Accuracy of a smartwatch based single-lead electrocardiogram device in detection of atrial fibrillation
Level 2 - randomized trial
Prospective multi-centre diagnostic validation study against a reference standard 12-lead ECG
OpenAlex W3000250023 · doi:10.1136/heartjnl-2019-316004
What was done
A prospective, multi-centre validation study across three tertiary hospitals evaluated the diagnostic accuracy of the AliveCor KardiaBand (a smartwatch-based single-lead ECG recording device) compared with a gold-standard 12-lead ECG. The device generated automated readings of atrial fibrillation (AF) or sinus rhythm (SR). When unclassified or no-analysis readings occurred, repeat recordings were obtained, and two blinded electrophysiologists independently reviewed the unclassified tracings.
What was found
In 200 hospitalized patients (38 with AF, 162 with SR; 439 total ECGs recorded), the automated smartwatch algorithm demonstrated a sensitivity of 94.4%, specificity of 81.9%, positive predictive value of 54.8%, and negative predictive value of 98.4%. Agreement between the 12-lead ECG and the automated device diagnosis was moderate when unclassified tracings were included (κ=0.60, 95% CI 0.47 to 0.72). Combining the automated diagnosis with blinded electrophysiologist interpretation of unclassified tracings improved overall agreement (EP1: κ=0.76, 95% CI 0.65 to 0.87; EP2: κ=0.74, 95% CI 0.63 to 0.86).
Why it matters
Smartwatch-based single-lead ECGs show high sensitivity for AF detection, but moderate specificity leads to frequent false positives, meaning physician review of unclassified tracings remains essential.
Limits
The study evaluated an inpatient tertiary-care sample rather than a broader outpatient screening population. Only 38 patients had AF at the time of testing. Ambulatory motion artifacts and long-term diagnostic utility were not evaluated.
Cited by
- supports Smartwatches can detect atrial fibrillation and measure electrocardiograms.