CMS indicates Centers for Medicare & Medicaid Services; pHTN, pulmonary hypertension; and TVT, Transcatheter Valve Therapy registry. Group 1 includes patients with mean pulmonary artery pressure (mPAP) of less than 25 mm Hg; group 2, mPAP of 25 to 34 mm Hg; group 3, mPAP of 35 to 44 mm Hg; and group 4, mPAP of at least 45 mm Hg.
Kaplan-Meier curves depict increased rates of the composite outcome of mortality and heart failure readmissions (A), all-cause mortality (B), and readmissions for heart failure (C) at 1 year in patients with more severe pulmonary hypertension. Group 1 includes patients with mean pulmonary artery pressure (mPAP) of less than 25 mm Hg; group 2, mPAP of 25 to 34 mm Hg; group 3, mPAP of 35 to 44 mm Hg; and group 4, mPAP of at least 45 mm Hg.
eTable 1. Baseline and 30-Day TVT Outcomes in the CMS-Linked vs Non–CMS-Linked Cohorts
eTable 2. Baseline and 30-Day TVT Outcomes in the Missing PAP vs Study Cohort
eTable 3. Procedural Details and Initial In-hospital Outcomes
eTable 4. 1-Year Mortality and Heart Failure Outcomes Using the TVT/CMS Linked Data
eTable 5. Unadjusted and Adjusted Hazard Ratios for 30-Day and 1-Year Mortality (TVT/CMS Linked Data)
eFigure 1. Mortality and Heart Failure After MitraClip in Patients With Primary Mitral Regurgitation
eFigure 2. Mortality and Heart Failure After MitraClip in Patients With Secondary Mitral Regurgitation
Customize your JAMA Network experience by selecting one or more topics from the list below.
Al-Bawardy R, Vemulapalli S, Thourani VH, et al. Association of Pulmonary Hypertension With Clinical Outcomes of Transcatheter Mitral Valve Repair. JAMA Cardiol. 2020;5(1):47–56. doi:10.1001/jamacardio.2019.4428
What is the association of pulmonary hypertension with clinical outcomes in patients undergoing transcatheter mitral valve repair?
In this cohort study of 4071 patients in the Society of Thoracic Surgery/American College of Cardiology Transcatheter Valve Therapy registry, pulmonary hypertension was common and associated with increased mortality and readmissions for heart failure after transcatheter mitral valve repair for severe mitral regurgitation. However, even in patients with severe pulmonary hypertension, transcatheter mitral valve repair was safe and effective and resulted in improved functional capacity.
These findings emphasize the adverse association of pulmonary hypertension with clinical outcomes after transcatheter mitral valve repair for mitral regurgitation and the need for efforts to determine whether earlier intervention before pulmonary hypertension develops will improve clinical outcomes.
Pulmonary hypertension (pHTN) is associated with increased risk of mortality after mitral valve surgery for mitral regurgitation. However, its association with clinical outcomes in patients undergoing transcatheter mitral valve repair (TMVr) with a commercially available system (MitraClip) is unknown.
To assess the association of pHTN with readmissions for heart failure and 1-year all-cause mortality after TMVr.
Design, Setting, and Participants
This retrospective cohort study analyzed 4071 patients who underwent TMVr with the MitraClip system from November 4, 2013, through March 31, 2017, across 232 US sites in the Society of Thoracic Surgery/American College of Cardiology Transcatheter Valve Therapy registry. Patients were stratified into the following 4 groups based on invasive mean pulmonary arterial pressure (mPAP): 1103 with no pHTN (mPAP, <25 mm Hg [group 1]); 1399 with mild pHTN (mPAP, 25-34 mm Hg [group 2]); 1011 with moderate pHTN (mPAP, 35-44 mm Hg [group 3]); and 558 with severe pHTN (mPAP, ≥45 mm Hg [group 4]). Data were analyzed from November 4, 2013, through March 31, 2017.
Patients were stratified into groups before TMVr, and clinical outcomes were assessed at 1 year after intervention.
Main Outcomes and Measures
Primary end point was a composite of 1-year mortality and readmissions for heart failure. Secondary end points were 30-day and 1-year mortality and readmissions for heart failure. Linkage to Centers for Medicare & Medicaid Services administrative claims was performed to assess 1-year outcomes in 2381 patients.
