ECG analysis may predict risk of death from all causes

Nov 6 2012

Research by Johns Hopkins cardiologists suggests that electrocardiograms (ECGs), which are widely used tests to assess the heart’s electrical activity, may have a greater and more profound future role in predicting the risk of death from any cause, not just heart problems.

In a study initially designed to determine which patients with heart failure were at highest risk of a dangerous heart rhythm disorder called ventricular tachycardia, and death, the researchers analyzed the ECGs of 850 patients using a new computer algorithm that detects abnormal heart rate patterns. They found a significant correlation between the ECG findings using the special analysis and the patients who died within five years of the test.

“From our research, we conclude that a specialized analysis of heart rate patterns using this algorithm holds the potential to tell us about a patient’s overall health, not just his heart, and even ascertain which patients may be at highest risk of death from all causes,” says Deeptankar DeMazumder, M.D., Ph.D., a cardiology fellow at Johns Hopkins Medicine.

DeMazumder will present the findings of the study in a poster session at the American Heart Association Scientific Sessions on November 5, 2012. The presentation is titled “A Novel Strategy Based on Dynamic Analysis of Cardiac Repolarization Predicts Mortality in Patients With Implantable Cardioverter-Defibrillators.”

The findings emerged from a subset of 1,200 patients studied at four medical institutions: The Johns Hopkins Hospital, the University of Maryland Medical Center, the Medical College of Virginia and Washington Hospital Center. All of the patients were participants in the PROSE-ICD study, which stands for PRospective Observational Study of the Implantable Cardioverter Defibrillator in the prevention of sudden death. The goal of the overall study, led by Gordon Tomaselli, M.D., chief of the Division of Cardiology at Johns Hopkins, is to understand the biological mechanisms that increase the risk of dying suddenly from an abnormal heart rhythm. 

All 850 of the patients in the study had heart failure and a normal rhythm on an ECG. With heart failure, the heart is unable to pump enough blood to meet the needs of the body because it becomes progressively weakened and inefficient. The condition usually can be managed effectively with medications, but patients are still at risk of developing dangerous heart rhythms that cause sudden death.

The researchers were trying to predict who would develop potentially fatal heart rhythms and therefore which patients truly needed the implanted defibrillator to shock the heart back into normal rhythm.

“Subtle differences in heart rate patterns on an ECG are not readily apparent without the computerized analysis that we used,” says DeMazumder. He explains this novel method of computer analysis, which is being called EntropyX, can pick up disorganized patterns that occur randomly, and may reflect dysfunction in the interplay among organs and the body as a whole.

“All of our organs are constantly talking with each other and this computer analysis, or algorithm using the ECG, allows us to listen in on that conversation,” DeMazumder says. “The Entropy X algorithm has the potential to determine who may need an implantable defibrillator to shock the heart back into a normal rhythm.”

The findings from the patients analyzed so far showed a significant correlation between the specialized ECG analysis and deaths from all causes within five years of the ECG test. “We accounted for all of the major predictors of death, including age, kidney failure and diabetes, and the results of standard laboratory and imaging tests, as well as the ECG. Results from EntropyX were the strongest predictor,” DeMazumder says. About one-third of the patients in the study died within the five-year time frame.

“A fundamental question is why the prognosis varies among patients with heart failure — why some will do well while others will not,” says Steven Jones, M.D., a cardiologist at Johns Hopkins Medicine and co-author of the study. “This study represents an important observation, but right now we can’t explain why certain variations in heart beat patterns picked up from the ECG predict a higher risk of death.”

The researchers say that they need to validate the findings with larger studies and with different patient populations. An estimated 5 million Americans have heart failure and 550,000 new cases are diagnosed in the United States each year. About 290,000 people die each year from heart failure.

The study is funded by the National Heart, Lung, and Blood Institute, part of the National Institutes of Health, under grant number R01 091062.

For more information:  http://www.hopkinsmedicine.org/heart_vascular_institute/