Apple Watch Helps Detect AF: Is This the Future?

March 16, 2019

NEW ORLEANS — In the first study of its kind involving more than 400,000 members of the general population, using an Apple Watch to identify irregular pulse was found to help in the identification of atrial fibrillation (AF).

The watch identified an irregular rhythm in 0.5% of the whole study population and in those who were subsequently monitored with an electrocardiogram (ECG) patch, it showed a positive predictive value for detecting AF of 0.84.

"The positive predictive value of 84% supports the ability to correctly identify AF among those notified of an irregular rhythm by the watch," co-principal investigator, Marco Perez, MD, Stanford University, California, concluded during his presentation today at the American College of Cardiology 68th Annual Scientific Session 2019 (ACC.19).

"A positive notification by the watch could therefore help inform clinical evaluation along with clinical history and examination."

Fellow co-principal investigator, Mintu Turakhia, MD, Stanford School of Medicine, added: "This study improves our understanding of how this wearable technology and this app works in the real-world setting and how well the technology can detect long periods of AF."

Study chair, Kenneth Mahaffey, MD, Stanford University, commented to | Medscape Cardiology: "I think the results confirm that the algorithm and the precision around the measurements we made are a suitable foundation to move this forward and study this further in how to integrate it into clinical care."

Delegates at the ACC.19 meeting had widely differing opinions on how this technology should fit into clinical practice.  

Discussant of the study at an ACC press conference, Matthew Martinez, MD, chair of the ACC Sports & Exercise Council and a cardiologist at Lehigh Valley Health Network, Allentown, Pennsylvania, was enthusiastic about the new technology. 

"This study shows that using an Apple Watch is a feasible option to help detect AF. This technology is coming whether we like it or not. Like many other instances, the technology here is ahead of the science but we need to try and keep up with it as best we can," he told | Medscape Cardiology.

Martinez said he was "stunned and encouraged" by the 0.5% notification rate of irregular rhythms in the study. "That is far less than I would have anticipated, and makes this approach workable in my opinion." On the positive predictive value of 85% he said, "This is a good number  — it's a great place to start."

But giving the opposite perspective, Ray Gibbons, MD, Mayo Clinic, Rochester, Minnesota, and past ACC president, was clearly not impressed with the Apple study.

"This is very preliminary and not ready for prime time," he told | Medscape Cardiology. "Monitoring for AF is only a good idea if a patient has reasonable likelihood of having AF. The patient group who has AF are older people who don't wear Apple watches. The vast majority of participants in this study were young people, but AF is not major health problem in young people.

"It's like ordering a coronary calcium score for a 35th birthday present," he said. "And for the older population we have multiple alternative technologies to detect AF which have been better validated, so it is not clear where this fits in."

Gibbons believes the new Apple technology will raise healthcare costs without public health benefits. "This is a gift to hospital administrators. It will encourage young people who are very unlikely to have any serious health issues to go to their doctor and have multiple tests that are highly reimbursed." 

He also says it could overload primary care physician who are already besieged with other problems of much greater health consequences. "I would rather they dedicate their time and energy to treating high blood pressure and ensuring compliance for secondary prevention medications for people with known coronary disease."

For the study, 419,297 US residents aged 22 years or older (mean age, 40 years) self-enrolled during an 8-month period and wore the watch, which passively measured heart rate with periodic spot measurements.

Occurrence of irregular rhythm in five of six measurements triggered a notification for the user to contact a telehealth doctor and the mailing of an ECG patch to the user when appropriate. The patch was worn together with the watch for a further week to detect AF.

Results showed that 2162 participants (0.5% of those enrolled) received an irregular pulse notification. This was, as expected, highly age dependent. Of the 200,000 participants under age 40, the notification rate was 0.16% compared with 3.2% in the 25,000 people age 65 or over.   

Turakhia described these figures as "very encouraging, as it tracks with our understanding of AF being more common with age."

In total, 658 participants were sent an ECG patch; 450 patches were returned and included in the analyses.

When comparing pulse detections on the watch with simultaneous ECG patch recordings among 450 participants who wore both at the same time, researchers found that the positive predictive value for an individual irregular rhythm measurement was 71%, and the positive predictive value for the notification (triggered by irregular rhythms detected in five of six measurements) was 84%.

However, only 34% of those who had received an irregular pulse notification on their watch were found to have AF based on wearing an ECG patch over a week later.

Turakhia said this was not entirely unexpected. "AF can come and go, particularly early on in the course of the disease. It's not surprising for it to go undetected in subsequent ECG patch monitoring," he said.

"So while only 34% of people who were still having AF on the ambulatory ECG, that doesn't mean that the other 66% didn't have AF.  It just means that AF may not have been there at the time," he said. "These parameters help us understand how we, as clinicians, should think about these notifications."

Of patients in whom AF was detected, 89% had episodes lasting more than 1 hour. 

Of those who received an irregular pulse notification via the app, about half ended up initiating contact with the study doctor, but researchers said the others may have sought care elsewhere. In subsequent surveys, 57% of people who got an alert said they sought medical attention outside of the study regardless of whether they had been seen virtually by a study doctor.

About one third of people who received a notification had a CHADS-VASc score of 2 or greater.

The study investigators said the obvious next step is to study how this technology can be used to complement other technologies, such as real-time ECG and other validated tools, to manage heart rhythm problems.

Martinez said patients were already embracing this new technology. "I have diagnosed two patients with AF in the last few weeks who came in because they detected abnormal rhythm on their Apple watches. If someone calls me and says they are having an abnormal rhythm of their watch, I ask them to email a copy of the pulse monitoring strip from the phone app," he said.

"I would say to people if you have an abnormal rhythm on their phones, go to see a doctor. If you already have a cardiologist, go and see them," Martinez said. "If you don't, a primary care doctor is a good place to start as there is really no reason they can't order the same tests, such as an ECG patch or other monitoring. To primary care doctors — I would say, Don't ignore it. I would say at least it is worth further assessment for a short period."

Mahaffey added: "My message to primary care physicians is to continue to evaluate patients with a good history, physical exam and assessment of risk factors. The notification in the watch can be used as another data point in their evaluation as to whether further work up needs to be done."

In a more general discussion on how digital technology will affect clinical practice, participants were generally extremely upbeat.

Turakhia made the point that this Apple study "really represents a paradigm shift for how clinical studies can be conducted. We don't have to bring people into a brick and mortar clinic and give the study intervention."

Paul Stoffels, MD, chief scientific officer, Johnson & Johnson, described a new randomized controlled trial starting this year comparing the Apple watch with a control for the diagnosis of AF, which will also track hard clinical outcomes, such as myocardial infarction, stroke, and death. 

"This study will look at whether digital health can have a real impact on events," Stoffels said.

Harlan Krumholz, MD, Yale School of Medicine, New Haven, Connecticut, said, "We are about to see a digital transformation, and there has never been a more exciting time in medicine."

Moderating the panel discussion, John Rumsfeld, MD, ACC chief innovation officer, concluded: "I think we had all better believe that in the future, digital health and medicine will just be called health and medicine."

The Apple study was funded by Apple Inc. Turakhia reports honoraria and consultancy fees from AliveCor and Cardiva Medical. Perez reports consultancy and honoraria from Apple.

American College of Cardiology 68th Annual Scientific Session 2019 (ACC.19). Presented March 16, 2019. 

 For more from | Medscape Cardiology, follow us on Twitter and Facebook


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.