The New England Journal of Medicine October 05, 2017

Rivaroxaban with or without Aspirin in Stable Cardiovascular Disease (COMPASS)

John W. Eikelboom, Stuart J. Connolly, Jackie Bosch, Gilles R. Dagenais, Robert G. Hart, et al. (COMPASS Investigators)

Bottom Line

In patients with stable atherosclerotic vascular disease, the addition of low-dose rivaroxaban (2.5 mg twice daily) to low-dose aspirin significantly reduced major adverse cardiovascular events and all-cause mortality compared to aspirin alone, though it was associated with an increase in major bleeding.

Key Findings

1. The primary composite efficacy outcome (cardiovascular death, stroke, or myocardial infarction) occurred in significantly fewer patients in the rivaroxaban plus aspirin group compared to the aspirin alone group (4.1% vs. 5.4%; HR 0.76, 95% CI 0.66-0.86; P<0.001).
2. Rivaroxaban alone (5 mg twice daily) did not significantly reduce the primary composite outcome compared to aspirin alone (4.9% vs. 5.4%; HR 0.90, 95% CI 0.79-1.03; P=0.12).
3. All-cause mortality was significantly lower in the rivaroxaban plus aspirin arm compared to the aspirin alone arm (3.4% vs. 4.1%; HR 0.82, 95% CI 0.71-0.96; P=0.01).
4. Major bleeding was significantly increased in the combination group compared to aspirin alone (3.1% vs. 1.9%; HR 1.70, 95% CI 1.40-2.05; P<0.001), which was primarily driven by gastrointestinal bleeding.
5. Importantly, there was no significant increase in the rates of fatal bleeding or intracranial bleeding between the combination group and the aspirin alone group.
6. The net clinical benefit outcome (composite of cardiovascular death, stroke, MI, fatal bleeding, or symptomatic bleeding into a critical organ) significantly favored the combination group (HR 0.80, P<0.001).

Study Design

Design
RCT
Double-Blind
Sample
27,395
Patients
Duration
23 mo
Median
Setting
33 countries
Population Patients with stable atherosclerotic vascular disease (coronary artery disease, peripheral artery disease, or both)
Intervention Rivaroxaban 2.5 mg twice daily + Aspirin 100 mg daily (a secondary arm evaluated Rivaroxaban 5 mg twice daily alone)
Comparator Aspirin 100 mg daily
Outcome Composite of cardiovascular death, stroke, or myocardial infarction (MACE)

Study Limitations

The trial was stopped early (after a mean follow-up of 23 months) due to clear superiority, which may have led to a slight overestimation of the treatment effect.
The study design included an active run-in phase that excluded patients who were non-adherent or experienced early adverse events on antithrombotics, potentially selecting for a population that tolerates therapy better than a real-world cohort.
Patients with a high risk of bleeding, a need for dual antiplatelet therapy, or severe heart failure were excluded, limiting generalizability to these complex clinical populations.

Clinical Significance

The COMPASS trial established a new standard of care—dual-pathway inhibition using 'vascular-dose' rivaroxaban (2.5 mg BID) alongside low-dose aspirin—for patients with stable coronary or peripheral artery disease. By simultaneously targeting thrombin generation and platelet activation, the regimen achieved robust reductions in ischemic risk and a rare, significant mortality benefit in a stable cardiovascular population, provided patients are appropriately selected to minimize bleeding risks.

Historical Context

Prior to COMPASS, the cornerstone of antithrombotic therapy for stable atherosclerotic disease was single antiplatelet therapy, almost universally aspirin. Previous trials attempting to intensify antithrombotic therapy in stable patients, such as with dual antiplatelet therapy (e.g., CHARISMA) or full-dose anticoagulation (e.g., WARIS II), often failed to show compelling ischemic benefits or resulted in unacceptable bleeding risks. Building on the success of very-low-dose rivaroxaban in the acute coronary syndrome setting (ATLAS ACS 2-TIMI 51), COMPASS proved the dual-pathway hypothesis in chronic disease, fundamentally altering international guidelines for secondary cardiovascular prevention.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Based on the COMPASS trials use of both rivaroxaban and aspirin, how do the mechanisms of these two drugs synergize to prevent atherothrombotic events in stable cardiovascular disease?

Key Response

Aspirin inhibits platelet aggregation via COX-1, addressing primary hemostasis. Rivaroxaban directly inhibits Factor Xa in the coagulation cascade, reducing thrombin generation. Because thrombin is also a potent platelet activator via PAR-1 receptors, dual pathway inhibition targets both the platelet and fibrin components of arterial thrombosis.

Resident
Resident

The COMPASS trial demonstrated reduced MACE but increased major bleeding with dual pathway inhibition. In a general cardiology clinic, what specific patient characteristics would make you prioritize the rivaroxaban-aspirin combination over aspirin alone, and who should be excluded?

