OCEAN
(2025)Objective
To determine whether continuing rivaroxaban is superior to aspirin for preventing stroke, systemic embolism, and covert embolic stroke in patients with stroke risk factors who had successful catheter ablation for atrial fibrillation at least 1 year prior.
Study Summary
• Annualized stroke/embolism rates were very low in both groups: 0.31 vs 0.66 events per 100 patient-years
• Major bleeding similar between groups (1.6% vs 0.6%), but clinically relevant nonmajor bleeding higher with rivaroxaban (5.5% vs 1.6%)
Intervention
Rivaroxaban 15mg daily vs aspirin 70-120mg daily for 3 years in patients with successful AF ablation ≥1 year prior and CHA2DS2-VASc score ≥1 (or ≥2 for women/vascular disease patients). Success defined as no atrial arrhythmia >30 seconds on Holter monitoring at 2-6 months and >6 months post-ablation.
Inclusion Criteria
Successful catheter ablation for nonvalvular atrial fibrillation ≥1 year prior, CHA2DS2-VASc score ≥1 (≥2 for women or vascular disease patients), no atrial arrhythmia >30 seconds on serial Holter monitors
Study Design
Arms: Rivaroxaban 15mg daily (n=641) vs aspirin 70-120mg daily (n=643)
Patients per Arm: 641 rivaroxaban, 643 aspirin
Outcome
• All stroke: 0.8% vs 1.1%, RR 0.72 (0.23-2.25)
• Major bleeding: 1.6% vs 0.6%, HR 2.51 (0.79-7.95)
Bottom Line
In patients with successful AF ablation ≥1 year prior and stroke risk factors, rivaroxaban did not significantly reduce the composite of stroke, systemic embolism, or covert embolic stroke compared to aspirin, with very low event rates in both groups.
Major Points
- No significant difference in primary composite outcome between rivaroxaban and aspirin (0.8% vs 1.4%, p=0.28)
- Extremely low annualized stroke rates in both groups (0.31 vs 0.66 events per 100 patient-years)
- 96% of patients in both groups had no evidence of new cerebral infarcts on 3-year MRI
- Similar major bleeding rates but higher clinically relevant nonmajor bleeding with rivaroxaban
- Trial stopped early for futility due to low event rates
Study Design
- Study Type
- International, open-label, randomized, blinded-outcome-assessment trial
- Randomization
- Yes
- Blinding
- Open-label with blinded outcome assessment
- Sample Size
- 1284
- Follow-up
- 3 years (median 36.0 months)
- Centers
- 56
- Countries
- Canada, US, Germany, Belgium, Australia, China
Primary Outcome
Definition: Composite of stroke, systemic embolism, or new covert embolic stroke (≥1 new infarct measuring ≥15mm on MRI) at 3 years
| Control | Intervention | HR/OR | P-value |
|---|---|---|---|
| 9 (1.4%) | 5 (0.8%) | - (0.19 to 1.65) | 0.28 |
Limitations & Criticisms
- A placebo could have been chosen over aspirin as a comparator
- The trial did not mandate extended monitoring of atrial fibrillation before enrollment or during follow-up, so precise incidence of asymptomatic AF recurrence is unknown
- Trial included patients with moderate stroke risk and minimal cardiac disease, so findings may not apply to very high stroke risk patients
- Event rates were much lower than anticipated, resulting in the trial being underpowered
- The 15mg rivaroxaban dose may have been suboptimal compared to the standard 20mg dose
Citation
N Engl J Med. 2025;DOI: 10.1056/NEJMoa2509688