Device-Detected Subclinical Atrial Fibrillation in the Wearable Era: When Is Anticoagulation Justified?

Introduction: Navigating the New Landscape of Wearable Arrhythmia Detection

Wearable technologies have fundamentally changed how frequently clinicians encounter possible atrial fibrillation. They have also created a distinct clinical challenge. Increasingly sensitive monitoring identifies rhythms whose clinical significance is not necessarily equivalent to conventional symptomatic or ECG-documented AF.

The first decision a clinician must make is diagnostic rather than therapeutic. An optical smartwatch alert, an ECG-confirmed episode, and an atrial high-rate episode (AHRE) recorded by an implanted cardiac device are related findings, but they are certainly not interchangeable.

This distinction matters immensely. The strongest randomized evidence supporting anticoagulation in SCAF comes principally from patients monitored continuously with implanted cardiac devices, rather than from otherwise healthy individuals who simply receive a single consumer smartwatch notification.

Why a Wearable Alert Does Not Equal an Atrial Fibrillation Diagnosis

The 2023 ACC/AHA/ACCP/HRS guideline recommends that an initial AF diagnosis be made by a clinician through visual interpretation of an electrocardiographic signal, regardless of the monitoring technology that generated the recording. Intracardiac device detections should also be confirmed by reviewing stored tracings because artifact and other arrhythmias can easily produce false-positive detections.

Photoplethysmography (PPG) provides entirely different information. Smartwatch PPG algorithms can identify irregular pulse patterns consistent with possible AF, but the ACC/AHA/ACCP/HRS guideline specifically states that PPG signals alone are not sufficiently reliable to establish an AF diagnosis. The FDA likewise describes cardiovascular notifications from consumer wearables as alerts that may prompt further medical evaluation rather than definitive diagnostic determinations.

Large wearable studies illustrate exactly why confirmation matters. In the Apple Heart Study, 34% of participants with an irregular-pulse notification who subsequently returned an analyzable ECG patch had AF recorded during later patch monitoring. Notifications that occurred concurrently with ECG monitoring showed substantially higher concordance with true AF.

Similarly, in the Fitbit Heart Study, 32.2% of participants who received an irregular-heart-rhythm notification and completed analyzable subsequent ECG-patch monitoring had AF identified. When a recurrent Fitbit irregular-rhythm detection occurred while the ECG patch was actively being worn, the positive predictive value for concurrent AF jumped to 98.2%.

These studies strongly support wearable technology as a useful detection tool. They do not, however, establish that anticoagulation should automatically follow an optical irregular-rhythm notification.

Distinguishing Clinical AF, AHRE, and Subclinical AF

The 2024 ESC guideline defines clinical AF as AF documented by an ECG. It defines device-detected SCAF as asymptomatic AF identified during continuous monitoring, which includes implanted cardiac electronic devices and appropriate consumer-based monitors, provided there is confirmation by professional review of an ECG-recorded rhythm or intracardiac electrogram.

An AHRE is commonly an automated device classification. It may represent AF, atrial flutter, atrial tachycardia, artifact, or another rhythm entirely. It should therefore be confirmed before any treatment decision is made.

A practical clinical sequence naturally follows: detection, rhythm confirmation, episode characterization, stroke-risk assessment, bleeding and safety assessment, and finally the treatment decision. Skipping rhythm confirmation risks applying potent anticoagulation evidence to a patient who may not actually have AF.

Why Subclinical AF Requires a Different Approach Than Clinical AF

Oral anticoagulation has a well-established role in appropriately selected patients with clinical AF. However, the absolute thromboembolic risk associated with brief device-detected SCAF is generally much lower.

Both ARTESiA and NOAH-AFNET 6 enrolled populations with substantial conventional stroke-risk factors, yet their untreated thromboembolic event rates remained near 1% per patient-year. That lower baseline event rate matters notably because a similar relative treatment effect produces a much smaller absolute reduction in stroke when the untreated risk is already low.

The LOOP trial reinforces this crucial distinction. Among 6,004 older adults with stroke risk factors but no known AF, implantable-loop-recorder screening substantially increased AF detection and oral anticoagulant use. AF was diagnosed in 31.8% of the monitored group versus 12.2% with usual care, and anticoagulation was initiated in 29.7% versus 13.1%, respectively. Despite substantially greater AF detection and treatment, the incidence of stroke or systemic arterial embolism was not significantly reduced.

Greater AF detection therefore does not automatically mean that every newly detected episode is a valid anticoagulation target.

The ARTESiA Trial: Evidence That Anticoagulation Can Prevent Stroke

ARTESiA randomized 4,012 patients with device-detected SCAF lasting 6 minutes to 24 hours to receive either apixaban or aspirin. The mean age was 76.8 years, the mean CHA₂DS₂-VASc score was 3.9, and mean follow-up was approximately 3.5 years.

Stroke or systemic embolism occurred at a rate of 0.78% per patient-year with apixaban and 1.24% per patient-year with aspirin, corresponding to a hazard ratio of 0.63.

This reduction in thromboembolism was accompanied by more major bleeding. In the prespecified on-treatment safety analysis, major bleeding occurred at 1.71% per patient-year with apixaban and 0.94% with aspirin, corresponding to a hazard ratio of 1.80.

ARTESiA therefore established two findings simultaneously. First, anticoagulation can reduce thromboembolic events in appropriately selected patients with SCAF. Second, that benefit carries a definitively increased risk of major bleeding.

The trial should not be interpreted to mean that six minutes of SCAF constitutes an automatic anticoagulation threshold. ARTESiA enrolled an older population with substantial baseline stroke risk and required qualifying SCAF plus clinical risk factors for enrollment.

The NOAH-AFNET 6 Trial: Evidence Against Routine Treatment for Every AHRE

NOAH-AFNET 6 randomized 2,536 patients aged at least 65 years with device-detected AHRE lasting at least 6 minutes and additional stroke-risk factors to receive edoxaban or placebo. Aspirin was permitted when otherwise clinically indicated.

The trial was terminated early because of safety concerns and an informal assessment of futility. The primary efficacy endpoint of cardiovascular death, stroke, or systemic embolism occurred at 3.2% per patient-year with edoxaban and 4.0% with placebo. This produced a hazard ratio of 0.81 that did not reach statistical significance. Stroke occurred at approximately 1% per patient-year in both groups.

The composite safety outcome of death or major bleeding occurred more frequently with edoxaban, yielding a hazard ratio of 1.31.

NOAH-AFNET 6 therefore argues strongly against treating every AHRE as though it carries the exact same stroke risk as established clinical AF.

Reconciling the Findings of ARTESiA and NOAH-AFNET 6

A study-level meta-analysis combining the two trials included 6,548 participants. Oral anticoagulation with apixaban or edoxaban reduced ischemic stroke, with a relative risk of 0.68, but it increased major bleeding, with a relative risk of 1.62.

The pooled analysis found no noteworthy reduction in cardiovascular death or all-cause mortality.

Taken together, the trials demonstrate a consistent clinical trade-off rather than a simple positive-versus-negative result. Anticoagulation can prevent ischemic stroke in device-detected AF, but the absolute value of that benefit depends heavily on baseline stroke risk and must always be weighed against the additional bleeding risk.

Episode Duration Matters, but There Is No Proven Biological Cutoff

Observational evidence initially placed substantial emphasis on episode duration. In the ASSERT trial, patients older than 65 years with hypertension and implanted pacemakers or defibrillators were followed for device-detected SCAF. Episodes lasting more than 24 hours were associated with an increased subsequent risk of stroke or systemic embolism, with an adjusted hazard ratio of 3.24. Episodes lasting 6 minutes to 24 hours were not associated with a statistically significant increase compared with no SCAF.

These findings helped support the duration-based framework subsequently incorporated into the ACC/AHA/ACCP/HRS guideline.

Randomized-trial secondary analyses, however, show exactly why episode duration should not be treated as a precise biological cutoff. Within the qualifying range of ARTESiA (below 24 hours), neither baseline SCAF frequency nor the longest episode duration notably modified the treatment effect of apixaban.

A prespecified NOAH-AFNET 6 analysis examined 259 patients with baseline AHRE lasting at least 24 hours. Only two strokes occurred in each treatment group, and no notable interaction between AHRE duration and the effect of edoxaban was demonstrated. The analysis was explicitly hypothesis-generating.

Duration remains clinically relevant, but current evidence simply does not identify a single number of minutes or hours at which anticoagulation abruptly becomes beneficial.

Current Recommendations from the ACC/AHA/ACCP/HRS Guideline

For patients without a previous diagnosis of AF, the 2023 ACC/AHA/ACCP/HRS guideline provides three duration-and-risk categories.