Key Response

Ideal candidates have high ischemic risk, such as multivessel CAD, polyvascular disease, or diabetes, and low bleeding risk. Patients with a history of severe bleeding, recent stroke, or those requiring therapeutic anticoagulation for atrial fibrillation should be excluded, as net clinical benefit relies on balancing ischemic reduction with hemorrhagic risk.

Fellow
Fellow

The COMPASS trial included both CAD and PAD patients. How do the results of the PAD subgroup, particularly regarding major adverse limb events, differentiate the clinical utility of rivaroxaban plus aspirin from traditional dual antiplatelet therapy in peripheral artery disease?

Key Response

In the COMPASS-PAD cohort, the combination significantly reduced major adverse limb events (MALE), including amputations. This establishes dual pathway inhibition as a foundational strategy in symptomatic PAD, contrasting with standard DAPT (like clopidogrel plus aspirin), which has traditionally lacked robust limb-salvage data in chronic stable PAD populations.

Attending
Attending

Given that the COMPASS trial was stopped early for efficacy but showed an absolute increase in major bleeding, how do you frame the net clinical benefit conversation with an asymptomatic patient who has stable CAD but is hesitant to add another medication?

Key Response

Early cessation can overestimate treatment effect, but the mortality benefit provides a hard endpoint for discussion. Attendings must teach shared decision-making by contrasting the absolute risk reduction in stroke and cardiovascular death against the absolute increase in bleeding, which was mostly gastrointestinal rather than fatal or intracranial, to align with patient values.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The COMPASS trial utilized a run-in phase and was stopped early for overwhelming efficacy. How might the early termination and the exclusion of non-adherent patients during the run-in phase bias the effect sizes for both efficacy and safety endpoints?

Key Response

An active run-in phase enriches the trial with tolerant, adherent patients, potentially underestimating real-world adverse events and dropout rates. Early stopping for efficacy (truncation bias) often captures a random high in treatment effect and might not allow enough time to capture cumulative long-term safety signals, limiting generalizability.

Journal Editor
Journal Editor

As a peer reviewer, how would you scrutinize the choice of the comparator arm in the context of contemporary medical therapy, considering many high-risk patients might otherwise receive alternative regimens?

Key Response

A rigorous reviewer would question if aspirin alone remains the correct gold standard for the high-risk subgroups enrolled, comparing it to prolonged DAPT from trials like PEGASUS-TIMI 54. Furthermore, varying baseline use of high-intensity statins or newer agents like SGLT2 inhibitors could alter baseline risk, potentially reducing the absolute benefit of dual pathway inhibition today.

Guideline Committee
Guideline Committee

Based on the COMPASS trial, how should current guidelines be updated regarding antithrombotic therapy for chronic coronary syndromes, and what Level of Evidence should be assigned to the rivaroxaban 2.5 mg BID plus aspirin regimen?

Key Response

ESC guidelines have incorporated this as a Class IIa (Level of Evidence A) recommendation for patients with chronic coronary syndromes at high ischemic risk and low bleeding risk. The committee must weigh the strong RCT evidence against the increased bleeding risk to recommend it specifically for patients with diffuse CAD or polyvascular disease, distinguishing it from standard single antiplatelet therapy.

Clinical Landscape

Noteworthy Related Trials

2012

ATLAS ACS 2-TIMI 51

n = 15,526 · NEJM

Tested

Rivaroxaban 2.5 mg or 5 mg twice daily

Population

Patients with recent acute coronary syndrome

Comparator

Placebo

Endpoint

Composite of cardiovascular death, MI, or stroke

Key result: Rivaroxaban significantly reduced the composite endpoint of cardiovascular death, myocardial infarction, or stroke compared to placebo, but increased major bleeding.
2015

PEGASUS-TIMI 54

n = 21,162 · NEJM

Tested

Ticagrelor 60 mg or 90 mg twice daily

Population

Patients with a history of MI 1 to 3 years prior

Comparator

Placebo

Endpoint

Composite of cardiovascular death, MI, or stroke

Key result: Adding ticagrelor to aspirin significantly reduced the risk of major cardiovascular events but increased the risk of major bleeding.
2020

VOYAGER PAD

n = 6,564 · NEJM

Tested

Rivaroxaban 2.5 mg twice daily

Population

Patients with symptomatic PAD undergoing lower-extremity revascularization

Comparator

Placebo

Endpoint

Composite of acute limb ischemia, major amputation, MI, ischemic stroke, or CV death

Key result: Rivaroxaban significantly lowered the incidence of the composite primary outcome compared to placebo, with a higher rate of major bleeding.

Tailored to your role

Want this tailored to you?

Add your specialty or training stage to get role-specific takeaways and more questions.

Personalize this analysis