Durham, NC - In a new study estimating the cost-effectiveness of defibrillator implantation in patients who met the Multicenter Automatic Defibrillator Implantation Trial II (MADIT II) eligibility criteria, Duke researchers have concluded that patients who receive the implantable cardioverter defibrillator (ICD) would be expected to live 1.8 years longer at a cost of $50 500 per extra year of life gained when compared with patients who received standard medical therapy.[1] Although the number of patients eligible for an ICD would nearly double, investigators say that based on conventional standards, the device is "economically attractive."

The results of the study and editorial are published in the April 19, 2005 issue of the Annals of Internal Medicine.

"If you compare this therapy with other therapies that are widespread, I think ICD therapy in this patient population, assuming the benefit of the ICD remains constant over time, is certainly cost-effective," Dr Sana Al-Khatib (Duke University Medical Center, Durham, NC) told heartwire. "But I think we can make this therapy more attractive if we can reduce the cost of the defibrillator and lengthen the life of the battery significantly."

In an editorial accompanying the published study, Drs Stephen Pauker, Mark Estes, and Deeb Salem (Tufts-New England Medical Center, Boston, MA) write that refining methods to select patients for ICDs will be necessary if the costs of the devices are to be kept in line with the benefits.[2]

"The clinician needs to know the risks, benefits, and costs of the ICD among the subgroups of patients who may be candidates for this therapy," write the editorialists. "As long as the risks, impact on the quality of life, and costs of ICDs continue to be more than minimal concerns, the logical approach is to be selective, focusing on patients who are at highest risk for sudden cardiac death after MI and who do not have substantial comorbid conditions."

As there are still no precise methods for identifying patients who will benefit most from ICD therapy, they note that many well-intentioned clinicians may choose to err on the side of caution by recommending ICD therapy in a patient at relatively low risk for sudden death. To keep cost-effectiveness in the acceptable range, Pauker, Estes, and Salem recommend using the cardiovascular profile of patients in MADIT II as their guide.

MADIT II demonstrated a significant mortality benefit for ICD therapy in patients with MI and advanced left ventricular dysfunction. To project beyond the 20-month follow-up reported in the study, the researchers performed a cost-effectiveness analysis on patients in the Duke cardiovascular database with cardiovascular profiles similar to those patients in MADIT II.

The study sample consisted of 1281 patients who were 21 years of age and older with a history of MI and an ejection fraction of less than 30% and who underwent cardiac catheterization. Despite meeting the MADIT II criteria, Al-Khatib and colleagues report that patients in the Duke database were younger, underwent revascularization less frequently, and had a higher mean ejection fraction and a shorter time between MI and enrollment.

None of these patients received an ICD, having been treated before the MADIT II results were published. The researchers used the 31% benefit seen in MADIT II and calculated the number of life-years gained had the patients been treated with an ICD, as well as the total cost of care for ICD patients compared with the total cost of care for patients receiving standard medical therapy.

Investigators found that the projected life expectancy for the ICD group was 10.88 years compared with 8.26 years for the medical-therapy group. After adjustment of these future health outcomes to present-day values, the ICD-treated group would be expected to live 1.8 years longer than those treated with conventional therapy. After estimating the in-hospital and ICD costs, the researchers report that the cost of ICD therapy per life-year gained works out to $50 500.

"We found that the number of patients eligible for an ICD is not trivial, but in terms of doing the cost-effectiveness analysis, it is good to see that the benefit that was observed in the MADIT II trial can be extrapolated to other settings like ours," said Al-Khatib. "The cost-effectiveness of implanting ICDs in these patients appears to be reasonable and comparable to other therapies that we currently use."

Al-Khatib and colleagues note that the cost-effectiveness of ICD therapy becomes even more favorable if the cost of the ICD is reduced to $10 000. In this hypothetical scenario, the cost-effectiveness ratio is $45 200 per life-year gained. The researchers note that if the battery life doubles to 10 years, the cost-effectiveness ratio is even better, decreasing to $42 500 per life-year gained.

In their paper, Al-Khatib and colleagues write that if the experiences of the Duke study were generalized, the results would imply that approximately 32 000 patients would meet MADIT II criteria for an ICD, doubling the number of ICDs implanted annually.

In their editorial, Pauker, Estes, and Salem write that this conservative estimation would increase the cost of implantation to $1 billion. They point out that other estimates have placed the financial burden of ICD therapy between $5 billion and $15 billion annually, depending on which eligibility criteria are used.

The editorialists note that the study is limited in that costs of ICD therapy are projected from the Duke database, a point the Duke researchers concede. A more complete analysis based on the actual patient care costs in MADIT II will provide better information about cost-effectiveness. That analysis is expected soon.