Whilst it is well established that patients with device-detected atrial fibrillation (DDAF) have an increased risk of stroke, there is considerable uncertainty as to whether this risk warrants treatment with oral anticoagulants (OACs). Patients with DDAF have an increased ischaemic stroke risk but still less than those with persistent AF. Balancing this with the significant risks of bleeding associated with OACs has made demonstration of benefit challenging through clinical trials. Similarly, the extent to which clinicians should be screening for DDAF in the general population remains unanswered. The trials within this area are summarised in table 1.
Screening for AF
Trials of AF screening have shown that there is a considerable prevalence of AF in the general population. The LOOP trial explored the benefit of using implantable loop recorders in the general population over age 70yrs to screen for AF. This trial showed that while the AF detection rate was significant at 31.8%, and more patients were started on OACs, there was no benefit in reduction of the primary outcome of stroke or systemic embolism. The STROKESTOP trial used a less invasive screening approach with regular handheld ECG monitoring in 75-76yr olds without a history of AF. Those invited for screening had a higher AF detection rate and consequent initiation of OACs and this was associated with a reduction on the primary endpoint which was a composite of stroke, systemic embolism, bleeding and all-cause death, suggestive of a benefit to screening for AF using devices.
Of course, where these two trials differ is in their method for detection of AF. The LOOP study used a cut-off of 6 minutes of implantable device-detected AF to start anticoagulation, whilst STROKESTOP used ECG screening. It is therefore likely that patients treated in STROKESTOP had to have a significantly higher AF burden for AF to be detected and they represent a higher risk group with a greater absolute stroke risk and hence more likely to benefit from anticoagulation. It still remains unclear whether patients with only 6 minutes of AF are at a raised risk of stroke, and there is a multiplying factor where increasing AF increases the risk. The LOOP study therefore includes patients with a significant heterogeneity in their risk of stroke, ranging from 6 minutes to patients with episodes >24hrs and this may explain the lack of benefit seen.
Dedicated anticoagulation trials
There are now two dedicated trials for using anticoagulation in device-detected AF – ARTESIA and NOAH-AFNET 6. The ARTESIA trial randomised patients with a mean CHA2DS2-VASc score of 3.9 and between 6 minutes and 24hrs of subclinical AF to anticoagulation with apixaban or aspirin. Stroke or systemic embolism was reduced in the apixaban group with a hazard ratio of 0.63, but major bleeding was increased, with a hazard ratio of 1.8. This trial was in contrast to the negative NOAH-AFNET 6 trial which compared edoxaban to placebo in patients with more than 6 minutes of DDAF. This trial was terminated after 21 months due to safety signals with bleeding and detected no significant benefit in the primary endpoint of cardiovascular death, stroke and systemic embolism. However, the event rate in NOAH-AFNET was lower and combined with early termination, this greatly reduces the likelihood of observing a clinically meaningful benefit. Additionally, edoxaban is known to have a higher bleeding rate than apixaban and this likely contributed to early termination of the trial. A combined meta-analysis of the two trials showed a statistically significant reduction in ischaemic stroke endpoint with a consequent increase in major bleeding when OACs were used in DDAF.
Clinical Implications
Decisions regarding anticoagulation in DDAF remain highly individualised. These patients are at increased stroke risk compared to the general population, but at lower risk than those with persistent AF, while bleeding risk from OACs remains similar. Dedicated risk scores are not validated in this setting, representing an ongoing research need. Clinicians must consider conventional stroke risk factors, AF burden and bleeding risk, recognising the complexity of their interaction. Contemporary care should prioritise shared decision-making, with patients central to discussions around risks and benefits.
