The experts on SCD-HeFT: Implement now, but seek more discriminating ICD patient-selection criteria

March 10, 2004

New Orleans, LA - The evidence base for using an implantable cardioverter defibrillator (ICD) in a large population of patients with HF expanded considerably with the recent release here of findings from the Sudden Cardiac Death in Heart Failure Trial (SCD-HeFT). Clinicians contacted after the trial's public presentation felt the 23% decrease in five-year mortality risk associated with the device was impressive enough to apply right away in clinical practicewith a few caveats.

Entry into SCD-HeFT required patients to be in NYHA class 2-3 HF, which could be of either ischemic or nonischemic etiology, and have an LVEF <35%. These expansive criteria could potentially apply to several million patients in the US alone.

Dr Alfred E Buxton

The ICD benefit in SCD-HeFT was "modest," said Dr Alfred E Buxton (Brown University, Providence, RI), "but clearly the effect was there. . . . A patient who meets SCD-HeFT criteria is a worthwhile patient to consider for an ICD, assuming there aren't correctable factors that could improve the ejection fraction and heart-failure status to class 1." Buxton notes, however, that the SCD-HeFT patients were exceptionally well treated with beta blockers and either ACE inhibitors or angiotensin receptor blockers. It is incumbent on the referring physicians, he said, to ensure the patient is receiving optimal medical therapy.

"Stop with the primary conclusion"

Dr Douglas P Zipes

When Dr Gust Bardy (Seattle Institute for Cardiac Research, WA) presented the results of SCD-HeFT here, he emphasized the tentative and inconclusive nature of the available outcomes data in various patient subgroups. Dr Douglas P Zipes (Indiana University School of Medicine, Indianapolis) goes further. The SCD-HeFT primary results, he said, "are totally consistent with previously published data," referring particularly to the first and second Multicenter Automatic Defibrillator Implantation Trial (MADIT and MADIT II)[1,2], and the Multicenter Unsustained Tachycardia Trial (MUSTT)[3]. Now, "I think the SCD-HeFT analysis should stop. . . . The study was wonderfully done, and we need to stop with the primary conclusion."

[The SCD-HeFT primary results]. . . are totally consistent with previously published data.

Further "slicing and dicing" of the data to glean possible treatment effects in patient subgroups would yield only hypothesis-generating insights, not anything on which to base treatment recommendations, according to Zipes. And not all of the tentative subgroup findings appear to make sense.

For example, the mortality hazard ratio with the ICD was reduced in NYHA class 2 patients, with confidence intervals both below one, and elevated in NYHA class 3 patients. "That the ICD was not as beneficial in class 3 vs class 2 is totally inconsistent with prior studies," Zipes said. "There's no reason to think that the ICD should be less effective in class 3."

Dr Barry M Massie

"I think it's a fluke," agreed Dr Barry M Massie (University of California, San Francisco). "Although the ICD seemed to work better in class 2 than in class 3 [patients], all other therapies work better the more advanced the heart failure. Sudden deaths account for most of the deaths in class 2 patients. In class 3 patients, a lower proportion of the deaths are accounted for by sudden death. A relative risk and a relative benefit would be smaller in class 3, but the absolute benefit may actually be bigger."

A relative risk and a relative benefit would be smaller in class 3, but the absolute benefit may actually be bigger.

That the nearly one half of the patients in SCD-HeFT with nonischemic cardiomyopathy shared in the survival benefit pleased Dr David L Hayes (Mayo Clinic, Rochester, MN). Such patients, at least those who have not had an out-of-hospital cardiac arrest, he said, are a group "in whom we've had trouble justifying [ICD treatment], short of performing an EP stimulation study." The SCD-HeFT primary outcome, he said, "lends more credence to changing the guidelines."

Role of risk stratification

Dr Kim A Eagle

The ICD should be used now in populations that meet the entry criteria from the MADIT II and SCD-HeFT trials, provided an effort is made to refine those criteria further, according to Dr Kim A Eagle (University of Michigan, Ann Arbor). "Because MADIT II and SCD-HeFT were both similar in the risk reduction that they reported, we would have to conclude that there is enough evidence to move forward."

We would have to conclude that there is enough evidence to move forward.

Zipes, agreeing that the SCD-HeFT criteria should now be a basis for selecting patients for ICD therapy, cautioned that the search for refinements in those criteria "needs to be done in a prospective, randomized fashion." The Centers for Medicare & Medicaid Services (CMS), he said, "should rescind their restriction of a QRS duration of at least 120 milliseconds." If any criterion is used to risk-stratify further, Zipes said, it should be T-wave alternans. "Its negative predictive value is superb, 98-plus percent."

The QRS-duration restriction on reimbursement from the CMS after the MADIT II trial seemed to please few clinicians. SCD-HeFT, observed Hayes, "should get the ball rolling" so that the federal agency drops the requirement and permits reimbursement for fully FDA-approved ICD indications in Medicare patients. "We'll have to jump through new hoops with the FDA and CMS to get first labeling and then reimbursement. We know how long that took with MADIT II, so I think we've got our work cut out for us."

We'll have to jump through new hoops with the FDA and CMS. . . . I think we've got our work cut out for us.

Leaner, meaner ICDs

Echoing other voices at the scientific sessions, Eagle suggested one way to reduce the potentially enormous cost of providing ICDs for all who meet the MADIT II and SCD-HeFT criteria. Using simpler, less costly devices in most patients would be a good start, he said.

Massie noted that the SCD-HeFT investigators "used a device with no bells and whistles. My understanding is that such a device is available via the internet for $9000. . . . We have beta blockers that cost $2000 per year. If you put in a device, and the patient has it for five years, well, then [ICD cost] doesn't sound so bad, does it?"

Model 7223 units from Medtronic Inc (Minneapolis, MN) were used in SCD-HeFT, Bardy said in his formal presentation of the trial.

NASPE-Heart Rhythm Society Responds to SCD-HeFT

The SCD-HeFT findings are "welcome news," said Dr Michael E Cain (Washington University, St Louis, MO), president of the North American Society for Pacing and Electrophysiology-Heart Rhythm Society, in a statement issued after the trial's presentation here. "The positive results underscore the benefit of implantable cardioverter defibrillators as effective early intervention tools for more patients with heart disease, even those with mild to moderate heart failure without previous heart attacks."

In light of the new findings, the statement continued, "we recommend that CMS expand ICD coverage to those Medicare beneficiaries who meet the criteria established in the MADIT II and SCD-HeFT trial."

"These results also emphasize the national need to continue to develop and implement new methods to further identify patients who will benefit from life-saving ICD therapy."





Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.
Post as: