Cleveland Clinic Studies Accuracy of Apple Watch 4 for Atrial Fibrillation Detection

Cleveland Clinic researchers have found that the Apple Watch 4 (AW4) identified only 41% of atrial fibrillation (AFib) instances when viewing the watch display alone. In addition to the watch display, the AW4 provides a downloadable PDF to detect AFib, which detected AFib – a type of irregular heartbeat – 98% of the time.

In the study, researchers compared both types of readings to a telemetry electrocardiogram (EKG or ECG), which is a test that continuously measures the electrical activity of the heartbeat while the patient is the hospital.

“A standard ECG remains the gold standard for detecting Afib,” said Marc Gillinov, M.D., Cleveland Clinic chair of thoracic and cardiovascular surgery and the Judith Dion Pyle Chair in Heart Valve Research. “At this point, consumer wearables and watches don’t have the accuracy to replace the ECG. A diagnosis of Afib requires input from a physician.”

The study was published today in Circulation.

The study looked at 50 postoperative cardiac surgery patients at Cleveland Clinic on telemetry ECG monitoring. Heart rhythm assessments were obtained three times a day (morning, early/late afternoon, and early/late evening) over two days, resulting in six assessments per patient. Of the possible 300 rhythm assessments, researchers obtained 292 readings. Two patients were discharged before the completion of the study. The researchers chose to study cardiac surgery patients because these patients frequently develop Afib while in the hospital. Afib is a common after cardiac surgery.

“The data suggests that further technological advances are necessary before these wearables can be fully incorporated into current clinical management,” says Milind Desai, M.D., Cleveland Clinic cardiologist and director of clinical operations and Haslam Family Endowed Chair in cardiovascular medicine.

During each reading, patients were provided an Apple Watch 4 at random and provided with instructions on how to activate the heart rhythm reading on the app. At the same time, an automatically-generated PDF of the waveform on the Apple Health App was exported and a rhythm strip were saved. The ECGs were generated and viewed by a cardiologist according the instructions provided by Apple.

Half of the patients had more than one AFib occurrence during their hospital stay. The telemetry ECG picked up all 90 AFib instances.

The results:

• AW4 app display only
  • Correctly identified 34 of the 90 (41%) instances of atrial fibrillation
  • In the 25 patients who had more than one instance of AFib, AFib was identified in 19
  • In patients whose hearts were in sinus, or normal rhythm, there were no notifications of afib, indicating there were no false positives
• AW4 with a retrievable PDF of the waveform strips
  • 284 rhythm assessments were gathered
  • In eight instances a reading failed to generate, this was counted as incorrect rhythm interpretations
  • Correctly identified 84 of the 90 (96%) instances of atrial fibrillation
  • In the 25 patients who had more than one instance of AFib, AFib was shown in 24
  • In patients whose hearts were in sinus, or normal rhythm , zero instances of Afib were picked up, indicating there were no false positives

According to the American Heart Association, about 2.7 million Americans are living with AFib. Atrial fibrillation (AF or AFib) is the most common irregular heart rhythm that starts in the atria. Some people live for years with atrial fibrillation without problems. Atrial fibrillation is associated with an increased risk of stroke, heart failure and even death.