Apple’s AFib history FDA qualified for trials

Can it reliably estimate AFib burden, and what are its key limitations?

On 1 May 2024, the US Food and Drug Administration (FDA) qualified Apple Watch’s atrial fibrillation history feature (AFib History) as a biomarker test to estimate AFib burden in clinical trials.^1,2 It is the first digital health technology to receive this qualification under the Medical Device Development Tools (MDDT) programme, meaning the FDA considers the feature reliable enough for use in clinical trials of cardiac ablation devices without needing additional review.^1-4

The AFib History feature is now FDA qualified to evaluate AFib burden estimates, for use as a secondary effectiveness endpoint in clinical trials of the safety and effectiveness of cardiac ablation devices²

It is intended for use in people ≥22 years of age diagnosed with AFib.²

How does the AFib History feature estimate AFib burden?

• The Apple Watch’s photoplethysmography (PPG) sensor opportunistically monitors the user’s pulse rate and identifies episodes of irregular heart rhythm that are considered consistent with AFib²
• The AFib History feature analyses the data and retrospectively estimates weekly AFib burden (the proportion of time spent in AFib during the previous week of Apple Watch wear)^2,5

The Apple Watch must be worn for 12 hours a day, 5 days a week to generate AFib burden estimates.⁶

How accurate is the AFib burden estimate?

The performance of the AFib History feature was evaluated in a prospective clinical study of 280 participants with a history of AFib (paroxysmal or permanent).

• The average difference between the AFib weekly burden estimate from the AFib History feature and the reference device (Cardea SOLO Wireless ECG Patch) was 0.67% (95% CI, −0.05 to 1.38%)²
• Most participants (92.9%; 260/280) had paired weekly AFib burden differences within ±5% of the reference device²

Subgroup analysis revealed a difference in performance among those with prior ablation; however, the study was not powered to demonstrate equivalent performance in people with and without prior ablation.² The FDA acknowledged that reduced heart rate variability after AFib ablation could adversely affect the accuracy of the Apple AFib History feature, so further studies are needed to compare its performance with other methods of determining AFib burden in people with prior ablation².

Overall, the FDA considered the retrospective AFib burden estimates from the AFib History feature to be clinically acceptable.²

Advantages and limitations of wearable AFib burden estimates

Accurate assessment of AFib burden can be challenging as it ideally involves continuous recording of the date, time of onset and duration of each arrhythmia episode,^2,7 previously necessitating use of methods such Holter ECG monitoring or implantable devices.^8,9

The FDA noted the Apple AFib History feature may:²

• Address challenges such as patient adherence, potential placebo effects, and technical difficulties in measuring AFib burden without an implantable device
• Reduce barriers to incorporation of AFib history into clinical studies
• Eliminate the need for medical device developers to provide a rationale for AFib history collection methods and cadence, as it provides a weekly AFib burden estimate to compare between treatments (versus other devices, such as Holter monitors, that require justification for the type of device and duration of monitoring). However, the FDA also noted that investigators still need to justify their study design and endpoints
• Allow AFib history data to be captured more easily and consistently between studies

Conversely, limitations of the feature include:

• A lack of data on timing and duration of episodes, ventricular rate during episodes, and other types of arrhythmias²
• Potential for short AFib episodes to not be captured in AFib burden estimates^2,5
• The impact of factors such as motion and ambient temperature on PPG-derived pulse data⁵
• The requirement to correctly fit the device on the wrist, and the absence of monitoring when the user is not wearing the device (e.g., when bathing or charging the device)^2,5

Paving the way for more wearable technologies in clinical trials

Overall, the FDA’s qualification of the AFib History feature opens up opportunities to use more non-invasive technologies in cardiology research, and represents an initial step towards approving other novel and innovative digital technologies for use in clinical trials.⁷

With other non-invasive AFib detection technologies becoming available or in development, including a smart ring¹⁰ and adhesive ambulatory ECG patches,¹¹ this is an exciting time for wearables in clinical trials and healthcare.