT and GDMT alone groups, stratified by groups. In addition, the results of TMVr plus GDMT and GDMT alone on the end points of MR severity, QOL (assessed using the Kansas City Cardiomyopathy Questionnaire [KCCQ]), and functional capacity (assessed by 6MWD) were also compared in groups 1 and 2.
Two additional analyses14,15 were performed by creating 6 subgroups of the COAPT cohort and examining the the association of the end point of ACM or HFH at 24 months by subgroup. The first analysis14 defined the 6 subgroups by EROA cutoffs typically used to distinguish moderate, moderate to severe, and severe MR (≤0.3, >0.3 to <0.4, and ≥0.4 cm2) and LVEDVi (≤96 mL/m2 or >96 mL/m2), and the second analysis15 used 6 subgroups defined by (RV) based on similar cutoffs using the guidelines of the American Society of Echocardiography (<45 regurgitant volume, 45-60, and >60 mL/beat) and LVEDVi (≤96 mL/m2 or >96 mL/m2).15 Regurgitant volume in COAPT was determined using the proximal isovelocity hemispheric surface area method.16
For time to first event analyses, estimated event rates were compared with Cox proportional hazards regression models. Interaction testing was performed to determine whether the relative impact of TMVr vs GDMT was different across the 2 groups. Categorical variables were compared with the Fisher exact test. The Bowker test was used to compare pairwise changes in categorical variables over time. Continuous variables were compared with 2-tailed, paired t tests or the Wilcoxon rank sum test for nonnormally distributed data. Analysis of covariance was used to compare mean changes in continuous outcome measures from baseline to follow-up between groups. A 2-sided P < .05 was considered statistically significant. All statistical analyses were performed with SAS software, version 9.4 (SAS Institute Inc).
A total of 548 participants (mean [SD] age, 71.9 [11.2] years; 351 [64%] male) were included in the study. The Table compares the baseline characteristics between patients in group 1 (EROA≤0.30 cm2 and LVEDVi >96 mL/m2) and group 2 (EROA>0.30 cm2 and/or LVEDVi ≤96 mL/m2). Patients in group 1 tended to be younger, had a lower Society for Thoracic Surgery replacement score, and were less likely to have atrial fibrillation (AF). By the group definition, EROA was smaller and LV dimensions were larger in group 1 compared with group 2. A smaller percentage of patients in group 1 had severe MR (4+). Group 1 closely resembled the entire MITRA-FR study cohort in terms of EROA (mean [SD] EROA, 0.26 [0.04] cm2 in COAPT vs approximately 0.31 [0.10] cm2 in MITRA-FR) and LVEDVi (mean [SD] EROA, 127.6 [29.8] mL/m2 vs approximately 135  mL/m2).
As shown in Figure 1, group 1 patients (n = 56) had no significant difference at 24 months in ACM or HFH between TMVr plus GDMT vs GDMT alone (27.8% vs 55.5%; hazard ratio [HR], 0.53; 95% CI, 0.21-1.35; P = .18), and there was no difference in ACM. In group 2 (n = 492), ACM or HFH (44.3% vs 70.0%; HR, 0.48; 95% CI, 0.38-0.61; P < .001) and ACM (27.3% vs 47.9%; HR, 0.51; 95% CI, 0.37-0.60; P < .001) were significantly reduced with TMVr compared with GDMT alone at 24 months.
As shown in Figure 2, group 1 TMVr-randomized patients experienced a significant improvement in MR compared with GDMT alone. At 30 days, MR grade was reduced to 2+ or less in 95% of patients with TMVr plus GDMT compared with 45% with GDMT alone (P < .001). These results were sustained through 12 months. Similarly, TMVr-randomized patients in group 2 experienced greater improvement in MR severity compared with GDMT. At 30 days, MR grade was reduced to 2+ or less in 95% of patients with TMVr plus GDMT vs 33% with GDMT alone (P < .001). These results were sustained through 12 months.
Patients randomized to receive TMVr vs those treated with GDMT alone in group 1 had significantly greater improvement in mean (SD) KCCQ overall summary scores at 12 months (mean [SD] paired improvement: 18.36 [5.38] vs 0.43 [4.00] points; P = .01) (Figure 3). A similar benefit of TMVr occurred in group 2 patients at 12 months (mean [SD] paired improvement: 16.54 [1.57] vs 5.78 [1.82] points; P < .001).