Among the 4071 patients included in the analysis, the median age was 81 years (interquartile range, 73-86 years); 1885 (46.3%) were women and 2186 (53.7%) were men. The composite rate of 1-year mortality and readmissions for heart failure was 33.6% (95% CI, 31.6%-35.7%), which was higher in those with pHTN (27.8% [95% CI, 24.2%-31.5%] in group 1, 32.4% [95% CI, 29.0%-35.8%] in group 2, 36.0% [95% CI, 31.8%-40.2%] in group 3, and 45.2% [95% CI, 39.1%-51.0%] in group 4; P < .001). Similarly, 1-year mortality (16.3% [95% CI, 13.4%-19.5%] in group 1, 19.8% [95% CI, 17.0%-22.8%] in group 2, 22.4% [95% CI, 18.8%-26.1%] in group 3, and 27.8% [95% CI, 22.6%-33.3%] in group 4; P < .001) increased across pHTN groups. The association of pHTN with mortality persisted despite multivariable adjustment (hazard ratio per 5-mm Hg mPAP increase, 1.05; 95% CI, 1.01-1.09; P = .02).
Conclusions and Relevance
These findings suggest that pHTN is associated with increased mortality and readmission for heart failure in patients undergoing TMVr using the MitraClip system for severe mitral regurgitation. Further efforts are needed to determine whether earlier intervention before pHTN develops will improve clinical outcomes.
Pulmonary hypertension (pHTN) often results from long-standing left-sided valvular heart disease and affects 15% to 32% of patients undergoing mitral valve surgery for mitral regurgitation.1 Pulmonary hypertension contributes to decisions regarding candidacy for mitral valve surgery and is a known risk factor for short- and long-term mortality after mitral valve surgery.2-6 Transcatheter mitral valve repair (TMVr) using a commercially available system (MitraClip system; Abbot Vascular, Inc) has emerged as an effective therapeutic intervention for patients with primary mitral regurgitation who are perceived to be at high risk for surgical mitral valve repair or replacement,7 and recent data suggest that TMVr is similarly effective in selected patients with secondary mitral regurgitation.8 Although TMVr is known to improve pulmonary artery pressures (PAP) and right ventricular performance,9 data on the association of preprocedure pHTN with clinical outcomes after TMVr for mitral regurgitation are limited.
We sought to assess the association of pHTN with clinical outcomes after TMVr using the MitraClip system within the Society of Thoracic Surgery/American College of Cardiology (STS/ACC) Transcatheter Valve Therapy (TVT) registry. Specifically, we evaluated the association of pHTN severity with the composite outcome of mortality and readmissions for heart failure at 1 year.
The STS/ACC TVT registry was launched in 2011 as a joint effort between the ACC and the STS to monitor safety and clinical outcomes of emerging transcatheter valve replacement and repair procedures. Described in detail previously,10 the registry satisfies the Centers for Medicare & Medicaid Services (CMS) national coverage determinations. The STS/ACC TVT registry has captured data on TMVr using the MitraClip system within the United States since its approval by the US Food and Drug Administration on October 24, 2013, for symptomatic severe (grades ≥3) primary mitral regurgitation in patients with prohibitive surgical risks (https://www.accessdata.fda.gov/scripts/cdrh/cfdocs/cfpma/pma.cfm?id=P100009). The details of the mitral repair procedure have been previously described.11 The CMS has linked the STS/ACC TVT registry to CMS administrative claims using direct patient identifiers. Registry activities have been approved by a central institutional review board. The Duke University School of Medicine institutional review board granted a waiver of informed consent, which was not feasible for this registry study. This cohort study followed the Strengthening the Reporting of Observational Studies in Epidemiology (STROBE) reporting guideline.
The STS/ACC TVT registry has been linked by the CMS to Medicare administrative claims data via direct patient identifiers to evaluate long-term clinical outcomes, including hospitalizations and survival.10 Data quality checks are implemented at the National Cardiovascular Data Registry data warehouse and Duke Clinical Research Institute, Durham, North Carolina, to optimize data completeness and accuracy. In addition, random 10% remote or on-site audits of STS/ACC TVT data are conducted annually by an independent third party.
