CRT Is a Life RAFT for Class II and III HF Patients With Low EF and Wide QRS

Reed Miller

November 14, 2010

November 14, 2010 (Chicago, Illinois) — Results of the first trial to show that cardiac resynchronization therapy (CRT) provides a survival benefit beyond that of implantable cardioverter defibrillator (ICD) therapy alone, in patients with moderate HF, were announced during a late-breaking trial session today here at the American Heart Association (AHA) 2010 Scientific Sessionsand are simultaneously published in the New England Journal of Medicine [1].

Results of the Resynchronization/Defibrillation for Ambulatory Heart Failure Trial (RAFT) trial show that adding CRT to ICD therapy improves survival in patients with moderate heart-failure symptoms, a wide QRS complex, and left ventricular systolic dysfunction. RAFT primary investigator Dr Anthony Tang (University of Ottawa Heart Institute, ON) announced results at the AHA meeting.

Dr Anthony Tang

"[RAFT] demonstrated that in people with fewer symptoms of heart failure, this therapy also helped to prevent them from developing further heart-failure symptoms or going into the hospital because of HF and also improved their longevity," Tang told heartwire . "Many of these people are just getting an ICD . . . because that's the proven therapy, and we've now proven that many of these people probably should now at least be given the option of CRT."

Commenting on the RAFT findings, Dr Gregg Fonarow (University of California, Los Angeles) told heartwire , "This is the important randomized clinical-trial evidence that we needed. There were certainly data from REVERSE and MADIT-CRT that showed benefit as far as hospitalizations, but many physicians remained on the fence about whether CRT would be worth the additional lead, with the potential complications, and the expense compared with just giving these patients an ICD or leaving them with just medications and not providing a device," he said.

Dr John GF Cleland (University of Hull, Kingston-upon-Hull, UK) calls the RAFT findings "a great result that is much clearer than with MADIT-CRT," because the mortality rate of patients with only an ICD was much higher in RAFT than in MADIT-CRT (26.1% vs 7.3%). "MADIT-CRT had a weirdly low event rate, perhaps reflecting slightly younger, less symptomatic patients. Almost 30% of MADIT-CRT patients were not receiving a diuretic, and they much less ischemic heart disease than in RAFT."

"This is a pretty important finding across a wide spectrum of patients," Fonarow said. "It will likely lead to revisions in the guidelines and changes in physician practice patterns, with more eligible patients being treated and benefiting from this important therapy." Fonarow stressed that the survival benefit of CRT-D over ICD alone was fairly consistent across most of the subgroups in the RAFT study. "There was really broad-based benefit."

RAFT randomized 1798 patients with NYHA class II or III heart failure, left ventricular ejection fraction <30%, and a QRS duration >120 ms (or paced QRS >200 ms) to either ICD therapy alone or an ICD with CRT (CRT-D). The mean follow-up time for all patients was 40 months.

The primary outcome, all-cause death or heart-failure hospitalization, occurred in 33.2% of the CRT-D group and in 40.3% of the ICD-only group (hazard ratio 0.75, p<0.001). By the last follow-up, 186 out of the 894 CRT-D patients died compared with 236 of the 904 patients in the ICD-alone group. The five-year actuarial death rates were 28.6% for the CRT-D group vs 34.6% for the ICD-alone group (hazard ratio [HR] 0.75, p=0.003). At these rates, one life is saved for every 14 patients treated with CRT-D instead of ICD alone over five years. There were 174 heart-failure hospitalizations in the CRT-D group and 236 heart-failure hospitalizations in the ICD-alone group. One heart-failure hospitalization is prevented for every 11 patients treated with CRT-D instead of ICD-alone over five years.

RAFT builds upon the findings of the previous landmark heart-failure-device therapy studies. COMPANION showed that CRT-D improved survival over optimal medical therapy alone in patients with left ventricular dysfunction, a wide QRS complex, and severe heart-failure symptoms. In a similar population, CARE-HF showed CRT pacing without an ICD (CRT-P) improved survival over optimal medical management. MADIT-CRT showed that CRT-D reduced heart-failure medical encounters over medical therapy alone in New York Heart Association (NYHA) class I and II patients but did not show a mortality improvement.

There were 118 device- or implantation-related periprocedural complications in the CRT-D group vs 61 in the ICD-alone group, but no complication-related fatalities. Most of the additional complications in the CRT-D group were related to the implantation of the extra lead, including reinterventions to reposition a lead that had dislodged or device pocket problems requiring revision.

