Prospective evaluation of elastic restraint to lessen the effects of heart failure (PEERLESS-HF) trial.

BACKGROUND: Left ventricular (LV) remodeling predicts poor outcomes in heart failure (HF) patients. The HeartNet(®) cardiac restraint device (Paracor Medical Inc., Sunnyvale, CA) may reduce LV remodeling and improve functional capacity, quality of life, and outcomes in HF patients. To evaluate the safety and efficacy of the HeartNet Ventricular Support System in HF patients receiving optimal medical therapy. METHODS AND RESULTS: Prospective, randomized, controlled, multicenter trial in patients with symptomatic HF and LV ejection fraction ≤35% on optimal medical and device therapy. The primary efficacy end points were changes in peak VO(2), 6-minute walk (6MW) distance, and Minnesota Living with Heart Failure (MLWHF) quality of life score at 6 months. The primary safety end point was all-cause mortality at 12 months. Because the planned adaptive interim analysis of the first 122 subjects with a completed 6-month follow-up indicated futility to reach the peak VO(2) end point, trial enrollment was suspended. Hence, the results on the 96 treatment and 114 control subjects are reported. Groups were similar at baseline. At 6 months, responder frequency for a prespecified improvement was similar between groups for peak VO(2) (P = .502) and MLWHF score (P = .184) but borderline higher for improvement in 6MW distance in the treatment compared with the control group (33 [38%] vs. 25 [25%]; P = .044). At 6 months, the treatment group had a significantly greater improvement in Kansas City Cardiomyopathy Questionnaire (KCCQ) (P < .001) and decrease in LV mass (P = .032), LV end-diastolic diameter (P = .015), LV end-systolic diameter (P = .032), and LV end-diastolic volume (P = .031) as compared with controls. At 12 months, all-cause mortality and responder rates were similar in the 2 groups. Success rate for the HeartNet implantation was 99%. CONCLUSION: Enrollment in the trial was stopped because an interim analysis showed futility of reaching the peak VO(2) end point. However, because of the device safety and favorable signals for LV remodeling and quality of life, further investigation of this device is warranted.