| Trial | Type of trial | Patient population | Number of patients | Method for AF detection | Definition of AF | Incidence of new AF | Treatment arm | Control arm | Primary outcome | Results | Safety |
|---|---|---|---|---|---|---|---|---|---|---|---|
| STROKESTOP | Open-label RCT | Individuals aged 75-76yrs without AF | 27975 | Handheld ECG device twice daily over a 14 day period | AF on ECG | 3% in the screening arm | Offered AF screening and consequent treatment | Usual care | Combined endpoint of ischaemic or haemorrhagic stroke, systemic embolism, bleeding or all-cause death | Primary endpoint reduced in screening group, 31.9% vs 33.0%, HR 0.96, p=0.045 | Included in primary outcome |
| LOOP | Open-label RCT | Individuals aged 70-90yrs without AF and with at least one additional stroke risk factor | 6004 | Implantable loop recorder | AF episode lasting 6 minutes or longer | 31.8% in the ILR group vs 12.2% in the standard care group | Implantation of ILR and anticoagulation as per clinician choice if AF episode detected | No ILR and standard care | Time to first stroke or systemic arterial embolism | No significant difference in primary outcome, 4.5% vs 5.6%, HR 0.8, p=0.11 | No significant difference in rate of major bleeding, 4.3% of ILR group versus 3.6% of control group, HR 1.26, p=0.11 |
| ARTESIA | Double-blind RCT | Patients over 55yrs with an implanted device which had detected AF and a CHA2DS2-VASc score ≥3 | 4012 | Any implanted device | AF episode lasting 6 minutes – 24hrs | N/A | Apixaban | Aspirin | Stroke or systemic embolism | Primary event reduced in apixaban group, 0.78% per patient-year vs 1.24% per patient year for aspirin, HR 0.63, p=0.007 | Increased major bleeding with apixaban, 1.71% per patient year vs 0.94% per patient year, HR 1.80, p=0.001 |
| NOAH-AFNET 6 | Double-blind RCT | Patients over 65yrs with an implanted device which had detected new AF and one or more risk factor for stroke | 2536 | Any implanted device | Atrial high-rate-episode lasting 6 minutes or longer | N/A | Edoxaban | Placebo | First occurrence of cardiovascular death, stroke or systemic embolism | Trial terminated early No change in primary event, 3.2% per patient year for edoxaban vs 4.0% per patient year for placebo, HR 0.81, p=0.15 |
Death or major bleeding increased in edoxaban group, 5.9% per patient-year vs 4.5% per patient-year, HR 1.31, p=0.03 |
References
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. The Lancet. 2021;398(10310):1507-1516. doi:10.1016/S0140-6736(21)01698-6
Svennberg E, Friberg L, Frykman V, Al-Khalili F, Engdahl J, Rosenqvist M. Clinical outcomes in systematic screening for atrial fibrillation (STROKESTOP): a multicentre, parallel group, unmasked, randomised controlled trial. The Lancet. 2021;398(10310):1498-1506. doi:10.1016/S0140-6736(21)01637-8
Healey JS, Lopes RD, Granger CB, et al. Apixaban for Stroke Prevention in Subclinical Atrial Fibrillation. N Engl J Med. 2024;390(2):107-117. doi:10.1056/NEJMoa2310234
Kirchhof P, Toennis T, Goette A, et al. Anticoagulation with Edoxaban in Patients with Atrial High-Rate Episodes. N Engl J Med. 2023;389(13):1167-1179. doi:10.1056/NEJMoa2303062
Chiv R, Beradid S, Suissa S, Renoux C. Effectiveness and Safety of Edoxaban Compared With Apixaban in Elderly Patients With Nonvalvular Atrial Fibrillation: A Real-World Population-Based Cohort Study. Stroke. 2024;55(5):1161-1170. doi:10.1161/STROKEAHA.123.045098
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. doi:10.1161/CIRCULATIONAHA.123.067512
Author Sub-editor: Dr Atmadeep Banerjee Dr Atmadeep Banerjee is an ST4 Cardiology Registrar in the Kent, Surrey and Sussex deanery. He graduated from Calcutta National
Authors Heartbeat Sub-Editor: Sandeep Singh Dr Sandeep Singh is a Cardiology Clinical Fellow at Nottingham University Hospitals NHS Trust. After graduating from India, he completed
Authors Heartbeat Sub-Editor: Khin Kay Kay Kyaw I am currently working as cardiology specialist registrar ST6 at southwest peninsula region while also working as an
Authors Heartbeat Sub-Editor: Anindya Mukherjee Dr. Anindya Mukherjee is a ST5 trainee registrar in Cardiology at Mid Yorkshire Teaching NHS Trust, UK, with extensive experience
Authors Sub-editor: Sameer Zaman Heartbeat Sub-Editor: Aleena Haider Aleena is an academic cardiology trainee at Imperial College London with a strong commitment to all aspects
