Study Boosts FDA Approval Prospects for Mayo-Backed Wearable ECG Device

Powered by the Mayo Clinic’s AI, research claims it can read blood potassium with a touch.

Study Boosts FDA Approval Prospects for Mayo-Backed Wearable ECG Device
A new study has added momentum to a commercialization push for a Mayo Clinic-backed wearable device and smartphone app which analyzes user-generated electrocardiograms to detect high blood potassium levels – something which can now only be done in doctors’ offices.
The device is the KardiaBand for Apple Watch and an associated smartphone app, made by AliveCor of Mountain View, California, a privately-held medtech firm and Mayo Clinic Ventures portfolio company. Its collaboration with the Rochester clinic includes the licensed use of predictive artificial intelligence algorithms developed by Mayo researchers to interpret electrocardiograms generated merely by a user’s touch.
The KardiaBand has already been cleared by the U.S. Food and Drug Administration to record, store and transmit high-quality electrocardiogram (ECG) rhythms, which can be used to detect irregular heartbeats, or atrial fibrillations, in users with heart conditions.
But the company and its Mayo collaborators have loftier goals for the device: They are also working toward FDA approval for its use to detect high blood potassium levels, a condition known as hyperkalemia. This condition is commonly seen in congestive heart failure, chronic kidney disease and diabetes – it can also be caused by the very medications used to treat these conditions.
Doctors say hyperkalemia is associated with significant mortality and heart arrhythmias, but because there are frequently no symptoms, detecting it is difficult. ECGs so far have been useful only in spotting severe cases; thus, testing for hyperkalemia now requires a blood draw in a clinical setting, making it inconvenient for patients to get regular check-ups.
The need for such regular testing, however, is becoming more acute with the growing prevalence of heart and kidney diseases and their risk factors such as high blood pressure and diabetes. And that, in turn, means the population of patients at risk for abnormally high or low concentrations of blood potassium is growing along with it.
Dr. Paul Friedman, a Mayo cardiologist and one of AliveCor’s key medical collaborators, recently told the Harvard Business Review the researchers are moving swiftly to gain FDA approval for a method of using the KardiaBand to address this hyperkalemia dilemma.
“Our goal was to create a test that people could do at home, without drawing blood,” he said. “Detecting abnormal blood potassium levels earlier would allow medicines to be more effectively titrated, problems to be addressed before they occurred, hospitalizations to be prevented, and costs to be reduced.”
Now, a new study co-authored by Friedman is adding to evidence claiming to validate the ability of Mayo’s algorithms to non-invasively diagnose high blood potassium levels via the app’s readings.
Part of the study data was generated by analyzing records from Mayo’s vast archives, including 709,000 patients, 2.1 million ECGs and four million serum potassium values. It found that using the proprietary AI, high blood potassium levels could be accurately identified from “two-lead” ECG readings with 85 percent sensitivity and 72 percent specificity.
Also as part of the study, a group of 10 patients undergoing hemodialysis were supplied with a prototype AliveCor two-lead smartphone app. In that group, the study claimed, the device’s sensitivity for hyperkalemia was 94 percent, with specificity at 74 percent.
“The AI algorithm is a viable tool for detecting even mild hyperkalemia from a two-lead ECG,” the study authors concluded. “In combination with a smartphone ECG, this may enable a new means for rapid, non-invasive monitoring and screening for hyperkalemia at home and in the clinic.”
Mayo has made several rounds of financial investments in AliveCor, including participation in a $30 million Series D venture capital round last year, led by Omron Healthcare.
Newsletter Sign Up