Apple Heart Study: Not just for atrial fibrillation

The Apple Heart Study, a prospective, single-group, open-label, site-free, pragmatic study, demonstrated that wearable technology (Apple Watch) is a feasible means of detecting not only atrial fibrillation (AF) but also other types of arrhythmias in the general population.

Prof. Marco Perez (Stanford University, USA) presented new findings from this study, showing that the pulse sensor paired with a mobile app identified important arrhythmias apart from AF with a high positive predictive value [1]. The study had two co-primary endpoints: (1) AF >30 seconds as validated by ECG patch monitoring, and (2) simultaneous ECG patch-validated AF and irregular patient tachogram. Key secondary outcomes were simultaneous AF on ECG patch monitoring while the pulse notification algorithm detected an irregular pulse, and post-study healthcare provider intervention within 3 months after an irregular pulse notification.

The study, published in the New England Journal of Medicine [2], showed that an irregular pulse was detected in 0.52% of the 419,297 participants between November 2017 and February 2019. Participants were required to own an Apple Watch and iPhone and to install the study app that used an irregular pulse notification algorithm. If the pulse algorithm registered an irregular tachogram, sampling frequency was increased. If 5 out of 6 tachograms within a subsequent 48-hour period were irregular, a notification was sent to the participant, who would be recommended to contact a study physician by video consultation. As a result, 658 participants (0.16% of total) received an ECG patch by mail to wear for up to 7 days, of which 450 were returned for analysis (23% women; 40% >65 years).

Of the 450 participants who returned ECG patches, 18 patients had heart rates >200 bpm, 1 had a pulse pause of >6 seconds, and 1 had ventricular tachycardia for >6 seconds. In this study, 153 (34%) of the returned ECG patches confirmed AF. Among participants who were notified of an irregular pulse, the positive predictive value was 0.84 (95% CI 0.76-0.92) for observing AF on the ECG simultaneously with a subsequent irregular pulse notification. Among 297 participants who were shown to not have AF, 74 (25%) did have premature atrial contraction burden between 1-15%, and 4 (1.3%) participants had a burden of >15%. High-grade atrioventricular block episodes were identified in 3 patients that were linked to transient sinus slowing that lasted <4 seconds. In addition, 2 individuals were identified with premature ventricular contractions of >15%, and 11 participants (3.7%) had an episode of ventricular tachycardia for >8 heartbeats. The investigators concluded that the wearable technology had adequate sensitivity to detect AF as well as rarer arrhythmias in the general population and that this approach could potentially identify asymptomatic individuals at significant risk and direct them to specialised care.

1. 
   
   1. Perez MV, et al. Apple Watch App Identifies Clinically Important Arrhythmias Other Than Atrial Fibrillation: Results From the Apple Heart Study. FS.AOS.04, AHA Scientific Sessions 2019, 14-18 November, Philadelphia, USA.
   2. Perez MV, et al. Large-scale assessment of a smartwatch to identify atrial fibrillation. N Engl J Med 2019;381:1909-1917.

 



Posted on 26 February 202024 February 2021