We identified 9263 patients within the STS/ACC TVT registry who underwent commercial TMVr using the MitraClip system from November 4, 2013, until March 31, 2017, for symptomatic severe mitral regurgitation (Figure 1). Patients with missing preprocedure invasive mean PAP (mPAP) (n = 4683), cardiogenic shock (n = 16), or cardiac arrest (n = 6) and those undergoing emergent, urgent, or salvage procedures (n = 487) were excluded. The remaining 4071 patients across 232 sites made up the primary study cohort and were used for determination of in-hospital and 30-day outcomes. For clinical events after hospital discharge, 2381 patient records were linked to CMS claims data as previously described.11
Baseline characteristics plus procedural, in-hospital, and 30-day outcomes were collected within the STS/ACC TVT registry. Thirty-day outcomes, including mortality, stroke, myocardial infarction, bleeding complications, acute kidney injury, and vascular complications, were defined as per Valve Academic Research Consortium 2 criteria.12 Clinical outcomes occurring beyond the 30-day point were collected from CMS administrative claims data.11 Rehospitalization events were determined from CMS administrative claims data using the following codes from the International Classification of Diseases, Ninth Revision, Clinical Modification: 398.x, 402.x1, 404.x1, 404.x3, and 428.x for rehospitalization for heart failure. Our primary end point was a composite of 1-year mortality and readmissions for heart failure. Secondary end points were 30-day and 1-year mortality and readmissions for heart failure.
Our study cohort was divided into the following 4 groups based on severity of pHTN, measured by invasive right heart catheterization performed immediately or any time before the procedure: 1103 with no pHTN (mPAP, <25 mm Hg [group 1]); 1399 with mild pHTN (mPAP, 25-34 mm Hg [group 2]); 1011 with moderate pHTN (mPAP, 35-44 mm Hg [group 3]); and 558 with severe pHTN (mPAP, ≥45 mm Hg [group 4]).1,13 Among the 2381 patients in the CMS-linked cohort, 669 were in group 1, 837 in group 2, 561 in group 3, and 314 in group 4 (Figure 1). Baseline characteristics, treatment profiles, procedure details, and clinical outcomes were compared across groups. Continuous variables were presented as medians with interquartile range (IQR) percentiles, and categorical variables were expressed as frequencies with percentages. To test for independence of patient baseline characteristics, in-hospital care patterns, and outcomes with respect to the 4 mPAP groups, we used the Kruskal-Wallis test for continuous variables and Pearson χ2 test for categorical variables. The primary outcomes of interest were death, rehospitalization for heart failure, and the composite end point of death or rehospitalization for heart failure within 1 year. The Kaplan-Meier method was used for 1-year mortality and composite end point rates, and the incidence rate was attained from the Fine-Gray subdistribution hazard model for the heart failure, thus taking the competing risk of death into account. Unadjusted and adjusted Cox proportional hazards regression models were generated using robust standard errors. This procedure addresses the issue that patients from the same centers are more likely to respond in a similar fashion (within-site clustering). Proportional hazards regression assumption was met, and testing for linearity between risk factors and each end point of interest was performed. The multivariable models were adjusted for the following covariates: age, sex, race, body surface area, hemoglobin level, left main coronary artery disease, prior myocardial infarction, dialysis, endocarditis, prior transient ischemic attack or stroke, carotid stenosis, prior peripheral artery disease, smoking, diabetes, New York Heart Association (NYHA) class IV, atrial fibrillation or atrial flutter, conduction defect, chronic lung disease, use of home oxygen, hostile chest (prior chest radiotherapy or presence of adhesions from a prior cardiothoracic procedure precludes an open chest procedure), porcelain aorta, use of a pacemaker, previous implantable cardioverter defibrillators, prior percutaneous coronary intervention, prior coronary artery bypass graft, prior cardiac operations, prior aortic procedure, prior nonaortic procedure, moderate or severe aortic insufficiency, moderate or severe mitral regurgitation, moderate or severe tricuspid regurgitation, glomerular filtration rate, left ventricular ejection fraction, mitral regurgitation etiology (primary vs secondary), and procedure date. Within sensitivity analyses, multiplicative interaction terms were created to assess whether mitral regurgitation etiology affected the association of pHTN and clinical outcomes, and subgroup analyses were conducted separately to investigate the association of pHTN with clinical outcomes after TMVr in patients with primary and secondary mitral regurgitation. Mixed mitral regurgitation was included within the primary mitral regurgitation subgroup.