"There is a price to pay, so to speak, for the patient. [CRT] is a lengthier procedure; it includes more hardware, and therefore more things can go wrong. However, what you gain out of it is longevity and [fewer] hospitalizations," Tang said. CRT-D implants are also more expensive than ICDs alone. Tang said his group is investigating cost-effectiveness data to determine whether the reduction in the cost of heart-failure hospitalizations offsets the additional procedure and device costs of implanting a CRT-D instead of an ICD.

The improvements in survival and hospitalization with CRT-D over ICD alone were similar for both the class II and class III patients, analyzed separately, to the improvement seen in the study population as a whole. However, the improvement in survival for the class III patients (hazard ratio 0.79) did not reach statistical significance (p=0.14). Tang said, "The magnitude of improvement appears to be the same [in the class III and class II patients] from a statistical point of view, but we didn't reach statistical significance, only because we didn't have enough patients. So we can't be absolutely 100% as sure [of the benefit] as we are for the class II patients, but we're relatively sure."

Why Does CRT Work?

The only statistically significant interaction between treatment and a subgroup in the RAFT results was QRS duration. Patients with a QRS of 150 ms or longer showed a greater benefit with CRT-D over ICD alone than patients with a QRS under 150 ms. This subgroup result concurs with MADIT-CRT but "deserves further research," Tang et al note. There was also a nonsignificant trend toward patients with left bundle branch block benefiting more from CRT than the study population as a whole.

"It's a matter of taste whether you go for the QRS story or left bundle branch block story, but despite my earlier doubts, it now seems to be stacking up in favor of QRS being important [as a marker of who will benefit from CRT], although final proof is awaited," Cleland said.

"The QRS duration is really an indicator of electrical dyssynchrony, and it does seem that greater the degree of electrical dyssynchrony, the greater the magnitude of benefit," Fonarow said. "It's very important to recognize that in the prior studies with the narrow QRS, we have not seen a benefit of cardiac resynchronization in those patients. There were many who speculated that the target [of CRT] was actually mechanical dyssynchrony, and that that could be measured by echocardiograms or other techniques and that you didn't need to necessarily have a wide QRS. . . . But the trials that were done to look at this population with narrow QRS, in marked distinction to the trials that have focused on patients with wide QRS, showed no benefit."

Commenting on the study at the AHA conference, Dr Clyde Yancy (Baylor University Medical Center, Dallas, TX) said, "A compelling argument can be made that a sign of significant likelihood of responsiveness and benefit begins at a QRS duration of 150 ms, which is strikingly different from what currently appears in clinical practice guidelines. We may ultimately need to revisit which thresholds we recommend that should trigger the indication for CRT."

Dr Clyde Yancy

Dr Leslie Saxon (University of Southern California, Los Angeles) agreed that the exact mechanism whereby CRT improves heart-failure outcomes is not entirely clear, but "one thing is really clear, and that is that as positive as this group of studies are--REVERSE, RAFT, MADIT-CRT--it's equally clear that when you start backing up into a less sick group with a more preserved EF and a narrow QRS, the data are pretty weak. And it's also very clear that echo measures are horrible predictors of response."

Saxon believes that CRT works "because it does a lot of the things like a beta blocker does. It causes neurohormonal remodeling, decreases the sympathetic tone, and decreases the downregulation of beta receptors into the heart. It really causes molecular repair, but you have to have some kind of electrical issue to trigger that effect."

In an editorial accompanying the RAFT results [2], Dr Arthur Moss (University of Rochester, NY) predicts that ongoing substudies from the major CRT trials may answer some of the important questions about CRT and its applications, such as whether CRT can prevent recurrent heart failure or inhibit ventricular tachycardia or fibrillation. Also, future research may also help to refine the CRT device implant technique by identifying the best branch of the coronary vein in which to locate the left ventricular pacing lead.

Cleland also suggests that, since CRT-D patients had much better outcomes than ICD-only patients and sudden cardiac death is so rare in patients with mild heart failure symptoms, it is worth considering whether CRT-P (no ICD) would be the best option for patients with mild heart-failure symptoms, a wide QRS, and low left ventricular ejection fraction. "I wonder if CRT alone would have done just as well or even better than CRT-D. I think the added value of ICD to CRT is small," he said. "But we don't really know that."

Saxon agrees that the benefits of the ICD in this population are probably small and implanting CRT-P instead of CRT-D devices in this population would save money, but the risk of sudden death in even class I patients is not zero. "If you lose one person to sudden death in class I heart failure because they've gotten so much better on CRT, it's pretty much a tragedy, and how many less-than-60-year-old tragedies are you going to accept to save however many dollars?"

RAFT was supported by the Canadian Institutes of Health Research and Medtronic of Canada. Tang also reports grants from St Jude Medical, Sanofi-Aventis, and Boehringer Ingelheim.

Comments

3090D553-9492-4563-8681-AD288FA52ACE

processing....