MADIT-CRT Analysis Ties Echo Changes to Clinical Benefit in CRT for Mild HF

June 02, 2010

June 2, 2010 (Berlin, Germany) — It didn't come as a shocker, but it's nice to be a little surer of it: cardiac resynchronization therapy (CRT) improves clinical outcomes in appropriately selected patients with heart failure at least partly because it can shrink enlarged hearts and boost ventricular function, the echocardiographic substudy of a major randomized trial suggests.

One-year declines in LV systolic and diastolic volumes and gains in LVEF were at least three times as great in patients with CRT-defibrillator (CRT-D) as in those with defibrillator-only (ICD) devices in an analysis from the Multicenter Automatic Defibrillator Implantation Trial with Cardiac Resynchronization Therapy (MADIT-CRT).

"These findings were concordant with and predictive of the primary outcome of death or heart-failure events and suggest a compelling cardiac structural and functional mechanism by which CRT improves outcomes," Dr Scott D Solomon (Brigham and Women's Hospital, Boston, MA) said when presenting the analysis here at the Heart Failure Congress 2010 sessions of the Heart Failure Association of the European Society of Cardiology.

I think this is powerful stuff and that this will have an impact on practice. It will confirm, at least for me, assumptions that we've had, that resynchronization leads to reverse remodeling.

The overall trial's primary result, published last year [1] and reported at the time by heartwire , was a 34% drop in mortality/HF-event risk over a mean 2.5 years. The benefit was driven primarily by a steep reduction in HF events and all the more striking because MADIT-CRT entered only patients with NYHA class 1-2 heart failure; they also had pronounced ventricular dyssynchrony as defined by a QRS duration of at least 130 ms. The 1820-patient trial was conducted at 110 centers in Europe and the US.

Its findings are consistent with and expand on earlier trials, such as CARE-HF and REVERSE, that showed similar effects of CRT on cardiac structure and helped make it a mainstream therapy for patients with more severe heart failure.

"My task is easy," said Dr Kenneth Dickstein (University of Bergen, Stavanger Norway), who had been invited to formally discuss Solomon's presentation. "Their data speak for themselves. This is a large sample size, . . . there was a very rigorous data collection and experienced core lab, and the process was performed well. The results are very convincing," he said, "both for the echo changes and their concordance with the clinical outcomes." He pointed out that all patients in the study had devices with defibrillators, so the benefits seen from CRT in the trial were on top of not only optimal medications but also ICD therapy.

Solomon's conclusion that the observed relationships between echocardiographic changes and clinical outcomes provide insights about how CRT works, Dickstein said, was "appropriately modest. I can be slightly more enthusiastic. I think this is powerful stuff and that this will have an impact on practice. It will confirm, at least for me, assumptions that we've had, that resynchronization leads to reverse remodeling."

The trial's echocardiographic substudy included patients who had echocardiography performed at both baseline and one year, the latter instance with the CRT device activated. About 85% of the substudy's patients were in NYHA class 2; the remainder was in class 1--that is, asymptomatic on their existing therapy.

Changes From Baseline to One Year by Treatment Group in The MADIT-CRT Echocardiographic Substudy

Echocardiographic measure CRT-D, n=749 ICD-only, n=623
LVEF (% points) +11 +3
LVEDVI (%) -21.0 -5.9
LVESVI (%) -32.5 -10.2
LA volume (%) -26.2 -10.0

All differences p=0.0001

LVEF=left ventricular ejection fraction; LVEDVI=left ventricular end-diastolic volume index; LVESVI=left ventricular end-systolic volume index; LA=left atrial

The significant CRT effects on LV end-systolic and end-diastolic volumes were observed in all of the study's prospectively defined subgroups, which separated the patients by age, sex, NYHA class, HF etiology, QRS duration, and whether they had diabetes or left bundle-branch block (LBBB). Moreover, the benefits were significantly more pronounced (p<0.001) in women and patients with nonischemic cardiomyopathy, LBBB, or a QRS >150 ms, Solomon reported.

Improvements in ventricular end-systolic and -diastolic volumes were significantly and independently predictive of reduced risk of the primary end point as well as all-cause mortality.

Clinical-Outcomes Hazard Ratios (HR) Associated With Echocardiographic Improvements at One Year in MADIT-CRT*

Parameter HR (95% CI) per 10% LVEDV decrease p HR (95% CI) per 10% LVESV decrease p HR (95% CI) per 5 %-point LVEF increase p
Death or HF event 0.60 (0.51–0.72) <0.001 0.72 (0.65–0.84) <0.001 0.60 (0.50–0.72) <0.001
All-cause mortality 0.79 (0.59–1.06) NS 0.83 (0.68–0.99) 0.047 0.69 (0.51–0.93) 0.014

*Adjusted for age, sex, diabetes, HF etiology; renal function; QRS width; baseline NYHA class; use of ACE inhibitors, angiotensin-receptor blockers, beta blockers, diuretics, and statins; and treatment group

LVEF=left ventricular ejection fraction; LVEDV=left ventricular end-diastolic volume; LVESV=left ventricular end-systolic volume

Dickstein observed that the benefits in MADIT-CRT were to some extent dependent on the population's characteristics. The patients tended to be young (the mean age was about 64 years), in NYHA class 2, and male (only about 25% were women). "They also had very wide QRS complexes [averaging 159 ms], which I think made it more likely that they would demonstrate a positive result."

MADIT-CRT, Dickstein said, "might encourage clinicians to select patients with mild symptoms and wide QRS complexes [for CRT]. But remember, patients were not selected for this trial based on echo criteria. So the degree of extrapolation should be somewhat conservative."

MADIT-CRT was funded by Boston Scientific, for which Solomon has consulted. Dickstein had no disclosures.