Data were analyzed from November 4, 2013, to March 31, 2017. A 2-sided P < .05 signified statistical significance. All analyses were performed using SAS statistical software, version 9.4 (SAS Institute Inc) at the TVT Registry Analysis Center at the Duke Clinical Research Institute.
In our study cohort of 4071 patients, the median age was 81 (IQR, 73-86) years; 1885 (46.3%) were women and 2186 (53.7%) were men; and 3443 (84.6%) were white. The median mPAP was 31 (IQR, 24-39) mm Hg, and systolic PAP was 48 (IQR, 37-60) mm Hg. Table 1 summarizes baseline characteristics among the different pHTN groups. Patients with more severe pHTN were more likely to be younger and have diabetes, hypertension, lung disease, oxygen dependency, and higher body mass index. When compared with group 1 (patients without pHTN), patients with mild, moderate, and severe pHTN had a higher STS predicted risk of mortality score for mitral valve repair (group 1, 5.1% [IQR, 3.1%-7.7%]; group 2, 5.9% [IQR, 3.9%-8.5%]; group 3, 6.2% [IQR, 4.0%-9.4%]; and group 4, 5.4% [IQR, 3.1%-8.5%]; P < .001) and mitral valve replacement (group 1, 7.5% [IQR, 4.8%-11.0%]; group 2, 8.2% [IQR, 5.7%-11.5%]; group 3, 8.6% [IQR, 5.6%-12.4%]; and group 4, 8.1% [IQR, 5.0%-11.6%]; P < .001). More severe pHTN was associated with lower median Kansas City Cardiomyopathy Questionnaire scores (group 1, 43.8 [IQR, 26.0-63.0]; group 4, 34.4 [IQR, 17.7-53.1]; P < .001) and shorter 6-minute walk test distances before TMVr (group 1, 237 m [IQR, 126-322 m]; group 4, 183 m [IQR, 101-274 m]; P < .001).
Comparisons of baseline patient characteristics in the CMS-linked vs non–CMS-linked cohorts (eTable 1 in the Supplement) demonstrated that the CMS-linked cohort was older (median age, 82 [IQR, 76-86] vs 78 [IQR, 67-84] years) and had higher rates of prior permanent pacemaker (486 [20.4%] vs 296 [17.5%]), coronary artery bypass graft (702 [29.5%] vs 412 [24.4%]), and atrial fibrillation (1533 [64.4%] vs 990 [58.6%]). However, rates of prior myocardial infarction (555 [23.3%] vs 450 [26.6%]), dialysis dependency (298 [12.5%] vs 100 [5.9%]), diabetes (593 [24.9%] vs 510 [30.2%]), NYHA classes III (1552 [65.2%] vs 1136 [67.2%]) and IV (407 [17.1%] vs 318 [18.8%]) at presentation, and cardiomyopathy (29 [1.2%] vs 35 [2.1%]) were less frequent within the CMS-linked cohort. Because 1690 (41.5%) of the STS/ACC TVT registry patients were excluded from this analysis owing to missing preprocedure PAP, we compared baseline patient characteristics of our study cohort with those of patients with missing preprocedure PAP (eTable 2 in the Supplement) and found the cohorts to be comparable.
