Show Some Restraint: Heartnet Randomized Trial Halted Early, Results Mixed

September 16, 2010

September 16, 2010 (San Diego, California) — A "superelastic" metal mesh that slides around the heart to provide "just the right amount of cardiac support" to rein in remodeling without constricting hemodynamics, according to researchers, may slow or stop ventricular dilatation or even prevent some heart-failure hospitalizations, hints a randomized study that isn't conclusive because enrollment was halted prematurely [1].

Dr William T Abraham

As described by the trial's principal investigator, Dr William T Abraham (Ohio State University, Columbus, OH), here at the Heart Failure Society of America 2010 Scientific Meeting, the Prospective Evaluation of Elastic Restraint to Lessen the Effects of Heart Failure (PEERLESS-HF) trial was stopped after a prespecified interim analysis suggested that six months of treatment with the HeartNet (Paracor Medical) device had no effect on some functional end points, particularly peak VO2.

On the other hand, based on the patients who did enter the study, using the device is safe (it didn't increase 12-month all-cause mortality), it may stop ventricular enlargement, and it may have had a more pronounced benefit in the 50% of patients already treated with cardiac resynchronization therapy (CRT), Abraham said. "These findings, we believe, support undertaking a clinical trial that targets patients with CRT but with an inadequate CRT response."

PEERLESS-HF randomized 217 patients with stage C heart failure of either etiology for at least six months who had been on optimal medications for at least three months to receive the HeartNet (by minithoracotomy) on top of optimal meds or to optimal meds only; inclusion required an LVEF <35%, LV end-diastolic diameter <85 mm, and LV end-diastolic diameter index <40 mm/m2.

There was a really quite impressive reduction in left ventricular mass.

The population had been intentionally enriched with patients with defibrillating or pacing-only CRT devices. "We tried to get CRT nonresponders or inadequate responders based on animal data of synergy between the two [devices] and preliminary feasibility data of a better response in a CRT population," Abraham said.

HeartNet (Source: Paracor Medical)

The trial was stopped after no significant difference was seen in peak VO2 after six months for the first 50% of patients enrolled; at that time, 39% in the treatment group and 41% of controls had improved in peak VO2; 28% and 25%, respectively, had worsened (p=0.7)

Survival at 12 months, the prespecified safety end point, was "equivalent," Abraham said, at 93.4% in the treatment group and 92.4% among controls (p=0.056). And 39% and 26%, respectively, had improved their six-minute-walk distance (p=0.08); 41% and 28%, respectively, had improved in NYHA class (p=0.056); and 63% and 41%, respectively, improved in Kansas City Cardiomyopathy Questionnaire (KCCQ) quality-of-life scores (p=0.002).

The HeartNet group also showed significant improvements in some echocardiographic measures of remodeling, but not all those measured.

PEERLESS-HF: Changes in Echocardiographic Measures of Remodeling at Six Months, Heartnet vs Optimal Medical Therapy

Echocardiographic end point HeartNet, n=103 Control, n=114 p
LVEDD (mm) -2 +1 0.02
LVESD (mm) -1 0 0.04
LVEDV (mL) -3 +10.5 0.06
LVESV (mL) -7 +2.4 0.23
LV mass (g) -20 +8 0.03
LVEF (% points) +1 +1.9 0.10

LVEDD=left ventricular end-diastolic diameter; LVESD=left ventricular end-systolic diameter; LVEDV=left ventricular end-diastolic volume; LVESV=left ventricular end-systolic volume

"There was a really quite impressive reduction in left ventricular mass seen," Abraham said. "So there seems to be something going on here in terms of reverse remodeling, but in some instances the changes are relatively modest."

PEERLESS-HF wasn't powered for clinical outcomes, but reductions in risk of a first heart-failure event (defined as CV death, transplantation, ventricular assist device [VAD] implantation, or HF hospitalization) favored the HeartNet group. The same was seen for "major cardiac procedures," which included transplantation and VAD implantation.

Dr Eric J Velazquez [Source: Duke]

As the featured discussant following Abraham's presentation, Dr Eric J Velazquez (Duke Clinical Research Institute, Durham, NC) said the device was associated with modest improvements with no major adverse safety signal, but it should be considered in context with other heart-failure device therapies.

If the same can be done by exercise, you wonder if this is an important step forward.

"If you look at these results relative to standards of care in NYHA class 2–3 heart failure, I think it is sobering to see the lack of improvements, particularly in remodeling and functional end points," he said. "Using CARE-HF as an example, there were obvious CRT improvements leading to rapid and progressive sustained improvement in LV end-diastolic volume, LV end-systolic volume, and LVEF. In comparison, although it was over a shorter period of time, these improvements were not seen in the PEERLESS-HF study."

Also, he said, "relative to exercise, if you use that as a comparator, the impact of the [HeartNet device] on functional impairment was very modest, and in fact if the same can be done by exercise, you wonder if this is an important step forward."

Concluding, Velazquez said he believes "targeting CRT nonresponders is a very rational approach, but it's important to target those patients after significant time has elapsed since CRT implantation, to allow the most improvement that can be obtained."

The study was supported by Paracor Medical. Abraham reports he is a consultant for Paracor Medical as well as CardioKinetix, CardioMEMS, Medtronic, and St Jude Medical. Velazquez reports he has received research grants from Novartis, Johnson & Johnson, CardioKinetix, and Boehringer Ingelheim and fees for speaking or honoraria from Novartis; he has been a consultant to or on an advisory board for Novartis, Philips, and Boehringer Ingelheim.