Fact checked byRichard Smith

Read more

February 03, 2023
2 min read
Save

Manual review necessary to rule out AF in one-fourth of smartwatch ECG tracings

Fact checked byRichard Smith
You've successfully added to your alerts. You will receive an email when new content is published.

Click Here to Manage Email Alerts

We were unable to process your request. Please try again later. If you continue to have this issue please contact customerservice@slackinc.com.

An analysis of five ECG-capable smartwatches found comparable sensitivity and specificity for detection of atrial fibrillation between devices; however, tracings were labeled inconclusive in about one in four cases.

Following manual review by cardiologists, rhythm was determined in nearly all inconclusive tracings, according to results of the prospective, single-center, diagnostic BASEL Wearable study published in JACC: Clinical Electrophysiology.

Smart watch
An analysis of five ECG-capable smartwatches found comparable sensitivity and specificity for detection of AF between devices; however, tracings were labeled inconclusive in about one in four cases.
Source: Adobe Stock

“In this clinical validation of five direct-to-consumer smart devices, we found a high diagnostic accuracy among all assessed devices when excluding inconclusive tracings,” Diego Mannhart, MD, of the department of cardiology and the Cardiovascular Research Institute Basel, University Hospital Basel, University of Basel, Switzerland, and colleagues wrote. “We found differences in the number of inconclusive tracings. In a clinical setting, manual review of tracings is required in about one-fourth of cases for all assessed devices.”

Researchers enrolled 201 patients (median age, 67 years; 31% women) presenting to the University Hospital Basel and scheduled for catheter ablation procedures, electric cardioversions, pacemaker or implantable cardioverter defibrillator implantation. At enrollment, participants underwent a preprocedural 12-lead ECG followed by 30-second, single-lead ECG recordings using one of five smartwatches (Apple Watch 6, AliveCor KardiaMobile or KardiaMobile 6L, Fitbit Sense, Samsung Galaxy Watch 3 and Withings Scanwatch ECG).

All smartwatch and 12-lead ECGs were anonymized and distributed to two masked cardiologists who independently interpreted each ECG and assigned it as sinus rhythm, AF or inconclusive. The automated interpretations were compared with the cardiologist-reviewed 12-lead ECGs, using an intention-to-diagnose approach, and then compared with the cardiologist-interpreted single-lead ECGs.

AF was present in 31% of the cohort.

Researchers reported that the sensitivity and specificity for the detection of AF were comparable between the five devices:

  • 85% and 75% for the Apple Watch 6;
  • 85% and 75% for the Samsung Galaxy Watch 3;
  • 58% and 75% for the Withings Scanwatch;
  • 66% and 79% for the Fitbit Sense; and
  • 79% and 69% for the AliveCor KardiaMobile.

Single-lead ECGs were labeled inconclusive by the smartwatches at a rate between 17% and 26% (P for pairwise comparison < .01).

Upon manual review of inconclusive tracings, the masked cardiologists were able to determine rhythm in 99% of the single-lead ECGs.

“A careful validation of results should be made by an experienced physician before conferring the diagnosis of AF based on a single-lead ECG, especially with regard to possible therapeutic consequences and measures based on these findings,” the researchers wrote. “In addition, patient perception differed between devices. This clinical validation study of these smart devices may help to better advice patients and physicians in the usage and validity of single-lead ECG devices for everyday use.”