Pulmonary hypertension was associated with indices of heart failure, including higher prevalence of NYHA functional class IV (147 [13.3%] in group 1 vs 118 [21.1%] in group 4; P < .001), greater use of implantable cardioverter defibrillators (119 [10.8%] in group 1 vs 111 [19.9%] in group 2; P < .001), greater use of loop diuretics (692 [62.7%] in group 1 vs 460 [82.4%] in group 4; P < .001), and greater levels of brain-type natriuretic peptide (297 [IQR, 159-579] pg/mL in group 1 vs 618 [IQR, 331-1227] pg/mL in group 4; P < .001) or N-terminal pro–brain-type natriuretic peptide (1805 [IQR, 748-3354] pg/mL in group 1 vs 3111 [IQR, 1396-6266] pg/mL in group 4; P < .001) levels. Although median left ventricular ejection fraction was similar in all 4 groups, patients with moderate and severe pHTN (131 [13.2%] and 97 [17.5%], respectively) were more likely to have left ventricular ejection fraction of less than 30% compared with patients with mild or no pHTN (146 [10.6%] and 73 [6.7%], respectively) (P < .001). Similarly, patients with worse pHTN had lower median cardiac output (group 1, 4.2 [IQR, 3.4-5.1] L/min; group 4, 3.8 [3.0-4.8] L/min), higher median pulmonary capillary wedge pressure (group 1, 12.0 [IQR, 9.0-16.0] mm Hg; group 4, 30.0 [IQR, 25.0-36.0] mm Hg), and higher median right atrial pressures (group 1, 6 [IQR, 4-8] mm Hg; group 4, 16 [IQR, 12-21] mm Hg) (Table 2). After accounting for 133 patients with missing data, 2989 (75.9%) had primary mitral regurgitation.
Most patients received a single clip (2149 [52.8%]), and 126 (3.1%) had an unsuccessful TMVr procedure (eTable 3 in the Supplement). The number of clips was consistent across pHTN groups. Patients with worse pHTN had higher rates of severe mitral regurgitation after TMVr (group 1, 28 [2.5%]; group 2, 39 [2.8%]; group 3, 42 [4.2%]; and group 4, 29 [5.2%]; P < .001) and lower rates of mild mitral regurgitation (group 1, 608 [55.1%]; group 2, 711 [50.8%]; group 3, 492 [48.7%]; and group 4, 269 [48.2%]; P < .001). Conversion to open heart surgery and use of mechanical support or cardiopulmonary bypass were similar regardless of pHTN severity. Length of stay was also similar across groups, with a median length of 2.0 (IQR, 1.0-4.0) days. In-hospital mortality was 1.6% (n = 65) in the entire cohort, with higher mortality noted in patients with pHTN (group 1, 4 [0.4%]; group 2, 29 [2.1%]; group 3, 17 [1.7%]; and group 4, 15 [2.7%]; P < .001).
The 30-day outcomes in the STS/ACC TVT registry are summarized in Table 3. The composite of mortality and heart failure readmission at 30 days was 5.7% (n = 232) within the entire cohort of 4071 patients, with progressively higher rates observed with worse pHTN (group 1, 37 [3.4%]; group 2, 80 [5.7%]; group 3, 65 [6.4%]; group 4, 50 [9.0%]; P < .001). All-cause 30-day mortality was 2.9% (n = 118) within the entire TVT cohort, with progressively higher mortality rates seen across pHTN groups (16 [1.5%] in group 1 to 24 [4.3%] in group 4; P = .004) (Table 3). More severe pHTN was not associated with increased rates of hospitalization for heart failure (99 [2.8%] in group 1 vs 20 [4.2%] in group 4; P = .08) but was associated with persistent NYHA functional class III (group 1, 91 [9.5%]; group 4, 63 [13.3%]) to class IV (group 1, 11 [1.2%]; group 4, 20 [4.2%]; P < .001 for both comparisons) after TMVr. However, robust improvements in NYHA functional class were observed across levels of pHTN. Device-related complications were rare and not statistically different between groups. The 30-day outcomes were comparable in the CMS-linked vs non–CMS-linked cohorts (eTable 1 in the Supplement) and in the study cohort vs the missing preprocedure PAP cohort (eTable 2 in the Supplement).
After multivariable adjustment, pHTN remained independently associated with the composite of 30-day all-cause mortality and hospitalization for heart failure (hazard ratio [HR] for group 4 vs 1, 1.56; 95% CI, 0.99-2.42; P = .05) (eTable 5 in the Supplement). Although 30-day mortality was similarly associated with more severe pHTN (HR for group 4 vs 1, 2.81; 95% CI, 1.28-6.16; P = .01), adjusted rates of 30-day hospitalization for heart failure did not differ between groups (HR for group 4 vs 1, 1.23; 95% CI, 0.71-2.14; P = .46) (eTable 5 in the Supplement).
