Highlight
– In patients with ischemic stroke or transient ischemic attack (TIA) who have nonvalvular atrial fibrillation (AF) and concomitant atherosclerotic cardiovascular disease, adding an antiplatelet to anticoagulant therapy provided no net clinical benefit over anticoagulant monotherapy at 2 years.
– The combination strategy produced a numerically smaller (but statistically nonsignificant) reduction in ischemic events (11.1% vs 14.2%) but caused a marked increase in major or clinically relevant nonmajor bleeding (19.5% vs 8.6%; HR 2.42, 95% CI 1.23–4.76, P = .008).
– Absolute effects translate to roughly one additional major/clinically relevant bleed for every 9 patients treated with combination therapy (NNH ≈ 9) versus an uncertain prevention of ischemic events (NNT ≈ 32, not statistically significant).
Background
Patients with ischemic stroke or TIA and concurrent nonvalvular atrial fibrillation are at high risk of recurrent ischemic stroke from cardiac embolism. At the same time, a substantial proportion have coexisting atherosclerotic cardiovascular disease — carotid or intracranial stenosis, prior noncardioembolic stroke, ischemic heart disease, or peripheral artery disease — that may be expected to respond to antiplatelet therapy. Clinicians commonly face the dilemma whether combining antiplatelet therapy with oral anticoagulation yields incremental ischemic protection that outweighs bleeding harm.
Guideline documents generally recommend oral anticoagulation for stroke prevention in AF and reserve addition of antiplatelet therapy to situations with specific indications (for example, recent acute coronary syndrome or percutaneous coronary intervention). However, high absolute ischemic risk from coexisting atherosclerotic disease has prompted uncertainty and heterogeneity in practice. The ATIS-NVAF randomized trial was designed to address the net clinical benefit of combination antiplatelet plus anticoagulant therapy versus anticoagulant monotherapy in this high-risk population.
Study design
ATIS-NVAF was a multicenter, open-label randomized clinical trial conducted at 41 sites in Japan (November 2016 to March 2025). The trial enrolled adults with ischemic stroke or TIA 8 to 360 days before randomization, documented nonvalvular AF, and at least one manifestation of atherosclerotic cardiovascular disease (carotid or intracranial artery stenosis, noncardioembolic stroke, ischemic heart disease, or peripheral artery disease). Participants were randomized to either combination therapy (oral anticoagulant plus antiplatelet agent) or anticoagulant monotherapy. The anticoagulant regimen (vitamin K antagonist or direct oral anticoagulant) and the specific antiplatelet agent were chosen by treating clinicians per standard practice; details of regimens are reported in the primary manuscript.
The primary outcome was a composite of ischemic cardiovascular events and major bleeding within 2 years. Secondary and safety outcomes included ischemic cardiovascular events separately, and major and clinically relevant nonmajor bleeding events. The trial was open-label and planned to enroll a larger sample, but an interim analysis led to early termination for futility on July 18, 2023. Data were analyzed through October 2024.
Key findings
316 patients were randomized (159 to combination therapy; 157 to monotherapy). The cohort was elderly (mean age 77.2 years) with 28.5% female participants. Key outcomes at 2 years were:
- Primary composite outcome (ischemic events plus major bleeding): 17.8% in the combination group vs 19.6% in the monotherapy group (hazard ratio [HR] 0.91; 95% CI 0.53–1.55; P = .64). No statistically significant difference.
- Ischemic cardiovascular events alone: 11.1% vs 14.2% (HR 0.76; 95% CI 0.39–1.48; P = .41). Numerically fewer ischemic events with combination therapy but not statistically significant.
- Major and clinically relevant nonmajor bleeding: 19.5% vs 8.6% (HR 2.42; 95% CI 1.23–4.76; P = .008). Combination therapy more than doubled the risk of clinically important bleeding.
Absolute event differences illustrate the clinical trade-offs: the absolute increase in clinically important bleeding with combination therapy was 10.9% (19.5% minus 8.6%), yielding a number needed to harm (NNH) of approximately 9. The absolute reduction in ischemic events was 3.1% (14.2% minus 11.1%), corresponding to a number needed to treat (NNT) of about 32 to prevent one ischemic event — a benefit that did not reach statistical significance and must be interpreted cautiously given early termination and limited power.
Interpretation and clinical implications
The ATIS-NVAF trial provides randomized evidence that routine addition of an antiplatelet agent to anticoagulation in patients with ischemic stroke/TIA, AF, and atherosclerotic disease does not confer a net clinical advantage and substantially increases clinically important bleeding. The findings are consistent with prior randomized data and guideline recommendations that oral anticoagulation is the mainstay for preventing cardioembolic stroke in AF and that antiplatelet therapy should not be added routinely unless there is a separate, compelling atherosclerotic indication (for example, recent coronary stenting or acute coronary syndrome where short-term dual therapy is indicated).
