REDUCE FMR was the first ever double-blinded, randomized, parallel, sham-controlled RCT of a percutaneous heart valve therapy. The objective was to assess the safety and effectiveness of the Carillon Mitral Contour System for treatment of secondary (functional) mitral regurgitation (FMR) associated with heart failure. Key eligibility criteria were similar to the prior trials of the Carillon system including NYHA class II-IV, MR grade 2+ to 4+, LVEDD >55mm and an ejection fraction <50%, despite the use of stable guideline-directed medical therapy. The primary endpoint was change in regurgitant volume (RV) at 1 year by intention-to-treat analysis. Being double-blind, the study was of the highest scientific rigor – patients and assessors were blinded throughout follow-up and the independent echo core lab not only was blinded to treatment and the timing of the images but also read the images in random order.

In total, 120 patients were randomly allocated to treatment (87) or the sham-control (33) arm at 31 sites in Europe, Australia and New Zealand.

Echocardiographic evaluation was performed at the site prior to enrollment, and 35 patients (29.2%) were later judged by the echo core laboratory to have had MR 1+ at baseline. In 14 (16%) of those allocated to treatment, no device was implanted, due most commonly to coronary impingement (n=8) which resolved following device removal in all cases. The average device implant time was 64.2±34.8 minutes and total procedure time averaged 102.7±54.9 minutes. Freedom from device-related major adverse events (MAE) was 98.9%, with no device embolizations, device fractures, cardiac perforations or intra-procedural ischemic events. There were two deaths within 30 days, both in implanted patients attributed to progressive cardio-renal deterioration. The incidence of MAE at 1 year was not statistically different between the treatment and control groups (16.1% in treatment group vs. 18.2% in control group).

The primary endpoint was met with a statistically significant reduction in mitral RV in the treatment group as compared to the control group (decrease of 7.1 mL/beat vs. an increase of 3.3 mL/beat, p=0.049). A secondary analysis of patients with more severe FMR (MR grade ≥3+) demonstrated a similar pattern of changes. Patients in the treatment group (n=15) experienced a reduction in mitral RV of 12.8 mL/beat whereas patients in the control group (n=8) experienced an increase of 0.6 mL/beat (p=0.14). In addition to improvements in RV, favorable reverse LV remodeling was demonstrated, with a significant reduction in LV volumes in patients receiving the device vs. the control group.

Patients in the treatment group also experienced a significant improvement in 6-minute walk distance at 12 months compared with their baseline (p=0.002), whereas patients allocated to the control group did not (p=0.29). Likewise, the treatment group had a significant improvement in NYHA class at 12 months compared with baseline (p=0.002) whereas patients allocated to the control group did not (p=0.75).

With REDUCE FMR, the Carillon system became the first TMVr therapy to demonstrate significant favorable reverse LV remodeling at 1 year and significant reduction in RV in a blinded, randomized, sham-controlled trial.