As shown in Figure 4, group 1 TMVr-randomized patients vs those treated with GDMT alone had significantly greater improvement in 6MWD at 12 months (mean [SD] paired improvement: 39.0 [28.6] vs approximately −48.0 [18.6] m; P = .02). Group 2 TMVr-randomized patients vs those treated with GDMT alone tended to have greater improvement in 6MWD at 12 months, although the difference did not reach statistical significance (mean [SD] paired improvement: 35.0 [7.7] m vs 16.0 [9.1] m,; P = .11).
To further identify subgroups derived from EROA and LVEDVi in which TMVr response may be limited, group 2 patients were subdivided into 5 subgroups based on EROA cutoffs of 0.30 cm2 or less, 0.30 to 0.40 cm2, and 0.40 cm2 or more and an LVEDVi cutoff of 96 mL/m2. Baseline differences among the 6 subgroups were noted in age, percentage of women, and extremely high risk for mitral valve surgery (eTable 1 in Supplement 1). The findings for the composite end point of ACM or HFH at 24 months for the 5 COAPT-like subgroups are given in eFigure 1 in Supplement 1. We also compared the outcome of ACM or HFH for TMVr plus GDMT vs GDMT alone by RV cutoffs of less than 45, 45 to 60, and greater than 60 mL/beat and an LVEDVi cutoff of 96 mL/m2 (eFigure 2 in Supplement 1). In COAPT, the mean (SD) RV as determined by the proximal isovelocity hemispheric surface area method was 59.7 (21) mL in the TMVr group and 59.9 (23.5) mL in the GDMT group. Compared with the EROA and LVEDVi analysis, this analysis found no differerence between TMVr plus GDMT vs GDMT alone for the MITRA-FR–like subgroup, but each of the other 5 subgroups had reductions in ACM plus HFH that favored TMVr plus GDMT vs GDMT alone (eFigure 2 in Supplement 1).
To further assess the contributions of EROA and LVEDVi on initial and long-term reduction in MR, we analyzed MR severity (grade of ≤2+) at 30 days for the EROA subgroups arranged in increasing order of effectiveness for both the TMVr plus GDMT and the GDMT alone arms (eFigure 3 in Supplement 1). Although significant MR reduction was achieved across all subgroups with TMVr, the proportion of patients with an MR grade of 2+ or higher was lower in patients with LVEDVi greater than 96 mL/m2 compared with those with LVEDVi of 96 mL/m2 or less.
The different outcomes of TMVr treatment of SMR reported for the MITRA-FR and COAPT trials has led to a flurry of discussion about the reasons for this difference, but substantial controversy remains.5-12,17,18 Two of the most observable differences between the trials were the smaller mean EROA and larger mean LVEDVi in patients enrolled in the MITRA-FR trial.5 Grayburn et al7 and Packer and Grayburn12 attempted to reconcile the differences between MITRA-FR and COAPT by introducing the concept of disproportionate vs proportionate SMR, suggesting that the EROA to LVEDV ratio identifies proportionate and disproportionate MR and that those with proportionate MR will not benefit from TMVr compared with GDMT alone. In this analysis of the COAPT cohort, TMVr had no benefit on the composite rate of ACM or HFH at 24 months in the subgroup of COAPT patients who were similar to MITRA-FR patients, as was observed in the MITRA-FR study.3 The reduction of SMR to grade 2+ or less at 30 days was similar with TMVr in both group 1 (MITRA-FR–like) and group 2 (COAPT-like) patients. However, in the MITRA-FR–like subgroup of patients in this analysis, TMVr plus GDMT resulted in significant improvements in QOL and 6MWD at 12 months compared with GDMT alone, which was quantitatively similar to the COAPT-like cohort (EROA>0.3 cm2 and/or LVEDVi ≤96 mL/m2). Patient-related outcomes, and specifically the KCCQ, have demonstrated reproducible and independent associations with mortality and HFH.19-22 The incremental KCCQ improvement of TMVr on background GDMT in both groups 1 and 2 were of similar magnitude and larger than the 5-point improvement that has been associated with improvements in HFH and mortality.23 The 6MWD also improved in both groups, although the improvement was only significant in the MITRA-FR–like group. Why TMVr-treated patients in group 1 had improved QOL and 6MWD but not reduced death or HFHs is uncertain. In MITRA-FR, reported outcomes for the 6MWD favored the TMVr group at both 12 and 24 months,3,24 and similar trends were seen for the global QOL score at 12 months.3 It is possible that there may be a benefit in patient-centered outcomes even when ACM or HFH are not reduced at 1 year as seen in patients in group 1. The KCCQ score includes a number of questions specifically related to dyspnea, a sensitive marker for reduced left atrial pressure that results from reduction in MR, providing a reason why QOL and 6MWD may be improved without changes in ACM or HFH.25,26 The improvements in the KCCQ score and 6MWD in group 1 may also be a function of the small numbers in this group, although both improvements were statistically significant.
