What You Should Know:
– The FDA granted Emergency Use Authorization to Eko for an ECG-based algorithm that screens for low ejection fraction, as part of an assessment of cardiac complications and co-morbidities associated with COVID-19.
– This ECG-based deep learning algorithm, developed on a large clinical dataset in collaboration with the Mayo Clinic, can help efficiently identify signs of possible heart failure in patients.
Eko, a digital health company building AI-powered screening and telehealth solutions to fight cardiovascular disease, today announced that the U.S. Food and Drug Administration (FDA) has issued the company an Emergency Use Authorization (EUA) for its novel ECG-based algorithm that can provide an easily accessible, rapid screening test for low ejection fraction (low EF), a weak heart pump.
COVID-19 Impact on Heart Disease
Awarded an FDA Breakthrough Device designation in December 2019, the algorithm’s regulatory review has been further accelerated due to its potential to assist providers in identifying patients with abnormal heart function during the COVID-19 pandemic.
It is known that patients with cardiovascular disease have a much greater risk of death from COVID-19 than those with normal hearts. Patients with pre-existing cardiovascular disease are estimated by the American College of Cardiology to have a 10.5% case fatality rate due to COVID-19, significantly higher than those without such conditions. COVID-19 infection may also directly cause heart muscle weakness. The Emergency Use Authorization allows healthcare providers to use Eko’s AI algorithm to analyze a standard 12-lead ECG, an easily accessible and painless test to assess the risk of a weakened heart pump.
Scope of Emergency Use Authorization (EUA)
The scope of the authorization is limited to the use of the algorithm by a healthcare provider to provide an assessment of LVEF for use as a diagnostic aid to screen for potential cardiac complications associated with COVID-19 or underlying cardiac conditions that may affect clinical management of COVID-19, in adult patients having or suspected of having COVID-19.
The solution should not replace an echocardiogram in cases where an echocardiogram is indicated. The software does not detect any possible cardiac abnormality besides low LVEF. It is not intended for monitoring of patients diagnosed with heart failure. The software is not intended as a sole means of diagnosis and is intended to be used when echocardiography is not yet available or is not indicated.
Eko Electrocardiogram (ECG) Low Ejection Fraction Tool
The algorithm is software as a medical device that detects whether a patient has an LVEF less than or equal to 40% based upon the analysis of ECG signals at the point of care. It analyzes a collected ECG, and within seconds displays a binary prediction of the likelihood of LVEF less than or equal to 40% on a smartphone, tablet, or PC device.
The algorithm will be part of Eko AI, a cloud-based software application program interface (API) that allows a user to upload 12-lead ECG data for analysis. The API can be electronically interfaced and perform analysis with data transferred from 12-lead ECG devices.
Impact of Early Detection of Heart Failure
“Early detection of heart failure patients can better inform treatments for these individuals at higher risk for severe illness from COVID-19,” said Connor Landgraf, CEO and co-founder, Eko. “With the FDA’s Emergency Use Authorization, Eko will be able to offer its low EF screening at Mayo Clinic immediately, the first step in expanding the offering to other providers on the Eko platform.”
Heart failure affects about 5.7 million people in the U.S. alone and is most commonly detected by echocardiogram, a test that is not normally conducted during a physical exam, requires specially trained technicians to record, and requires prolonged contact with the patient. Because of limited access to echocardiography, especially during the COVID-19 pandemic, and because many patients with reduced heart function do not have symptoms, or have symptoms that are attributed to COVID-19, heart failure is frequently diagnosed late, making life-prolonging treatment more challenging.
“Given the danger COVID-19 poses to patients with a weak heart pump, it’s important that we rapidly identify these individuals early and monitor them closely. By embedding the heart failure screening AI into a quick, widely available, and safe test using existing medical devices, we can detect heart failure early and start appropriate treatments,” said Dr. Paul Friedman, Chair of the Department of Cardiovascular Medicine, Mayo Clinic. “Additionally, for people with COVID-19, we may be able to identify when the virus causes the development of a weak heart pump quickly, safely, and easily using these AI tools.”
Peer-Reviewed Results
Eko’s low ejection fraction algorithm employs a deep neural network developed in collaboration with Mayo Clinic. The algorithm was first announced in a landmark publication in Nature Medicine in January 2019, describing for the first time the technology that permits a point of care ECG device combined with an AI algorithm to identify a low ejection fraction in patients.
“It is groundbreaking that an AI-based algorithm can identify signs of ventricular dysfunction in a 12-lead ECG, taking it beyond the ECG interpretation capabilities of even an expert cardiologist,” said Subbu Venkatraman, Ph.D., CTO, Eko. “We look forward to continuing our collaboration with Mayo Clinic to bring this technology to the front lines of care.”
