EVEREST II: Percutaneous mitral
valve repair compares favorably with open-heart surgery especially for select
patients
The MitraClip® - a tiny device threaded through an
artery to repair leaky heart valves - continues to compare favorably with conventional
open-heart surgery for treatment of select patients with mitral regurgitation,
according to updated research findings from the EVEREST II study presented at
the American College of Cardiology's 60th Annual Scientific Session.
Each year 250,000 people in the United States learn they have Mitral regurgitation
(MR), but only 20 percent eventually undergo standard treatment to repair or replace
the valve, which is open-heart surgery that puts patients on a heart-lung bypass
machine. For many, that procedure poses too great a risk. Instead they rely on
medications to reduce MR symptoms, and limit their activities as physical function
declines. Symptoms of MR can include palpitations, shortness of breath, fatigue,
lightheadedness, cough and swelling in the legs and feet from fluid buildup.
The percutaneous MitraClip® involves a minimally invasive procedure that
may make valve repair feasible for more people with MR. Cardiologists trained
to use this system guide a catheter-mounted device through an incision in the
groin, into the femoral artery and on to the heart. When the device is properly
placed, it clamps the edges of the faulty valve together like a clothespin (sometimes
a second clip is inserted for better control of MR). Standard MR surgery and MitraClip
insertion both take about two hours, but their hospitalization and recovery times
differ greatly.
"After getting a MitraClip®, patients spend one or two nights in the
hospital versus five to seven days after open-heart surgery, and they're back
to full activity immediately," said Ted Feldman, M.D., director of the cardiac
catheterization laboratory at NorthShore University HealthSystem, Evanston, Ill.,
and the study's co-principal investigator. "Traditional open-heart surgery
has a recovery time of one to three months. The contrast is pretty striking."
This Phase II study enrolled 279 patients in 37 North American centers who
met criteria for mitral valve surgery: grade 3+ (moderate to severe) or 4+ (severe)
MR. All patients had valve anatomy suitable for the procedure and were randomly
assigned in a 2-to-1 ratio to the MitraClip® or to standard surgery. Both
groups were well matched for baseline patient characteristics. Year 1 data have
been published. This presentation reports Year 2 data, and patients will be followed
for five years.
The effectiveness of the two treatments was measured by a composite of freedom
from death, no new mitral valve surgery and MR lower than pretreatment minimum
of 3+. In the treatment comparison, 101 patients (62.7 percent) in the percutaneous
group met the composite endpoint vs. 66 (66.3 percent) in the surgery group. At
two years, 78 percent of patients with the MitraClip did not need surgery.
Major adverse events at 30 days were significantly lower in the percutaneous
group (15.0 percent vs. 47.9 percent). Blood transfusions of two units or more
account for most of this difference: 13.3 percent in the percutaneous group vs.
44.7 percent for surgery patients.
Durability and anti-clotting drugs are other issues considered in this research.
A mechanical heart valve lasts 35 years and requires the patient to take warfarin
for life. Valve repair with a MitraClip® should last approximately 15 years,
and after implant, patients take clopidogrel for just one month and aspirin for
six months. Although the procedure's surgical version has demonstrated durability
for more than 12 years, long-term outcomes from MitraClip can be defined only
after further study.
"Both procedures reduced mitral regurgitation and produced meaningful
clinical benefits, with the MitraClip® valve repair increasing safety and
surgery decreasing mitral regurgitation more completely," Feldman said. "Our
two-year data indicate that the percutaneous procedure is a therapeutic option
for certain patients with significant mitral regurgitation."
The EVEREST II study (Endovascular Valve Edge-to-Edge REpair STudy) is funded
by Evalve, Inc., which also provides research funding to NorthShore University
HealthSystem. Feldman is a consultant to Abbott, which acquired Evalve in 2009.
These findings were published concurrently in the New England Journal of Medicine.
|