In our small subgroup of patients similar to those enrolled in MITRA-FR, ACM and HFH were not reduced at 24 months, as in MITRA-FR. The finding of improved ACM or HFH in this subgroup that emerged between 12 and 24 months was also reported in MITRA-FR at 24 months.24 In the MITRA-FR 2-year analyses, there was a divergence in the incidence of recurrent HFH between 1 and 2 years, with the landmark analyses demonstrating a statistically meaningful improvement (HR, 0.47; 95% CI, 0.22-0.98). These data suggest that there may be a delayed benefit in HFH or ACM in a MITRA-FR–like population. Although these post hoc data should be considered exploratory, they arose from a hypothesis driven by the MITRA-FR results and thus do provide additional evidence for determining patient selection for treatment of SMR.
A history of AF was present in approximately 30% of patients in MITRA-FR compared with nearly 60% in COAPT, which is similar to the prevalence in our 2 subgroups.3,4 This difference does not explain the divergent results of the 2 trials because AF does not alter the relative benefit of TMVr.27 However, it is possible that the higher prevalence of AF in COAPT contributed to the higher natriuretic peptide levels compared with MITRA-FR.28 Furthermore, the more severe SMR in COAPT (vs MITRA-FR) may be responsible the the higher prevalence of AF in COAPT.
In a recent publication,12 the COAPT cohort was analyzed using 6 EROA and LVEDVi subgroups identical to those described in this study. In that analysis,12 the EROA to LVEDVi ratio did appear to correlate with 12-month ACM or HFH, supporting the proportionate-disproportionate hypothesis. However, in the EROA groups of 0.3 to 0.4 cm2 and 0.4 cm2 or greater, a comparable reduction in death or HFH with TMVr was observed whether the mean LVEDVi was 96 mL/m2 or less or greater than 96 mL/m2. These data indicate that the association between EROA and LVEDV on the benefits of TMVr are complex. In the current analysis, we extend those findings in the 6 subgroups to 24 months, with several new observations. First, although small, the subgroup most closely resembling the MITRA-FR cohort suggests benefit of TMVr at 24 months similar to MITRA-FR results at 2 years.24 Second, when the analysis is extended to 24 months, the EROA to LVEDV ratio no longer strongly supports the proportionate-disporportionate hypothesis, which is similar to the conclusions of a similar analysis of 2-year results in the MITRA-FR trial12,28 (eTable 2 in Supplement 1). Third, when the association of RV and LVEDV is examined by dividing the COAPT cohort into 6 subgroups and extending follow-up to 24 months, the HR for a reduction in ACM and HFH is similar in all the subgroupse although not statistically significant for the MITRA-FR subgroup. Gaasch et al17,29,30 have suggested that the RV-LVEDV association may be much more predictive of outcomes in SMR than the EROA-LVEDV association, and our analysis suggest it may be at least as good.
The RV we report in patients in COAPT is much higher than the RVs in SMR suggested by Gaasch et al17 but is similar to the RV recently reported for patients in MITRA-FR.31 The proximal isovelocity hemispheric surface area method was used to determine the RV in COAPT, and it measures peak EROA, not mean EROA, thus overestimating RV. The Simpson method was used for measuring LV volumes, and it is difficult to compare volumes measured by different methods.32,33 Furthermore, the forward stroke volume and RV suggested by Gaasch et al17 in studies of SMR, including MITRA-FR and COAPT, likely underestimates the true RV in most patients with SMR.17 Using their estimates of forward stroke volume for both MITRA-FR and COAPT, most patients enrolled in these studies would have had cardiac outputs less than 2 L/min.
