Introduction

Patients suffering with subclinical atrial fibrillation (AF) are burdened with an increased risk of stroke, though to a lesser extent as compared to those with clinical AF. The ARTESiA trial demonstrated that the direct-acting oral anticoagulant apixaban reduced the risk of stroke or systemic embolism (SE) versus aspirin in people with subclinical AF. Nevertheless, there was an increase in the risk of major bleeding. Consequently, the beneficial role of anticoagulation in this patient subgroup with subclinical AF is under scrutiny. 

Aim

To determine if oral anticoagulation would offer a greater secondary stroke prevention benefit in patients with subclinical AF and a history of stroke or transient ischaemic attack (TIA) who are at an increased risk of recurrent stroke, as compared to those without a history of stroke or TIA.

Patient Profile

• Adult patient (aged ≥ 55 years) with device-detected subclinical AF lasting from 6 min to 24 h and having a CHA2DS2-VASc score of ≥3 (score range: 0-9, higher scores indicating a higher stroke risk).
• Patients aged ≥75 years or those with a history of stroke were eligible in the absence of other risk factors.

Methods

Study Design

• A prespecified subgroup analysis of the ATRESIA trial.
• ARTESiA was a double-blind, double-dummy, randomised controlled trial conducted across 247 sites in 16 countries across Europe and North America. 

Treatment Strategy

• The study subjects were randomized 1:1 to receive either oral apixaban 5 mg twice daily or oral aspirin 81 mg once a day. 
• The dosage was apixaban was reduced to 2·5 mg twice daily in patients meeting at least two of the following criteria: age ≥80 years, bodyweight ≤60 kg, or serum creatinine ≥133 μmol/L

Outcomes

Primary Efficacy Outcome 

Primary Safety Outcome 

Secondary Efficacy Outcomes 

• Incidence of myocardial infarction (MI), TIA (with motor deficit, speech disruption, or a duration of more than 5 min), and cause-specific mortality
• The composite of stroke, MI, SE, or death.

Secondary Safety Outcomes 

Results

• A total of 4012 patients with subclinical AF were randomized to either apixaban (n=2015) or aspirin (n=1997) between years 2015-2021. The median follow-up period for the study was 3.5 years.
• Of the entire study population, 346 (8.6%) subjects had a history of stroke or TIA (172 in the apixaban group and 174 in the aspirin group). 
• Median CHA2DS2- VASc score for subjects with a history of stroke or TIA was 6.0 (IQR 5.0–7.0) and for those without a history of stroke or TIA was 4.0 (IQR 3.0–4.0). Subjects with a history of stroke were at a greater risk of vascular outcomes and adverse events. 
• Amongst the patients with a history of stroke or TIA, those treated with apixaban had a 60% lower annual risk of stroke or SE, as compared to the patients treated with aspirin [1.20% (95% CI 0.48 to 2.48) vs. 3.14% (95% CI 1.86 to 4.96); hazard ratio (HR): 0.40 (95% CI 0.17 to 0.95)] (Fig. 1). 
• Amongst the subjects without a history of stroke or TIA (n=3666; 1843 in the apixaban group and 1823 in the aspirin group), those treated with apixaban had a 31% lower annual risk of stroke or SE vs. those treated with aspirin [0.74% (95% CI 0.55 to 0.98) vs.1.07% (95% CI 0.83 to 1.36); HR 0.69 (95% CI 0.48 to 1.00)] (Fig. 1). 

Fig. 1: The incidence of primary efficacy outcome in the study groups

• The absolute risk difference in incidence of stroke or SE during 3·5 years of follow-up was 7% (95% CI 2 to 12) in subjects with a history of stroke or TIA, as against 1% (0 to 3) in subjects without a history of stroke or TIA. 
• With regards to the annual incident rate of major bleeding amongst subjects with a history of stroke or TIA, the same was higher in patients treated with apixaban vs. aspirin [2.26% (95% CI 1.21 to 3.87) vs. 1.16% (95% CI 0.47 to 2.39); HR: 1·94 (95% CI 0·77 to 4·87). 
• Patients treated with apixaban vs. aspirin had greater absolute risk differences in the secondary outcomes of stroke, ischemic or unknown stroke, disabling or fatal stroke, MI, cardiovascular death, and the composite of stroke, MI, SE, or death in participants with a history of stroke or TIA relative to those without a history (Table 1).

Table 1: Incidence of secondary efficacy outcomes

Secondary Outcome	Patients with a History of Stroke/TIA			Patients without a History of Stroke/TIA
	Apixaban Annual rate (95% CI)	Aspirin Annual rate (95% CI)	HR (95% CI)	Apixaban Annual rate (95% CI)	Aspirin Annual rate (95% CI)	HR (95% CI)
Stroke	1·20% (0·48–2·48)	3·14% (1·86–4·96)	0·40 (0·17–0·95)	0·74% (0·55–0·98)	1·04% (0·80–1·32)	0·71 (0·49–1·03)
Ischemic stroke or unknown stroke	1·20% (0·48–2·48)	2·61% (1·46–4·31)	0·47 (0·19–1·16)	0·62% (0·44–0·84)	0·90% (0·68–1·16)	0·69 (0·46–1·03)
Disabling or fatal stroke (modified Rankin Scale 3–6)	0·50% (0·10–1·47)	2·03% (1·05–3·55)	0·26 (0·07–0·93)	0·24% (0·14–0·40)	0·39% (0·25–0·57)	0·63 (0·33–1·17)
MI	0·50% (0·11–1·49)	0·83% (0·27–1·95)	0·58 (0·14–2·41)	0·53% (0·36–0·73)	0·56% (0·39–0·78)	0·93 (0·58–1·49)
Cardiovascular death	1·33% (0·58–2·63)	2·63% (1·50–4·27)	0·52 (0·22–1·22)	1·48% (1·20–1·80)	1·42% (1·15–1·75)	1·03 (0·78–1·38)
Composite of stroke, MI, SE, and total death	7·18% (5·15–9·74)	8·28% (6·08–11·01)	0·88 (0·58–1·34)	5·91% (5·33–6·53)	5·90% (5·32–6·53)	0·99 (0·86–1·15)

• Amongst the patients with a history of stroke or TIA, the annual incident rates of symptomatic intracranial hemorrhage (ICH) were 0.33% (95% CI 0.04–1.21) with apixaban vs. 0.66% (0.18–1.70) with aspirin (HR: 0.50; 95% CI 0.09 to 2.71). 
• The absolute risk difference in major bleeding events at 3.5 years was 3% (–1 to 8) in subjects with a history of stroke or TIA, as against 1% (–1 to 2) in those without a history of stroke or TIA.
• Amongst patients without a history of stroke or TIA, the annual rates of symptomatic ICH were 0.23% (0.13–0.38) in those treated with apixaban and 0.30% (0.18–0.46) in those treated with aspirin (HR 0.77, 95% CI 0.39–1.52; interaction p=0.66). 
• Absolute risk differences in symptomatic ICH at 3.5 years of follow-up were 1% (95% CI –2 to 3) in people with a history of stroke or TIA and 0% (–0 to 1) for those without a history of stroke or TIA.
• The annual incident rates of hemorrhagic stroke for people with a history of stroke or TIA were 0.17% (95% CI 0.00–0.93) with apixaban vs. 0.66% (0.18–1.70) with aspirin (HR: 0.25, 95% CI 0.03 to 2.27). 
• The annual incident rates of hemorrhagic stroke in patients without a history of stroke or TIA were 0.14% (0.06–0.26) with apixaban vs. 0.14% (0.06–0.26) with aspirin (HR: 0.97, 0.39–2.45; interaction p=0.28). 
• Absolute risk differences in hemorrhagic stroke at 3.5 years of follow-up were 2% (95% CI –0 to 4) for people with a history of stroke or TIA and 0% (–0 to 1) in those without a history.
• The incidence of serious adverse events did not differ significantly between the two study groups. 

Conclusions

• The ARTESiA trial demonstrated apixaban to be consistently safe and effective as compared to aspirin in patients with subclinical AF, irrespective of prior incidence of stroke/TIA. 
• Patients with subclinical AF and prior stroke/TIA had higher stroke/SE risk, but absolute risk reduction with apixaban outweighed the increased major bleeding risk. 
• There was no increase in fatal or intracranial bleeding with apixaban.
• The findings from the study support the role of apixaban for secondary stroke prevention in this subgroup of patients.

Lancet Neurol 2025; 24: 140–51.