January 09, 2017
2 min read

Model helps predict which patients with HF may benefit from ICD

You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

A model designed to estimate mortality from sudden death predicted which patients with HF were likely to die of sudden death and could benefit from an implantable cardioverter defibrillator.

Wayne C. Levy, MD, professor of medicine in the division of cardiology at the University of Washington, Seattle, and colleagues developed the Seattle Proportional Risk Model (SPRM) to estimate the proportion of mortality due to sudden death vs. other causes of death. They validated the model prospectively in the HF-ACTION cohort and determined whether benefit from an ICD varied with SPRM score.

The variables in the SPRM score include age, sex, diabetes, BMI, systolic BP, ejection fraction, NYHA class, sodium, creatinine and digoxin use.

Wayne C. Levy

Risk prediction

“We found in 2006 that the mode of death changed with Seattle Heart Failure Model predicted mortality. In patients with a higher absolute mortality, the proportion of sudden death decreased,” Levy told Cardiology Today. “The question was whether we could improve on the Seattle Heart Failure Model to show who would benefit more from a defibrillator. We found in the SCD-HeFT cohort in 2009 that low-risk patients benefited from an ICD and high-risk patients had no benefit once the annual mortality was about 25%. Our concern was that a low-risk patient who is 60 years old is very different from a low-risk patient who is 80 years old. We devised a method to calculate the proportional risk of sudden death and published it in Heart Rhythm in 2015. We have now validated it in the HF-ACTION cohort.”

Levy and colleagues analyzed 1,947 patients with HF for a median of 2.5 years. An ICD or cardiac resynchronization therapy defibrillator was used by 62% of the population.

According to the researchers, SPRM was predictive of sudden death compared with nonsudden death in those without an ICD (P = .002).

Those with an ICD were less likely to die of any cause than those without one (HR = 0.63; P = .0002), and the ICD benefit related to all-cause mortality varied by SPRM score (P = .001), with those with a higher probability of sudden death deriving greater benefit from ICD use, the researchers wrote.

“The model discriminated reasonably well for mode of death,” Levy told Cardiology Today. “If you just use NYHA class and EF, the receiver operating characteristic is 0.52, which is virtually no different from a flip of a coin; the [receiver operating characteristic] for the model was 0.63. We found people who had a high proportional risk for sudden death had an enormous benefit from the [ICD], whereas people in the lowest quartile of sudden death risk had minimal benefit. Below approximately 32% proportional risk for sudden death, there was no benefit from the defibrillator in HF-ACTION.”

Decision aid

A website enabling anyone to use the model will be launched soon, Levy said.

“I’m hoping the SPRM will help patients and providers make decisions about whether or not to place defibrillators, or replace a generator in a patient that has never had their defibrillator go off,” he said.

Further research should be conducted to determine which younger patients with NYHA class IV HF, who are generally denied an ICD, should get one, Levy said. – by Erik Swain

For more information:

Wayne C. Levy, MD, can be reached at Division of Cardiology, University of Washington, Box 356422, 1959 NE Pacific St., Seattle, WA 98195; email: levywc@uw.edu.

Disclosure: Levy reports consulting for Biotronik, receiving a research grant from Medtronic and serving on a clinical endpoint committee for St. Jude Medical.