Background
Phase 3, double-blind, RCT (EINSTEIN CHOICE). 3,396 patients with symptomatic DVT or PE who had completed 6–12 months of anticoagulation therapy and were in clinical equipoise regarding the need for continued anticoagulation. Three-arm design: rivaroxaban 20 mg once daily (treatment dose), rivaroxaban 10 mg once daily (prophylactic dose), or aspirin 100 mg once daily for up to 12 months. This trial directly compared DOAC-based extended anticoagulation against aspirin — a commonly used but less potent alternative — testing whether a reduced DOAC dose provides superior VTE prevention while maintaining a favorable bleeding profile compared to aspirin.
Interventions and follow up
Arm A: Rivaroxaban 20 mg once daily (treatment dose) × up to 12 month
Arm B: Rivaroxaban 10 mg once daily (prophylactic dose) × up to 12 month
Arm C: Aspirin 100 mg once daily × up to 12 month
Primary endpoint: Symptomatic recurrent fatal or nonfatal VTE
mFollow up: Median 351 day
Arm B: Rivaroxaban 10 mg once daily (prophylactic dose) × up to 12 month
Arm C: Aspirin 100 mg once daily × up to 12 month
Primary endpoint: Symptomatic recurrent fatal or nonfatal VTE
mFollow up: Median 351 day
Results
Recurrent VTE (rivaroxaban 20 mg vs aspirin): 1.5% vs 4.4%, HR 0.34 (95% CI 0.20–0.59), P<.001
Recurrent VTE (rivaroxaban 10 mg vs aspirin): 1.2% vs 4.4%, HR 0.26 (95% CI 0.14–0.47), P<.001
Major bleeding (20 mg vs 10 mg vs aspirin): 0.5% vs 0.4% vs 0.3% — no significant differences
Clinically relevant nonmajor bleeding: 2.7% vs 2.0% vs 1.8%
Recurrent VTE (rivaroxaban 10 mg vs aspirin): 1.2% vs 4.4%, HR 0.26 (95% CI 0.14–0.47), P<.001
Major bleeding (20 mg vs 10 mg vs aspirin): 0.5% vs 0.4% vs 0.3% — no significant differences
Clinically relevant nonmajor bleeding: 2.7% vs 2.0% vs 1.8%
Adverse events
Main adverse events: Major bleeding was low and not significantly different across all three arms — the critical safety finding supporting rivaroxaban 10 mg. Intracranial hemorrhage: rare (<0.1%) in all arms. GI bleeding was slightly numerically higher with rivaroxaban 20 mg vs 10 mg. All-cause adverse events were similar across groups. No unexpected safety signals.
Conclusions
Both rivaroxaban doses (20 mg and 10 mg once daily) were significantly superior to aspirin in preventing recurrent VTE after initial anticoagulation, with comparable major bleeding rates, establishing rivaroxaban 10 mg as an effective, well-tolerated extended-treatment option with lower bleeding potential than the full treatment dose.
Key Limitations
Key Limitations: No placebo arm — cannot determine the absolute benefit of anticoagulation vs no treatment in this population (see WARFASA/ASPIRE for aspirin vs placebo data). No direct comparison to rivaroxaban 20 mg vs 10 mg for the primary endpoint — both were superior to aspirin, but the powered head-to-head was not pre-specified. Whether aspirin is an adequate alternative to no anticoagulation (as an intermediate option) in patients with high bleeding risk is not answered by this design. No comparison to warfarin or apixaban extended-phase therapy. Subgroup analysis by thrombophilia status not pre-specified. Median follow-up of 351 days does not address indefinite anticoagulation duration decisions.
Clinical Context
EINSTEIN CHOICE established rivaroxaban 10 mg once daily as the extended-treatment dose for VTE, FDA-approved for this indication in 2019. Together with apixaban 2.5 mg BID (AMPLIFY-EXT), reduced-dose DOAC regimens are now the standard approach for extended-phase VTE treatment in patients with moderate recurrence risk. The rivaroxaban 10 mg once daily is simpler dosing than apixaban 2.5 mg BID. Aspirin is not recommended over anticoagulation in patients who can tolerate extended anticoagulation (ACCP guidelines). For patients with active cancer, anticoagulation at treatment dose is maintained (do not reduce to prophylactic dose).
References