We also present data suggesting that the success in reducing SMR to grade 2+ or less with TMVr decreases as the size of the left ventricle increases. This finding is consistent with surgical data indicating that the recurrence of significant SMR after surgical MVR increases as the size of the ventricle increases.13,34 This finding might be an explanation for the poorer results in reducing SMR at 30 days (12%) and 1 year (17%) in MITRA-FR vs the lower rates in COAPT at 30 days (5%) and 1 year (5%). The reduction in SMR at 2 years in MITRA-FR has not been reported but remains at 5% for COAPT at 2 years. Indeed, data from COAPT have reported that reduction in SMR to grade 2+ or lower is a critical factor associated with outcomes with both TMVr and medical therapy (S.K., written communication, 2020).
The major limitations of this study are its post hoc nature and the small size of the MITRA-FR–like cohort and all the 5 other subgroups. Although most baseline variables are similar, some differences were found between groups that may influence the results. All patients enrolled in COAPT were qualified by the echocardiographic core laboratory as having true grade 3+ or 4+ SMR using an integrative multiparameter approach. In contrast, only a single echocardiographic parameter would qualify patients for MITRA-FR. Thus, the COAPT group 1 patients may be different from the MITRA-FR patients. An analysis of the combined MITRA-FR and COAPT databases is planned and will further clarify the factors associated with TMVr response in patients with SMR. Additional studies with larger cohorts and longer follow-up are also necessary to determine whether patients with less severe MR or greater LV dimensions may have a more modest clinical benefit.
In this exploratory, secondary analysis from the COAPT trial, a small subgroup of patients who most closely resembled patients enrolled in MITRA-FR did not achieve a significant improvement in ACM or HFH at 24 months but had improved QoL and functional capacity. Several analyses suggest that the proportionate-disproportionate hypothesis may not be enough to explain the divergent results of MITRA-FR and COAPT.
Accepted for Publication: November 30, 2020.
Published Online: February 3, 2021. doi:10.1001/jamacardio.2020.7200
Corresponding Author: JoAnn Lindenfeld, MD, Advanced Heart Failure, Vanderbilt Heart and Vascular Institute, 1215 21st Ave S, Ste 5309, Nashville, TN 37232 (firstname.lastname@example.org).
Author Contributions: Dr Stone had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Lindenfeld, Abraham, Grayburn, Lim, Sundareswaran, Mack, Weissman, Stone.
Acquisition, analysis, or interpretation of data: Lindenfeld, Abraham, Kar, Asch, Lim, Nie, Singhal, Sundareswaran, Weissman, Mack.
Drafting of the manuscript: Lindenfeld, Kar, Nie, Singhal, Sundareswaran.
Critical revision of the manuscript for important intellectual content: Lindenfeld, Abraham, Grayburn, Kar, Asch, Lim, Sundareswaran, Weissman, Mack, Stone.
Statistical analysis: Lindenfeld, Nie, Sundareswaran.
Obtained funding: Weissman.
Administrative, technical, or material support: Lindenfeld, Grayburn, Kar, Asch, Sundareswaran, Weissman.
Supervision: Lindenfeld, Abraham, Grayburn, Asch, Lim, Mack, Stone.