Among patients with linked CMS claims data, the composite end point of 1-year all-cause mortality and hospitalization for heart failure occurred in 33.6% (95% CI, 31.6%-35.7%), with higher rates noted with worse pHTN (group 1, 27.8% [95% CI, 24.2%-31.5%]; group 2, 32.4% [95% CI, 29.0%-35.8%]; group 3, 36.0% [95% CI, 31.8%-40.2%]; and group 4, 45.2% [95% CI, 39.1%-51.0%]; P < .001) (eTable 4 in the Supplement and Figure 2A). One-year mortality demonstrated a similar pattern and was 20.5% (95% CI, 18.8%-22.3%) in the entire cohort, with higher rates noted with worse pHTN (group 1, 16.3% [95% CI, 13.4%-19.5%]; group 2, 19.8% [95% CI, 17.0%-22.8%]; group 3, 22.4% [95% CI, 18.8%-26.1%]; and group 4, 27.8% [95% CI, 22.6%-33.3%]; P < .001) (eTable 4 in the Supplement and Figure 2B). The rate of readmission for heart failure at 1 year was 21.4% (95% CI, 19.6%-23.1%) in the whole cohort. Worse pHTN was associated with higher rates of readmission for heart failure at 1 year (group 1, 17.7% [95% CI, 14.8%-20.9%]; group 2, 20.2% [95% CI, 17.4%-23.1%]; group 3, 23.8% [95% CI, 20.1%-27.5%]; and group 4, 27.9% [95% CI, 22.7%-33.3%]; P < .001) (eTable 4 in the Supplement and Figure 2C).
The association of severe pHTN with increased rates of the composite of 1-year all-cause mortality and hospitalization for heart failure persisted despite multivariable adjustment for potential confounders (HR for group 4 vs 1, 1.44; 95% CI, 1.16-1.79; P < .001) (eTable 5 in the Supplement) and when modeled in a continuous manner (HR per 5-mm Hg increase in mPAP, 1.04; 95% CI, 1.01-1.07; P = .01) (eTable 5 in the Supplement). The association of severe pHTN with increased rates of 1-year all-cause mortality persisted after multivariable adjustment as well (HR for group 4 vs 1, 1.45; 95% CI, 1.10-1.93; P = .01) and in a continuous manner (HR per 5-mm Hg increase in mPAP, 1.05; 95% CI, 1.01-1.09, P = .02) (eTable 5 in the Supplement). The association of pHTN on the composite of 1-year all-cause mortality and hospitalization for heart failure (Wald χ2, 1.47; P = .51 for interaction) and also on the individual end points of 1-year all-cause mortality (Wald χ2, 3.77; P = .15 for interaction) and 1-year hospitalization for heart failure (Wald χ2, 2.55; P = .28 for interaction) after TMVr was consistent between primary and secondary mitral regurgitation (eFigure 1 and eFigure 2 in the Supplement).
In this study using the nationwide STS/ACC TVT registry of patients undergoing commercial TMVr with the MitraClip system for severe mitral regurgitation, we found that (1) pHTN was present in most patients with severe mitral regurgitation who underwent TMVr, with severe pHTN noted in 13.7% of patients; (2) more severe pHTN is associated with increased risk of all-cause mortality, hospitalization for heart failure, the composite of mortality and hospitalization in 1 year, and mortality in 30 days; (3) even mild elevation of PAP is associated with adverse clinical outcomes; (4) the risk associated with pHTN is consistent in patients undergoing TMVr for primary and secondary mitral regurgitation; (5) robust improvements in functional capacity occur after TMVr, even among patients with severe pHTN; and (6) TMVr in patients with pHTN is safe and effective. These observations highlight the importance of pHTN in patients with mitral regurgitation undergoing TMVr and, we believe, have clear implications for patient management.
