Annals of Internal Medicine | 2019
In HF with secondary mitral regurgitation, transcatheter mitral valve repair reduced HF hospitalizations at 2 years
Abstract
Question In patients with heart failure (HF) and symptomatic, secondary mitral regurgitation (MR), does adding transcatheter mitral valve repair to medical therapy improve clinical outcomes? Methods Design Randomized controlled trial (Cardiovascular Outcomes Assessment of the MitraClip Percutaneous Therapy for Heart Failure Patients with Functional Mitral Regurgitation [COAPT] trial). ClinicalTrials.gov NCT01626079. Allocation Concealed.* Blinding Blinded* (clinical events committee). Follow-up period Minimum 1 year; median 17 months (medical therapy alone group) to 23 months (mitral valve repair group). Setting 78 centers in the USA and Canada. Patients 614 patients 18 years of age (mean age 72 y, 64% men, mean left ventricular ejection fraction [LVEF] 31%) who had ischemic or nonischemic cardiomyopathy, LVEF 20% to 50%, echocardiography-confirmed moderate-to-severe (grade 3+) or severe (grade 4+) secondary MR, and ongoing symptoms (New York Heart Association functional class II to IVa) despite guideline-directed medical therapy at maximal doses. Patients who were eligible for mitral valve surgery were excluded. Intervention Transcatheter mitral valve repair, using the MitraClip device, plus guideline-directed medical therapy (n =302), or guideline-directed medical therapy alone (n =312). Outcomes Primary outcomes were HF hospitalizations at 24 months (efficacy) and freedom from device-related complications (single-leaflet device attachment, device embolization, endocarditis or mitral valve stenosis requiring surgery, left ventricular assist device implantation, heart transplantation, or other device-related event requiring nonelective cardiovascular surgery) at 12 months (safety). Other outcomes included all-cause mortality and MR grade 2+. Patient follow-up 96% at 1 year and 75% at 2 years (intention-to-treat analysis). Main results In the mitral valve repair group, 95% of patients had 1 clip implanted, and 97% (lower 95% CI = 95%) had no device-related complications at 12 months (P <0.001 vs prespecified performance threshold of 88%). Results for comparative outcomes are in the Table. Conclusion In heart failure with symptomatic secondary mitral regurgitation, adding transcatheter mitral valve repair to medical therapy reduced heart failure hospitalizations at 2 years. Transcatheter mitral valve repair + medical therapy (MT) vs MT alone in HF with moderate-to-severe or severe secondary mitral regurgitation (MR) Outcomes Transcatheter mitral valve repair + MT MT alone At 24 mo HR (95% CI) NNT (CI) HF hospitalizations (annualized rate) 36%/person-y 68%/person-y 0.53 (0.40 to 0.70) 4 person-y (2 to 8) All-cause mortality (% patients) 29% 46% 0.62 (0.46 to 0.82) 6 (4 to 12) Cardiovascular mortality (% patients) 24% 38% 0.59 (0.43 to 0.81) 8 (6 to 17) RBI at 12 mo (CI) NNT (CI) MR grade 2+ (% patients) 95% 47% 102% (74 to 140) 3 (2 to 3) HF = heart failure; HR = hazard ratio; other abbreviations defined in Glossary. RBI, NNT, and CI for MR grade 2+ calculated from event rates in article; NNT and CI for cardiovascular mortality calculated from hazard ratio and MT alone event rate in article; NNT and CI for HF hospitalizations and all-cause mortality reported in article. 4 person-years of follow-up would be needed to prevent 1 additional HF hospitalization. Based on Kaplan-Meier estimates. Based on 385 patients. Commentary Functional or secondary MR occurs when coaptation of the leaflets is poor due to changes in the left ventricle. Potential causes include displacement of the papillary muscles, mitral annular dilation, and regional wall motion abnormalities that result in restriction of the mitral valve leaflets. Severe secondary MR occurs in about 25% of patients with HF and reduced ejection fraction (HFrEF) and is independently associated with a high mortality rate (40% to 50% at 3 y) (1). Whether fixing MR with a percutaneous transcatheter approach would affect symptoms, rehospitalization rates, or mortality was not previously known. 2 new trials now address this question and have different conclusions. Both trials enrolled patients who had symptomatic, moderate-to-severe or severe MR with reduced ejection fraction and were not considered surgical candidates. Mean ejection fraction, intensity of medical management, use of resynchronization therapy, and rates of coronary artery disease and atrial fibrillation at baseline were similar in both trials, as were the rates of death in control groups at 1 year (22% in MITRA-FR and 23% in COAPT). However, the mean effective regurgitant orifice area at baseline was greater in COAPT than in MITRA-FR (mean 41 mm2 vs 31 mm2) (2). In addition, left ventricular end-diastolic volumes were smaller in COAPT than in MITRA-FR (101 mL/m2 vs 135 mL/m2) (2). This suggests that patients enrolled in COAPT had more severe MR but less compromised ventricles. In both trials, patient screening before randomization included core echocardiography laboratory assessment. COAPT included the likelihood of a successful procedure in the assessment. This contributed to a 58% exclusion rate in COAPT vs 32% in MITRA-FR. Both trials enrolled patients who remained symptomatic despite optimal guideline-directed medical therapy. In COAPT, the MitraClip group used more reninangiotensin system blockers than the control group at baseline (P =0.02), and the difference was consistent throughout the trial. Both trials used the same MitraClip device, which joins the anterior and posterior leaflet at the point of regurgitation, effectively creating a double orifice and reducing regurgitation. However, there were differences in technical success. In MITRA-FR, 9% of patients in the MitraClip group could not receive device implantation vs 5% in COAPT. Although COAPT followed patients for about twice as long (median 17 to 23 mo vs 1 y), differences in rehospitalization and all-cause mortality in COAPT were apparent early and persisted throughout the trial. In COAPT, we see the effect of MitraClip-mediated reduction in MR in carefully selected patients who were still symptomatic despite optimal use of devices and guideline-directed medical therapy. They also had worse MR (measured by severity in effective regurgitant orifice) but ventricles that were less dilated (in other words, decompensation was probably less advanced) than patients in MITRA-FR. Given how much better patients in the COAPT trial fared with MitraClip vs optimal medical management and how few treatment options remain (left ventricular assist device or transplantation) for patients with HFrEF and severe MR, we now believe that patients similar to those in COAPT should be referred to a heart valve team (including HF specialists) for consideration of this procedure. Results of the RESHAPE-HF2 trial (ClinicalTrials.gov NCT02444338) and the 2-year results of MITRA-FR should be available soon, as well as results from trials of new transcatheter devices (Carillon, Cardioband, Mitralign) and other mitral valve prostheses. Trials of longer duration are also needed to test the durability of these devices. Meanwhile, European guidelines (3) give devices for secondary MR in HFrEF a IIb rating (level of evidence C), and US guidelines do not recommend them yet, although, given these results, this will probably change for a select group of MR patients.