Conflict of Interest Disclosures: Dr Lindenfeld reported receiving personal fees from Abbott, Boehringer Ingelheim, CVRx, Impulse Dynamics, Edwards Lifesciences, and VWave; grants and personal fees from AstraZeneca; and grants from Sensible Medical and Volumetrix during the conduct of the study and personal fees from Abbott, Boehringer Ingelheim, CVRx, Impulse Dynamics, Edwards Lifesciences, and VWave; grants and personal fees from Astra Zeneca; and grants from Sensible Medical and Volumetrix outside the submitted work. Dr Abraham reported receiving personal fees from Abbott during the conduct of the study and personal fees from Edwards Lifesciences outside the submitted work. Dr Grayburn reported receiving grants and personal fees from Abbott Vascular and Edwards Lifesciences, echocardiography core laboratory contracts from Cardiovalve and Neochord, grants from Medtronic, personal fees from 4C Medical, and personal fees for serving as a consultant and an imaging core laboratory from W. L. Gore during the conduct of the study. Dr Kar reported receiving personal fees from Abbott, Boston Scientific, Medtronic, and WL Gore during the conduct of the study; serving as the principal investigator of the REPAIR MR (Percutaneous MitraClip Device or Surgical Mitral Valve Repair in Patients With Primary Mitral Regurgitation Who Are Candidates for Surgery) trial; and serving as a member of the steering committee for the Triluminate Trial. Dr Asch reported receiving grants from Abbott during the conduct of the study and grants from Edwards, Boston Scientific, Medtronic, MVRx, Livanova, Ancora GDS, Neovasc, Innovheart, and Polares Medical outside the submitted work. Dr Lim reported receiving grants from Abbott Vascular during the conduct of the study. Dr Mack reported receiving nonfinancial support from Abbott during the conduct of the study and nonfinancial support from Edwards Lifesciences and Medtronic outside the submitted work. Dr Weissman reported receiving grants from Abbott during the conduct of the study. Dr Stone reported receiving grants from Cardiovascular Research Foundation for trial-related services during the conduct of the study and personal fees from Terumo, Cook, Miracor, Neovasc, V-wave, Abiomed, MAIA Pharmaceuticals, Shockwave, Vectorious, Cardiomech, Applied Therapeutics, MedFocus, Biostar, Spectrawave, Valfix, Ancora, Orchestra Biomed, and Qool Therapeutics and equity/options from Aria, Cagent, Cardiac Success, Spectrawave, Valfix, Ancora, Orchestra Biomed, and Qool Therapeutics outside the submitted work. No other disclosures were reported.
Funding/Support: The Cardiovascular Outcomes Assessment of the MitraClip Percutaneous Therapy for Heart Failure Patients With Functional Mitral Regurgitation (COAPT) trial was sponsored by Abbott.
Role of the Funder/Sponsor: Abbott had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication. Abbott performed the statistical analysis.
Group Information: The COAPT investigators are as follows: Saibal Kar, MD, Smidt Heart Institute at Cedars-Sinai Medical Center, Los Angeles, California; Jamie Kennedy, MD, University of Virginia, Charlottesville; Jacob Mishell, MD, Kaiser Permanente–San Francisco Hospital, San Francisco, California; Brian Whisenant, MD, Intermountain Heart Center, Salt Lake City, Utah; Paul Grayburn, MD, Baylor University Medical Center, Baylor Heart and Vascular Institute, Dallas, Texas; Andreas Brieke, MD, University of Colorado Hospital, Aurora; Michael Rinaldi, MD, Sanger Heart & Vascular Institute/Atrium Health, Charlotte, North Carolina; Samir Kapadia, MD, Cleveland Clinic, Cleveland, Ohio; Ian Sarembock MB, ChB, MD, The Christ Hospital, Cincinnati, Ohio; Vivek Rajagopal, MD, Piedmont Hospital, Atlanta, Georgia; Robert Kipperman, MD, Gagnon Cardiovascular Institute, Morristown Medical Center, Morristown, New Jersey; Konstantinos Boudoulas, MD, The Ohio State University Wexner Medical Center, Columbus; Mubashir Mumtaz, MD, Pinnacle Health at Harrisburg Hospital, Harrisburg, Pennsylvania; Adam Greenbaum, MD, Henry Ford Hospital, Detroit, Michigan; Jason Rogers, MD, UC Davis Medical Center, Davis, California; Mark Riccardi, MD, Northwestern Memorial Hospital, Chicago, Illinois; Guy Reeder, MD, Mayo Clinic, Rochester, Minnesota; Alan Yeung, MD, Stanford University School of