Device finding	Stroke-risk context	Guideline approach
AHRE 24 hours or longer	CHA₂DS₂-VASc 2 or higher, or equivalent risk	Oral anticoagulation is reasonable within shared decision-making
AHRE 5 minutes to 24 hours	CHA₂DS₂-VASc 3 or higher, or equivalent risk	Oral anticoagulation may be reasonable within shared decision-making
AHRE less than 5 minutes	No other anticoagulation indication	Oral anticoagulation should not be given for the AHRE alone

These recommendations are Class 2a, Class 2b, and Class 3 (No Benefit), respectively.

The ACC framework deliberately combines episode duration and underlying thromboembolic risk rather than treating AF duration as an isolated trigger.

Current Recommendations from the ESC Guideline

The 2024 ESC guideline incorporates both ARTESiA and NOAH-AFNET 6 and takes a somewhat less threshold-driven approach.

It concludes that direct oral anticoagulant (DOAC) therapy may be considered in selected patients with asymptomatic device-detected SCAF who have a high estimated stroke risk and no major bleeding risk factors. The ESC guideline explicitly states that the duration and burden of SCAF that best identify patients likely to benefit from oral anticoagulation remain uncertain.

For conventional clinical AF, the 2024 ESC guideline emphasizes the CHA₂DS₂-VA score, which omits the sex category, rather than simply reproducing the US CHA₂DS₂-VASc framework. The ESC SCAF recommendation itself is deliberately framed around high estimated stroke risk rather than a single validated SCAF score cutoff.

The two guidelines can therefore be reconciled clinically. The US guideline supplies practical duration-and-risk thresholds, while the ESC guideline emphasizes that randomized evidence still does not define a universal SCAF-duration threshold.

Why Baseline Stroke Risk May Matter More Than SCAF Burden

A prespecified ARTESiA analysis examined outcomes according to the CHA₂DS₂-VASc score.

Among patients with a score greater than 4, stroke or systemic embolism occurred at 0.98% per year with apixaban and 2.25% per year with aspirin. Apixaban prevented an estimated 1.28 stroke or systemic-embolism events per 100 patient-years and was associated with an estimated 0.68 additional major bleeds per 100 patient-years.

Among participants with CHA₂DS₂-VASc scores below 4, the absolute reduction in stroke or systemic embolism was much smaller. Patients with a score of 4 represented an intermediate group in which individual treatment preferences assumed greater importance.

These results support risk refinement but should not be converted into a rigid new guideline rule stating that every patient with a CHA₂DS₂-VASc score above 4 and SCAF must receive anticoagulation.

How Previous Stroke or TIA Can Shift the Benefit-Risk Balance

A prespecified ARTESiA subgroup analysis included 346 participants with a previous stroke or TIA.

Stroke or systemic embolism occurred at 1.20% per year with apixaban versus 3.14% per year with aspirin, corresponding to a hazard ratio of 0.40. Over approximately 3.5 years, the estimated absolute reduction in stroke or systemic embolism was 7% in the prior stroke or TIA subgroup.

The estimated absolute increase in major bleeding in this subgroup was approximately 3% over the same interval.

These findings suggest that previous cerebral ischemia may identify a population in whom the absolute benefit from anticoagulation is substantially larger. Nevertheless, this remains a subgroup analysis and should not be generalized to every short-duration wearable-detected episode following a stroke.

Insights from the 2026 Implantable Cardiac Monitor Analysis

A final 2026 ARTESiA subgroup analysis examined the 209 trial participants whose SCAF was detected by an implantable cardiac monitor rather than a pacemaker or defibrillator.

Among these participants, stroke or systemic embolism occurred at 0.3% per year with apixaban and 2.6% per year with aspirin. The subgroup was small, patients monitored with implantable cardiac monitors differed clinically from the much larger pacemaker and ICD population, and treatment-by-device interaction testing did not establish that device type independently altered the treatment effect.

The finding is clinically relevant, particularly because implantable monitors are increasingly used after cryptogenic stroke, but it should not be used to create a new device-specific anticoagulation threshold.