Two practical points follow. First, clinicians should carefully review the indication for any antiplatelet therapy in a patient already on anticoagulation and avoid long-term combination therapy without clear benefit. Second, when combination therapy is required (for example, after percutaneous coronary intervention), the duration should be minimized and bleeding risk mitigation strategies (radial access, proton pump inhibitor for GI protection, use of DOAC rather than warfarin when appropriate) should be employed.
Strengths and limitations
Strengths of ATIS-NVAF include randomized allocation, a clinically relevant composite primary outcome that incorporated both ischemic and bleeding events to assess net clinical benefit, and enrollment of an older, comorbid population commonly encountered in practice.
Important limitations temper the findings. The trial was open-label, which could affect ascertainment and management of some outcomes, although objective endpoints like stroke and major bleeding are less susceptible to observer bias. The study was terminated early for futility, resulting in a smaller-than-planned sample size and reduced statistical power to detect modest differences in ischemic events. The specific composition and dosing of antiplatelet agents and the mix of anticoagulants (warfarin vs DOACs) were determined by treating clinicians and may vary from other settings; detailed subgroup outcome data are necessary to explore heterogeneity (for instance, patients with prior coronary stents vs carotid stenosis only). Finally, the trial population was Japanese and elderly (mean age 77); generalizability to younger populations or different ethnic groups should be considered cautiously.
How this fits with prior evidence
The trial’s risk–benefit trade-off echoes findings from other cardiovascular trials where combination antithrombotic strategies reduced ischemic events at the cost of higher bleeding (for example, COMPASS demonstrated that low-dose rivaroxaban plus aspirin reduced major adverse cardiovascular events in stable atherosclerotic disease but increased major bleeding). However, the AF population differs because anticoagulation alone provides robust prevention of cardioembolic stroke, so the incremental benefit of adding antiplatelet therapy is smaller and more likely to be offset by bleeding harm. AF management guidelines generally recommend anticoagulation as first-line stroke prevention and reserve antiplatelet agents for specific concomitant indications; ATIS-NVAF provides randomized trial evidence supporting this approach in patients with prior ischemic stroke/TIA and atherosclerosis.
Expert commentary and practice recommendations
For practicing clinicians, the trial supports the following pragmatic approach:
– Prioritize anticoagulant monotherapy for stroke prevention in AF unless a clear separate indication for antiplatelet therapy exists (e.g., recent coronary stenting or acute coronary syndrome).
– If short-term combination therapy is required, minimize duration, choose lower-risk strategies (use of DOACs when appropriate, employ gastroprotection), and reassess the ongoing need for antiplatelet therapy at follow-up.
– Use individualized bleeding-risk assessment (eg, HAS-BLED) and shared decision-making when discussing combination therapy risks and benefits with patients and caregivers.
From a research standpoint, adequately powered randomized trials or pooled patient-level meta-analyses are needed to address whether particular subgroups (for example, patients with high-burden intracranial atherosclerosis or recent coronary stent placement) might derive net benefit from carefully selected combination strategies and to determine optimal durations and drug choices.
Conclusion
In patients with ischemic stroke or TIA, nonvalvular AF, and concomitant atherosclerotic cardiovascular disease, adding an antiplatelet agent to anticoagulation did not reduce the composite of ischemic events and major bleeding and resulted in a substantially higher risk of clinically important bleeding. These data support anticoagulant monotherapy as the default strategy in this population and caution against routine long-term combination antithrombotic therapy except when specific atherosclerotic indications exist.
Funding and clinicaltrials.gov
Trial registration: ClinicalTrials.gov Identifier: NCT03062319.
Funding sources and detailed disclosures are reported by the authors in the primary publication (Okazaki et al., JAMA Neurol. 2025).
References
1. Okazaki S, Tanaka K, Yazawa Y, et al; ATIS-NVAF Trial Investigators. Optimal Antithrombotics for Ischemic Stroke and Concurrent Atrial Fibrillation and Atherosclerosis: A Randomized Clinical Trial. JAMA Neurol. 2025 Oct 6:e253662. doi: 10.1001/jamaneurol.2025.3662. PMID: 41051787; PMCID: PMC12501853.
2. Mant J, Hobbs FD, Fletcher K, et al. Warfarin versus aspirin for stroke prevention in an elderly community population with atrial fibrillation (BAFTA): randomized controlled trial. Lancet. 2007;370(9586):493–503.
3. Eikelboom JW, Connolly SJ, Bosch J, et al.; COMPASS Investigators. Rivaroxaban with or without aspirin in stable cardiovascular disease. N Engl J Med. 2017;377(14):1319–1330.
4. Kirchhof P, Benussi S, Kotecha D, et al. 2020 ESC Guidelines for the diagnosis and management of atrial fibrillation developed in collaboration with the European Association for Cardio-Thoracic Surgery (EACTS). Eur Heart J. 2020;42(5):373–498. (Guideline summary and management recommendations relevant to antithrombotic strategies in AF.)