Medicine, Stanford, California; Neal Stephen Kleiman, MD, Houston Methodist Hospital, Houston, Texas; Igor Palacios, MD, Massachusetts General Hospital, Harvard Medical School, Boston; Evelio Rodriguez, MD, St. Thomas Hospital, Nashville, Tennessee; Anita Asgar, MD, Montreal Heart Institute, Montreal, Quebec, Canada; John Cahill, MD, Vidant Medical Center, Greenville, North Carolina; Matthew Price, MD, Scripps Clinic, La Jolla, California; Vasilis Babaliaros, MD, Emory University Hospital, Atlanta, Georgia; James Hermiller, MD, St. Vincent Hospital/The Heart Center of Indiana, Indianapolis, Indiannapolis; Chad Rammohan, MD, El Camino Hospital; Nirat Beohar, MD, Surgeons, New York, New York; Bethany Austin, MD, St. Luke’s Hospital of Kansas City, Kansas City, Missouri; Firas Zahr, MD, Oregon Health and Science University, Portland; Mahammad Ghani, MD, Oklahoma Heart Hospital, Oklahoma City, Oklahoma; Shamir Mehta, MD, Hamilton Health Sciences, Hamilton, Ontario, Canada; John Lasala, MD, PhD, Washington University School of Medicine/Barnes-Jewish Hospital, St Louis, Missouri; Ted Feldman, MD, Evanston Hospital, Cardiology Division, Evanston, Illinois; Mario Goessl, MD, Mayo Clinic College of Medicine, Rochester, Minnesota; Robert Boone, MD, St. Paul's Hospital, University of British Columbia, Vancouver, British Columbia, Canada; Andrew Wang, MD, Duke University School of Medicine, Durham, North Carolina; Ethan Korngold, MD, Providence St. Vincent Medical Center, Portland, Oregon; Anson Jay Conrad Smith, MD, UPMC Presbyterian, Pittsburgh, Pennsylvania; Aziz Maksoud, MD, Ascension Via Christi, Wichita, Kansas; Garric Stewart, MD, Brigham and Women's Hospital, Boston, Massachusetts; Wayne Batchelor MD, MHS, Tallahassee Research Institute Inc, Tallahassee, Florida; Daniel Steinberg, MD, Medical University of South Carolina, Charleston, South Carolina; Sushell Kodali, MD, Columbia University Medical Center/New York Presbyterian Hospital, New York, New York; Clifford Kavinsky, MD, Rush University Medical Center, Chicago, Illinois; Mack Pirwitz, MD, Seton Medical Center Austin, Austin, Texas; Cezar Staniloae, MD, NYU Langone Medical Center, New York, New York; Gregory Helmer, MD, University of Minnesota, Minneapolis; Mark Vesely, MD, University of Maryland Medical Center, Baltimore; Andrew Berke, MD, St. Francis Hospital, Roslyn, New York; Lowell Satler, MD, MedStar Health Research Institute/Washington Hospital Center, Washington, DC; Steven Yakubov, MD, OhioHealth Riverside Methodist Hospital, Columbus; Jason Foerst, MD, Carilion Roanoke Memorial Hospital, Roanoke, Virginia; Mark Wiley, MD, Kansas City, Kansas; Shing-Chiu Wong, MD, Cornell Medical College, New York, New York; Robert Salley, MD, St. Joseph Hospital, Lexington, Kentucky; James Jenkins, MD, Ochsner Clinic Foundation, New Orleans, Louisiana; Raymond McKay, MD, Hartford Hospital, Hartford, Connecticut; George Hanzel, MD, William Beaumont Hospital, Royal Oak, Michigan; Joseph Wight, MD, Maine Medical Center, Scarborough; Navin Kapur, MD, Tufts Medical Center, Boston, Massachusetts; Massoud Leesar, MD, The University of Alabama at Birmingham; John Petersen MD, MHS, Duke University Medical Center, Durham, North Carolina; Howard Herrmann, MD, Philadelphia, Pennsylvania; Lawrence Ong, MD, North Shore University Hospital, Huntington, New York; David McAllister DO Iowa Heart Center PC, Des Moines; Neil Fam, MD, Division of Cardiology, St. Michael's Hospital, Toronto, Ontario, Canada; Christian Spies, MD, The Queen's Medical Center, Honolulu, Hawaii; Brian O’Neill, MD, Temple University School of Medicine, Philadelphia, Pennsylvania; Samin Sharma, MD, Mount Sinai School of Medicine, New York, New York; Richard Smallling MD, PhD, University of Texas Health Science Center at Houston; Zachary Gertz, MD, University of Pennsylvania, Philadelphia; Timothy Byrne, DO, Banner-University Medical Center Phoenix, Phoenix, Arizona; Omar Nass, MD, Nebraska Heart Institute Heart Hospital, Lincoln; Robert Welsh, MD, Division of Cardiology, University of Alberta, Edmonton, Alberta, Canada; Louis Heller, MD, Westchester Medical Center Health Network, Westchester, New York; John Sullebarger, MD, Tampa General Hospital, Tampa, Florida; James Maddux, MD, Providence St. Vincent Medical Center, Portland, Oregon.
Data Sharing Statement: See Supplement 2.