A Practical Step-by-Step Clinical Approach

Clinical situation	Interpretation	Practical response
PPG smartwatch irregular-rhythm alert only	AF not yet established	Obtain ECG-quality confirmation; do not anticoagulate from the notification alone
ECG or intracardiac electrogram confirms AF	Rhythm established	Determine whether this represents conventional clinical AF or device-detected SCAF and quantify episode characteristics
Confirmed AHRE or SCAF less than 5 minutes	Very brief episode	Do not anticoagulate solely for the AHRE in the absence of another indication
AHRE 5 minutes to 24 hours plus CHA₂DS₂-VASc 3 or higher	Intermediate-duration SCAF with elevated risk	Anticoagulation may be reasonable after individualized shared decision-making
AHRE 24 hours or longer plus CHA₂DS₂-VASc 2 or higher	Longer-duration AHRE with elevated risk	Anticoagulation is reasonable under the ACC framework after confirmation and individualized assessment
High baseline thromboembolic risk	Greater opportunity for absolute stroke reduction	Benefit-risk balance may favor anticoagulation more strongly
Previous stroke or TIA	Higher recurrent-stroke risk	ARTESiA subgroup evidence supports stronger consideration of apixaban if bleeding risk is acceptable
Major bleeding risk, interacting drugs, or competing comorbidity	Potentially reduced net benefit	Correct modifiable risks and reconsider net benefit before treatment

Why Bleeding Risk Must Be Central to the Primary Decision

The increased bleeding risk observed with anticoagulation is central to the SCAF decision because baseline stroke rates are relatively modest.

A 2025 prespecified ARTESiA bleeding analysis included 3,961 treated participants. Major bleeding occurred at 1.71 versus 0.94 events per 100 patient-years with apixaban and aspirin. Gastrointestinal major bleeding was higher with apixaban, at 0.89 versus 0.40 per 100 patient-years, whereas intracranial and fatal bleeding rates were not statistically different between groups.

This analysis provides additional context but does not erase the overall major-bleeding signal. The clinical objective remains to identify patients whose absolute reduction in disabling thromboembolism is likely to justify anticoagulant exposure.

Apixaban Dosing, Safety, and Regulatory Considerations

ARTESiA tested apixaban 5 mg twice daily, reduced to 2.5 mg twice daily when indicated.

Current US prescribing information for nonvalvular AF similarly recommends apixaban 5 mg orally twice daily for most adults. The labeled dose is 2.5 mg twice daily when at least two of the following are present: age 80 years or older, body weight 60 kg or less, or serum creatinine 1.5 mg/dL or higher.

SCAF itself does not create a separate reduced-dose regimen. A patient should not receive empiric low-dose apixaban merely because the AF is subclinical. Dose selection should strictly follow the applicable labeling criteria.

The apixaban label does not provide a separate FDA-defined duration threshold for device-detected SCAF. Selection of patients with SCAF for anticoagulation is therefore informed by randomized trial evidence and professional-society guidance, while dose selection and medication safety follow current prescribing information.

Apixaban is contraindicated in patients with active pathological bleeding or severe hypersensitivity. The prescribing information contains boxed warnings concerning increased thrombotic risk after premature discontinuation and spinal or epidural hematoma associated with neuraxial procedures.

Combined P-glycoprotein and strong CYP3A4 inhibitors can increase apixaban exposure and may require dose reduction or avoidance, depending on the existing dose. Combined P-glycoprotein and strong CYP3A4 inducers such as rifampin, carbamazepine, phenytoin, and St. John’s wort should be avoided because they decrease apixaban exposure. Concomitant antiplatelet agents, other anticoagulants, and chronic NSAID use increase bleeding risk.

Current labeling does not recommend apixaban in severe hepatic impairment. It is not recommended during pregnancy, and breastfeeding is not recommended during treatment.

Managing Patients When Anticoagulation Is Deferred

Deferring anticoagulation should never mean ending follow-up.

The ESC guideline reports progression from device-detected SCAF to clinical AF at approximately 6% to 9% per year. A 2025 ARTESiA analysis found progression to clinical AF or SCAF lasting more than 24 hours in 31.2% of participants during follow-up, corresponding to 9.3% per patient-year.

Stroke-risk factors are highly dynamic. Aging, hypertension, diabetes, heart failure, vascular disease, prior cerebral ischemia, increasing AF burden, or the development of conventional clinical AF can materially change the benefit-risk calculation.

Periodic reassessment is therefore an integral component of any conservative management strategy.

Acknowledging the Limitations of Current Evidence

ARTESiA and NOAH-AFNET 6 enrolled predominantly older patients with appreciable stroke-risk profiles. Their results should not be extrapolated directly to younger, low-risk smartwatch users.