Several smaller studies have investigated the clinical association of pHTN with outcomes after TMVr.14,15 Tigges and colleagues15 described 643 patients within the German Transcatheter Mitral Valve Interventions registry who underwent repair with the MitraClip system. The authors found that pHTN was associated with increased risk of major adverse cardiac and cerebrovascular events (composite of death, myocardial infarction, and stroke) at 1 year but not at the 30-day follow-up.15 In contrast to our findings, that study found no difference in rehospitalization rates after repair with MitraClip with more severe pHTN, although more than half of their patients had secondary mitral regurgitation.15 Matsumoto et al14 found that long-term mortality is increased in patients with pHTN despite improvements in mitral regurgitation grade and NYHA functional class after repair with the MitraClip system. Herein, within our large, nationally representative cohort, we found that pHTN is associated with not only increased post-TMVr mortality rates but also rates of rehospitalization for heart failure. Moreover, the association of pHTN is not reserved to those patients with severely elevated mPAP. Instead, our study establishes a graded association between mPAP and the risk of adverse clinical outcomes after TMVr.
Current valvular management guidelines from the American Heart Association and the ACC recommend mitral valve surgery for patients with severe mitral regurgitation if they are symptomatic or in the absence of symptoms if evidence of left ventricular dysfunction or dilatation is found. Surgical mitral valve repair is also reasonable in asymptomatic patients with severe primary mitral regurgitation complicated by new-onset atrial fibrillation or resting systolic PAP of greater than 50 mm Hg (class IIa; level of evidence, B).16 Herein, we demonstrate that even mildly elevated PAPs are associated with adverse clinical outcomes after TMVr, raising concerns that current PAP thresholds within the clinical guidelines may suggest mitral valve intervention too late in the disease course. Similar observations have been made in primary mitral regurgitation. Mentias and colleagues4 found that surgical mitral valve repair in patients with right ventricular systolic pressure of at least 35 mm Hg was associated with adverse outcomes. Matsumoto et al14 have also found that severity of mitral regurgitation is greater in patients with more severe pHTN, and pHTN, in turn, is associated with a greater burden of heart failure. Together, these observations within this large, nationwide sample lends further evidence suggesting that interventions for mitral regurgitation may be occurring too late in the disease course.17 Although secondary mitral regurgitation may be a consequence of adverse remodeling associated with cardiomyopathy, the preponderance of primary mitral regurgitation within this cohort and the observed higher pulmonary capillary wedge pressure within groups with more severe pHTN implicate mitral regurgitation as the causative factor for pHTN, although additional efforts are needed to determine whether earlier intervention for mitral regurgitation may disrupt and improve associated pHTN.
Transcatheter mitral valve repair significantly improves functional capacity, even in patients with severe pHTN at baseline, suggesting that TMVr should not be denied because of severe pHTN. We observed a decline in NYHA classes III to IV from 86.7% to 17.5% in those patients with severe pHTN who survived 30 days after TMVr, although approximately 4% of such patients died within that period. Furthermore, despite adverse outcomes after TMVr in patients with pHTN, these data appear to endorse the safety and effectiveness of TMVr for severe mitral regurgitation in the presence of moderate to severe pHTN. Even in patients with severe pHTN, 30-day mortality was lower than estimated using the STS predicted risk of mortality score. Further studies are needed to identify clinical or echocardiographic parameters associated with clinical response to TMVr in those patients with severe pHTN.
Data are limited regarding the efficacy of medical therapies to reduce pulmonary pressures in patients undergoing mitral valve intervention with normal left ventricular ejection fraction. Within a randomized trial, Bermejo et al18 assessed the role of sildenafil citrate in 200 patients who underwent surgical repair or replacement of valvular heart disease 1 year earlier and had persistent pHTN. Sildenafil given for 6 months was associated with harm, with a lower rate of survival free of hospitalization for heart failure noted within the sildenafil group compared with the placebo group.18 The use of silendafil in patients with corrected left heart valve disease is therefore discouraged. Although the use of diuretics may improve symptoms associated with pHTN, no pharmacological therapies have been proven to improve survival in patients with severe mitral regurgitation and pHTN.
