Key Points
- Wearable technologies based on photoplethysmographic measurements of irregular pulses can be ubiquitously distributed to perform opportunistic longitudinal monitoring of atrial fibrillation, the most common arrythmia worldwide, and potentially reduce its stroke risk by prompt initiation of anticoagulation.
- The eBRAVE-AF trial is the first trial to randomize elderly patients free of atrial fibrillation to them to digital AF screening via a smartphone app (n=2,860) or usual care (n= 2,691). After 6 months of observation, the digital care approach generated an odds ratio of identifying and treating atrial fibrillation of 2.12 (1.19–3.76, p=0.010), compared to usual care. The stroke event rate was too low during this short observation period to draw any definitive conclusions about the long term clinical impact of either strategy.
Atrial fibrillation (AF) remains the most common arrhythmia worldwide, and continues to be a growing epidemic. The contemporary management of AF relies on two pillars: stroke prevention and arrythmia control. While there are ongoing debates about the best way(s) to control the arrythmia, there is an established and streamlined pathway to successfully achieve stroke prevention, through prompt anticoagulation in patients at high risk of subsequent stroke. Given the often asymptomatic nature of the disease, prompt identification of atrial fibrillation has been an elusive target that may be best addressed by leveraging the potential for opportunistic longitudinal monitoring via wearable technologies.
Photoplethysmographic (PPG) sensors on smart devices such as mobile phones and watches can detect irregularities of pulse waves indicative of atrial fibrillation, and are ubiquitously distributed in many patient households worldwide. To date, a few consumer-oriented observational studies have proven the feasibility and unique potential of this approach, namely that a smart device can identify individuals with AF. However, none of these studies were randomized, and the central pillar of treatment-relevance of the detected AF remained unclear.
During the 2022 European Society of Cardiology Conference in Barcelona, Dr. Axel Bauer presented the results of the eBRAVE-AF trial (NCT04250220). Between 4/2020 and 7/2020, this study remotely identified and recruited 5,551 elderly patients from the pool of a large German health insurance, and randomized them to digital AF screening (n=2,860) or usual care (n= 2,691). Eligible patients had to be >50 years of age, free of known AF and on no anticoagulation at baseline, and with a projected CHAD2DS2VAsc score ≥1 for males and ≥2 for females.
The primary efficacy endpoint, analyzed by intention to treat principle, was newly diagnosed AF leading to initiation of anticoagulation. The digital screening for AF was achieved with the Preventicus Heartbeats app, which was remotely downloaded by patient on any of their smartphones. Then, patients were instructed to perform repetitive 1-min PPG self- measurements (by placing their index fingers on the camera), initially twice a day for 14 days, then bi-weekly over the course of 6 months. Any abnormal PPG measurement was then confirmed by external 14-day loop recorder or by clinical diagnosis.
Enrolled patients had a mean age of 65 years, 30% females with a mean CHAD2DS2VAsc score of 3. At the end of the 6 month study period, in the digital screening arm a total of 38 new cases of AF leading to anticoagulation were identified in the digital screening arm (1.3%), compared to 17 cases in the usual care arm (0.6%), generating an odds ratio of identifying and treating AF of 2.12 (1.19–3.76, p=0.010) in the digital screening arm, compared to usual care. In terms of hard outcomes, there were 12 strokes in the digital screening arm (0.4%) versus 11 strokes (0.4%) in the usual care arm (OR 1.0, 0.5 – 2.35, p=0.950).
In conclusion, the eBRAVE-AF Trial demonstrated that a scalable digital screening strategy using any smartphone significantly increases the odd of detecting new-onset atrial fibrillation and initiating anticoagulation. The very short observation time of <1 year likely is insufficient to detect a sufficient number of stroke events and draw any definitive conclusions at this time with regard to the impact of this digital screening technology on long-term outcomes, and further dedicated studies will be forthcoming.