ARTESiA studied qualifying SCAF lasting 6 minutes to 24 hours and therefore does not directly establish the efficacy of continuing the exact same strategy for substantially longer device-detected episodes.

Randomized evidence for patients presenting initially with AHRE lasting 24 hours or longer remains limited, and the NOAH-AFNET 6 analysis of episodes 24 hours or longer was underpowered for definitive treatment conclusions.

Consumer wearable trials primarily establish detection performance rather than clinical benefit from anticoagulation.

The CHA₂DS₂-VASc, previous stroke or TIA, bleeding, and implantable-monitor analyses are subgroup or secondary analyses. They are useful for refining treatment decisions but should not be converted into independent guideline mandates.

Neither randomized trial establishes a universally valid SCAF-burden threshold.

Clinical Implications for Everyday Practice

Device-detected AF is best approached as a continuum rather than as a binary smartwatch finding.

At one end is an unconfirmed optical irregular-rhythm notification. At the other is established clinical AF, for which conventional stroke-prevention recommendations apply. Between these categories lies confirmed device-detected SCAF, where episode characteristics and baseline thromboembolic risk must be interpreted together.

A clinically useful framework incorporates several questions simultaneously:

• Was the rhythm truly AF?
• How much SCAF occurred?
• What is the patient’s baseline probability of thromboembolism?
• Has the patient already had a stroke or TIA?
• What bleeding risk accompanies anticoagulation?
• Does the patient meet labeled dosing criteria?
• How does the patient value reduction in stroke risk relative to additional bleeding risk?

Conclusion: A Tailored Approach to Device-Detected AF

A smartwatch irregular-rhythm notification alone is never an indication for oral anticoagulation. Confirmation of AF using an interpretable ECG tracing or intracardiac electrogram must always come first.

Once SCAF is confirmed, current evidence supports selective anticoagulation rather than routine treatment of every episode. The ACC/AHA/ACCP/HRS guideline considers anticoagulation reasonable for AHRE lasting 24 hours or longer with a CHA₂DS₂-VASc score of 2 or higher, and notes it may be reasonable for AHRE lasting 5 minutes to 24 hours with a CHA₂DS₂-VASc score of 3 or higher. The ESC guideline takes a less threshold-driven position, stating that DOAC therapy may be considered in selected patients with high estimated stroke risk and no major bleeding risk factors, while acknowledging that the optimal SCAF duration and burden remain uncertain.

ARTESiA demonstrates that apixaban can reduce stroke or systemic embolism in selected patients with SCAF, whereas NOAH-AFNET 6 and the bleeding findings from both randomized trials show exactly why anticoagulation cannot be an automatic response to every device-detected episode.

The most defensible current approach is to confirm the rhythm, quantify the episode, assess thromboembolic and bleeding risk, and anticoagulate selectively only when the expected absolute benefit clearly justifies the risk.

