June 3rd, 2014

Prophylactic ICDs Appear Effective in Less Severe HF Patients

Implantable cardioverter-defibrillators are routinely implanted in heart failure patients with ejection fractions of 35% and lower to prevent sudden cardiac death. However, the benefits in patients at the higher end of the spectrum — EFs between 30% and 35% — have not been well demonstrated in clinical trials, since few patients in this range have been enrolled in such trials.

Now a study published in JAMA suggests that the benefits in this group are similar to the benefits in heart failure patients with more severely depressed EFs.

Sana Al-Khatib and colleagues analyzed data from the National Cardiovascular Data Registry ICD registry and the Get With The Guidelines–Heart Failure database. They compared the mortality benefit associated with ICDs in patients with EFs between 30% and 35% with the benefit in those with EFs below 30%. At follow-up, the ICD-associated reduction in mortality was similar in both groups. At 3 years, among patients with EFs 30%-35%, adjusted mortality rates were 47.1% in those with ICDs versus 58.o% in those without ICDs. Among patients with EFs <30%, mortality rates were 46.1% and 57.o%, respectively. The hazard ratio for mortality among ICD patients with EFs 30%-35% was 0.83 (CI 0.69 – 0.99, p = .04). The HR among ICD patients with EF <30% was 0.72 (CI 0.65 – 0.81, p < 0.001).

The authors write that their findings “support guidelines’ recommendations to implant a prophylactic ICD in eligible patients with an LVEF of 35% or less.” Although “the difference in absolute risk by 3 years was not large (3.6% at 3 years),” the add, “it was significant and close in magnitude to what was observed in the clinical trials of prophylactic ICDs.”


2 Responses to “Prophylactic ICDs Appear Effective in Less Severe HF Patients”

  1. David Powell , MD, FACC says:

    Larry, please check the quoted mortality rates, as the 3.6% does not jive with the 47 and 58%.
    Either way, I am impressed by how much higher mortality is compared with SCD HeFT….all a registry vs. controlled trial effect?
    I am very cynical about the significance of the findings, as the data analysis is fought with so many confounders.

    • David, the quote is correct but I understand why you might be confused. The 3.6% difference can be derived from Table 2, which reports the 3 year _unadjusted_ mortality rates (which I didn’t report in my news summary) in the EF 30-35% groups as 51.4% in the patients who had ICDs and 55% in the patients who did not have ICDs.