Advanced cardiac monitoring in patients and the use of implantable cardiovascular electronic devices may increase the detection of atrial fibrillation (AF), but the devices have limitations, including short battery life and lack of immediate feedback. Can new smartphone tools that can record an electrocardiogram (ECG) strip and make automated diagnosis overcome these limitations and facilitate timely diagnosis? The largest study to date, in Canadian Journal of Cardiology, published by Elsevier, found that using these devices is challenging in patients with abnormal EKGs. Better algorithms and machine learning could help these tools provide more accurate diagnoses, the researchers say.
“Earlier studies confirmed the accuracy of the Apple Watch for diagnosing AF in a limited number of patients with similar clinical profiles,” explained lead researcher Marc Strick, MD, PhD, LIRYC Institute, Bordeaux University Hospital, Bordeaux, France. “We tested the accuracy of the Apple Watch ECG app to detect AF in patients with various concomitant ECG abnormalities.”
The study included 734 consecutively hospitalized patients. Each patient underwent a 12-lead ECG, immediately followed by a 30-second recording on the Apple Watch. The smartwatch’s automated single-lead ECG AF detections were classified as “no evidence of atrial fibrillation,” “atrial fibrillation,” or “inconclusive reading.” Recordings from the smart watch were given to an electrophysiologist who performed a blinded interpretation, assigning each tracing a diagnosis of “AF”, “no AF”, or “unclear diagnosis”. A second blinded electrophysiologist interpreted 100 randomly selected traces to determine the degree to which observers agreed.
In approximately one in five patients, the smartwatch’s ECG failed to make an automatic diagnosis. The risk of false-positive autodetection of AF is higher in patients with premature atrial and ventricular contractions (PAC/PVC), sinus node dysfunction, and second- or third-degree atrioventricular block. In patients with AF, the risk of false-negative tracing (missed AF) is higher in patients with ventricular conduction abnormalities (interventricular conduction delay) or rhythms controlled by an implanted pacemaker.
Cardiac electrophysiologists had a high level of agreement in distinguishing between AF and non-AF. The smartphone app correctly identified 78% of patients who were in AF and 81% who were not in AF. Electrophysiologists identified 97% of patients who were in AF and 89% who were not.
Patients with PVCs were three times more likely to have false-positive AF diagnoses by smartwatch ECG, and identification of patients with atrial tachycardia (AT) and atrial flutter (AFL) was very poor.
“These observations are not surprising because automated smartwatch detection algorithms are based solely on cycle variability,” Dr. Strick noted, explaining that PVCs cause short and long cycles, which increases cycle variability. “Ideally, an algorithm would better distinguish PVCs from AF. Any algorithm limited to cycle variability analysis will have poor accuracy in detecting AT/AFL. Machine learning approaches can increase AF detection accuracy of smartwatch in these patients.”
In an accompanying editorial, Andres F. Miranda-Arboleda, MD, and Adrian Baranchuk, MD, Department of Cardiology, Kingston Health Science Centre, Kingston, Ontario, Canada, noted that this is the first “real world” study. focused on using the Apple Watch as a diagnostic tool for AF.
“This is extremely important because it allowed us to learn that the Apple Watch’s performance in diagnosing AF is significantly affected by the presence of underlying ECG abnormalities.” In a certain way, the algorithms of the smart watch for detecting AF in patients with cardiovascular diseases are not yet intelligent enough. But maybe they will be soon,” said Dr. Miranda-Arboleda and Dr. Baranchuk.
“With the increasing use of smartwatches in medicine, it is important to know which medical conditions and ECG abnormalities may affect and alter the smartwatch’s detection of AF in order to optimize care for our patients,” said Dr. Strick. “The detection of AF with a smart watch has great potential, but is more challenging in patients with pre-existing heart disease.”
Materials provided by Elsevier. Note: Content may be edited for style and length.