References

1. Joglar JA, Chung MK, Armbruster AL, et al. 2023 ACC/AHA/ACCP/HRS Guideline for the Diagnosis and Management of Atrial Fibrillation. J Am Coll Cardiol. 2024;83(1):109-279. https://doi.org/10.1016/j.jacc.2023.08.017 PMID: 38043043. Accessed August 7, 2026.
2. US Food and Drug Administration. Post-Market Evaluation of Smartwatch Cardiovascular Notifications. FDA. https://www.fda.gov/science-research/advancing-regulatory-science/post-market-evaluation-smartwatch-cardiovascular-notifications Accessed August 7, 2026.
3. Van Gelder IC, Rienstra M, Bunting KV, et al. 2024 ESC Guidelines for the management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J. 2024;45(36):3314-3414. https://doi.org/10.1093/eurheartj/ehae176 PMID: 39210723. Accessed August 7, 2026.
4. Healey JS, Lopes RD, Granger CB, et al; ARTESiA Investigators. Apixaban for Stroke Prevention in Subclinical Atrial Fibrillation. N Engl J Med. 2024;390(2):107-117. https://doi.org/10.1056/NEJMoa2310234 PMID: 37952132.
5. Kirchhof P, Toennis T, Goette A, et al; NOAH-AFNET 6 Investigators. Anticoagulation with Edoxaban in Patients with Atrial High-Rate Episodes. N Engl J Med. 2023;389(13):1167-1179. https://doi.org/10.1056/NEJMoa2303062 PMID: 37622677.
6. McIntyre WF, Benz AP, Becher N, et al. Direct Oral Anticoagulants for Stroke Prevention in Patients With Device-Detected Atrial Fibrillation: A Study-Level Meta-Analysis of the NOAH-AFNET 6 and ARTESiA Trials. Circulation. 2024;149(13):981-988. https://doi.org/10.1161/CIRCULATIONAHA.123.067512 PMID: 37952187.
7. Perez MV, Mahaffey KW, Hedlin H, et al; Apple Heart Study Investigators. Large-Scale Assessment of a Smartwatch to Identify Atrial Fibrillation. N Engl J Med. 2019;381(20):1909-1917. https://doi.org/10.1056/NEJMoa1901183 PMID: 31722151.
8. Lubitz SA, Faranesh AZ, Selvaggi C, et al. Detection of Atrial Fibrillation in a Large Population Using Wearable Devices: The Fitbit Heart Study. Circulation. 2022;146(19):1415-1424. https://doi.org/10.1161/CIRCULATIONAHA.122.060291 PMID: 36148649.
9. Svendsen JH, Diederichsen SZ, Højberg S, et al. Implantable loop recorder detection of atrial fibrillation to prevent stroke (The LOOP Study): a randomised controlled trial. Lancet. 2021;398(10310):1507-1516. https://doi.org/10.1016/S0140-6736(21)01698-6 PMID: 34469766.
10. Van Gelder IC, Healey JS, Crijns HJGM, et al. Duration of device-detected subclinical atrial fibrillation and occurrence of stroke in ASSERT. Eur Heart J. 2017;38(17):1339-1344. https://doi.org/10.1093/eurheartj/ehx042 PMID: 28329139.
11. McIntyre WF, Benz AP, Healey JS, et al. Risk of Stroke or Systemic Embolism According to Baseline Frequency and Duration of Subclinical Atrial Fibrillation: Insights From the ARTESiA Trial. Circulation. 2024;150(22):1747-1755. https://doi.org/10.1161/CIRCULATIONAHA.124.069903 PMID: 39229707.
12. Becher N, Toennis T, Bertaglia E, et al. Anticoagulation with edoxaban in patients with long atrial high-rate episodes ≥24 h. Eur Heart J. 2024;45(10):837-849. https://doi.org/10.1093/eurheartj/ehad771 PMID: 37956458.
13. Lopes RD, Granger CB, Wojdyla DM, et al. Apixaban vs Aspirin According to CHA₂DS₂-VASc Score in Subclinical Atrial Fibrillation: Insights From ARTESiA. J Am Coll Cardiol. 2024;84(4):354-364. https://doi.org/10.1016/j.jacc.2024.05.002 PMID: 39019530.
14. Shoamanesh A, Field TS, Coutts SB, et al. Apixaban versus aspirin for stroke prevention in people with subclinical atrial fibrillation and a history of stroke or transient ischaemic attack: subgroup analysis of the ARTESiA randomised controlled trial. Lancet Neurol. 2025;24(2):140-151. https://doi.org/10.1016/S1474-4422(24)00475-7 PMID: 39862882.
15. Xing LY, Lopes RD, McIntyre WF, et al. Apixaban for stroke prevention in subclinical atrial fibrillation detected by an implantable cardiac monitor: A subgroup analysis of ARTESiA. Heart Rhythm. 2026;23(7):1476-1482. https://doi.org/10.1016/j.hrthm.2025.07.048 PMID: 40752875.
16. Siegal DM, Sticherling C, Healey JS, et al. Major Bleeding With Apixaban vs Aspirin: A Subanalysis of the ARTESiA Randomized Clinical Trial. JAMA Cardiol. 2025;10(12):1305-1314. https://doi.org/10.1001/jamacardio.2025.4151 PMID: 41222953.
17. DailyMed. ELIQUIS (apixaban) tablets, film coated: Prescribing Information. Revised August 2025. https://dailymed.nlm.nih.gov/dailymed/drugInfo.cfm?setid=7be1f4c1-bb2f-4ded-ae9a-515d2a22f93e Accessed August 7, 2026.
18. Boriani G, McIntyre WF, Ramasundarahettige C, et al; ARTESiA Investigators. Atrial fibrillation progression in patients with device-detected subclinical atrial fibrillation: Insights from the ARTESiA trial. Heart Rhythm. 2025;22(12). https://doi.org/10.1016/j.hrthm.2025.07.002 PMID: 40645433.