Principal Findings:

Overall, 279 patients were enrolled. In the percutaneous mitral repair group, the mean age was 67 years, 62% were men, 34% had atrial fibrillation, the mean ejection fraction was 60%, 8% had diabetes, 73% had a degenerative mitral valve, and 27% had functional mitral regurgitation. The only baseline characteristic that was different between the groups was congestive heart failure, which was present in 91% of the percutaneous mitral repair group and 78% of the surgical repair/replacement group (p = 0.005).

The primary effectiveness outcome at 1 year (intention-to-treat), defined as freedom from death, mitral valve surgery (in the percutaneous group) or reoperation (in the surgical group), or 3 to 4+ mitral regurgitation occurred in 55% of the percutaneous group versus 73% of the surgical group (p = 0.007).

Rates of death were 6% versus 6%, surgery for mitral valve dysfunction was 20% versus 2.2%, and 3+ to 4+ mitral regurgitation was 21% versus 20%, respectively, for percutaneous versus surgical groups. In a per-protocol analysis, freedom from the primary outcome at 1 year occurred in 72% versus 88% (p = 0.02), respectively. The primary effectiveness outcome at 2 years (intention-to-treat) occurred in 52% versus 66% (p = 0.04), respectively. The primary effectiveness outcome at 5 years occurred in 44% versus 64% (p = 0.003), respectively.

The primary safety outcome at 30 days, defined as any major adverse event, occurred in 15% of the percutaneous mitral repair group versus 48% of the surgical repair/replacement group (p < 0.001). This was due to a lower frequency of blood transfusion in the percutaneous group: 13% versus 45% (p < 0.001). The rate of major adverse events excluding blood transfusion was 5% versus 10% (p = 0.23), respectively. Other individual components of safety were death: 1% versus 2%, urgent/emergent surgery for adverse event: 2% versus 4%, major stroke: 1% versus 2%, ventilation for >48 hours: 0% versus 4%, and new onset of permanent atrial fibrillation: 1% versus 0%, respectively.

Change in left ventricular end-systolic dimension from baseline to 12 months was -5.5 ml in the percutaneous group versus -5.6 ml in the surgical group (p = 0.97). Change in ejection fraction was -2.8% versus -6.8% (p = 0.005), respectively. Physical and mental quality of life improved to a similar degree in both groups.

Among high-risk participants (mean age 76 years), functional mitral regurgitation was present in 70%. One mitral valve device was implanted in 57.3%, while two mitral valve devices were implanted in 38.5%. Mitral regurgitation ≤2+ was present in 86% of participants at discharge and in 84% at 12 months.