This study has some limitations. Although the STS/ACC TVT registry 30-day data provide information on the cause of death and the clinical justification for hospital readmissions, 1-year CMS administrative claims data do not. Furthermore, the present study reports on a single-arm registry that does not allow for inferences regarding the causal role of pHTN in driving the risk of adverse clinical outcomes. Although extensive statistical adjustment was made to mitigate the effects of potential confounders, pHTN may remain a marker of another causal pathology. Data on PAP after TMVr are not available within the STS/ACC TVT registry, so we are not able to investigate the efficacy of TMVr in reversing pHTN. Similarly, the STS/ACC TVT registry does not possess data on right ventricular function, which may modify and confound the association of pHTN with clinical outcomes after TMVr. A large fraction of patients was excluded from the present analysis because invasive PAP measurements were missing, raising the possibility of selection bias affecting the reported observations. However, baseline characteristics and 30-day TVT outcomes were comparable in the study cohort and the cohort with missing preprocedure PAP. Finally, dependence on CMS claims data may limit the generalizability of our findings. Nevertheless, we did not find statistical differences in baseline and clinical 30-day outcomes between CMS-linked and non–CMS-linked cohorts. Another limitation is the lack of data on right ventricular sizes and functions that could be a marker of advanced pHTN along with obstructive sleep apnea.
Within a large national registry of patients undergoing TMVr for severe mitral regurgitation using the MitraClip system, we found that pHTN is highly prevalent and associated with adverse clinical outcomes, including an increased risk of mortality, hospital readmission, and their composite. Associations of pHTN with metrics of advanced heart failure in this cohort with largely primary mitral regurgitation suggest that treatment of mitral regurgitation occurs late in the disease course. Further efforts are needed to determine whether earlier intervention for mitral regurgitation may result in improved clinical outcomes.
Accepted for Publication: September 5, 2019.
Corresponding Author: Sammy Elmariah, MD, MPH, Cardiology Division, Department of Medicine, Massachusetts General Hospital, 55 Fruit St, Gray Building, Ste 800, Boston, MA 02114 (email@example.com).
Published Online: November 20, 2019. doi:10.1001/jamacardio.2019.4428
Author Contributions: Drs Al-Bawardy and Elmariah had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis.
Concept and design: Al-Bawardy, Vemulapalli, Thourani, Mack, Palacios, Inglessis, Sakhuja, Passeri, Yucel, Melnitchouk, Elmariah.
Acquisition, analysis, or interpretation of data: Al-Bawardy, Vemulapalli, Mack, Dai, Stebbins, Palacios, Sakhuja, Ben-Assa, Dal-Bianco, Yucel, Vlahakes, Jassar, Elmariah.
Drafting of the manuscript: Al-Bawardy, Dai, Elmariah.
Critical revision of the manuscript for important intellectual content: Al-Bawardy, Vemulapalli, Thourani, Mack, Stebbins, Palacios, Inglessis, Sakhuja, Ben-Assa, Passeri, Dal-Bianco, Yucel, Melnitchouk, Vlahakes, Jassar, Elmariah.
Statistical analysis: Dai, Stebbins, Palacios, Elmariah.
Administrative, technical, or material support: Al-Bawardy, Sakhuja, Ben-Assa, Vlahakes.
Supervision: Vemulapalli, Thourani, Palacios, Inglessis, Sakhuja, Dal-Bianco, Yucel, Vlahakes, Jassar, Elmariah.
Conflict of Interest Disclosures: Dr Vemulapalli reported receiving grants from Abbott Vascular during the conduct of the study; grants from Boston Scientific, the National Institutes of Health, Patient-Centered Outcomes Research Institute, and the Society of Thoracic Surgeons outside the submitted work; and personal fees from Boston Scientific, Janssen Pharmaceutica, Zafgen, Inc, and Premier Pharmacy Labs outside the submitted work. Dr Mack reported receiving nonfinancial support from Edwards Lifesciences, Abbott Laboratories, and Medtronic outside the submitted work. Dr Palacios reported receiving grants from Abiomed outside the submitted work. Dr Passeri reported receiving personal fees from Abbott Vascular outside the submitted work. Dr Elmariah reported receiving grants from Edwards Lifesciences and personal fees from Medtronic and AstraZeneca outside the submitted work. No other disclosures were reported.
Funding/Support: This study was supported by the American College of Cardiology’s National Cardiovascular Data Registry (NCDR).
Role of the Funder/Sponsor: The sponsor 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.
Disclaimer: The views expressed in this article represent those of the authors and do not necessarily represent the official views of the NCDR or its associated professional societies identified at https://cvquality.acc.org/NCDR-Home.
Create a personal account or sign in to: