People with subclinical atrial fibrillation are at increased risk of stroke, albeit to a lesser extent than those with clinical atrial fibrillation, leading to an ongoing debate regarding the benefit of anticoagulation in these individuals. In the ARTESiA trial, the direct-acting oral anticoagulant apixaban reduced stroke or systemic embolism compared with aspirin in people with subclinical atrial fibrillation, but the risk of major bleeding was increased with apixaban. In a prespecified subgroup analysis of ARTESiA, we tested the hypothesis that people with subclinical atrial fibrillation and a history of stroke or transient ischaemic attack, who are known to have an increased risk of recurrent stroke, would show a greater benefit from oral anticoagulation for secondary stroke prevention compared with those without a history of stroke or transient ischaemic attack.ARTESiA is a double-blind, double-dummy, randomised controlled trial conducted at 247 sites in 16 countries across Europe and North America. Adults aged 55 years or older with device-detected subclinical atrial fibrillation lasting from 6 min to 24 h and a CHA2DS2-VASc score of 3 or higher were randomly assigned using an interactive web-based system to oral apixaban 5 mg twice per day or oral aspirin 81 mg once per day. The primary efficacy outcome was stroke or systemic embolism, and the primary safety outcome was major bleeding, assessed as absolute risk differences. Analyses were by intention to treat. ARTESiA is registered with ClinicalTrials.gov (NCT01938248) and is completed; this report presents a prespecified subgroup analysis in people with a history of stroke or transient ischaemic attack.Between May 7, 2015, and July 30, 2021, 4012 people with subclinical atrial fibrillation were randomly allocated either apixaban (n=2015) or aspirin (n=1997). A history of stroke or transient ischaemic attack was present in 346 (8·6%) participants (172 assigned to apixaban and 174 to aspirin), among whom the annual rate of stroke or systemic embolism was 1·20% (n=7; 95% CI 0·48 to 2·48) with apixaban versus 3·14% (n=18; 1·86 to 4·96) with aspirin; (hazard ratio [HR] 0·40, 95% CI 0·17 to 0·95). In participants without a history of stroke or transient ischaemic attack (n=3666; 1843 assigned to apixaban and 1823 to aspirin), the annual rate of stroke or systemic embolism was 0·74% (n=48; 95% CI 0·55 to 0·98) with apixaban versus 1·07% (n=68; 95% CI 0·83 to 1·36) with aspirin (HR 0·69, 95% CI 0·48 to 1·00). The absolute risk difference in incidence of stroke or systemic embolism at 3·5 years of follow-up was 7% (95% CI 2 to 12) in participants with versus 1% (0 to 3) in participants without a history of stroke or transient ischaemic attack. The annual rate of major bleeding in participants with a history of stroke or transient ischaemic attack was 2·26% with apixaban (n=13; 95% CI 1·21 to 3·87) versus 1·16% with aspirin (n=7; 0·47 to 2·39; HR 1·94, 95% CI 0·77 to 4·87). The absolute risk difference in major bleeding events at 3·5 years was 3% (-1 to 8) in individuals with a versus 1% (-1 to 2) in those without a history of stroke or transient ischaemic attack.Treatment with the direct-acting oral anticoagulant apixaban in people with subclinical atrial fibrillation and a history of stroke or transient ischaemic attack led to a 7% absolute risk reduction in stroke or systemic embolism over 3·5 years, compared with a 1% absolute risk reduction for individuals without a previous history of stroke or transient ischaemic attack. The corresponding absolute increase in major bleeding was 3% and 1%, respectively. Apixaban could be considered for secondary stroke prevention in people with subclinical atrial fibrillation and a history of stroke or transient ischaemic attack.The Canadian Institutes of Health Research, Bristol-Myers Squibb-Pfizer Alliance, Heart and Stroke Foundation of Canada, Canadian Stroke Prevention and Intervention Network, Hamilton Health Sciences, Accelerating Clinical Trials Network, Population Health Research Institute, and Medtronic.