The New England Journal of Medicine AUGUST 27, 2017

Rivaroxaban with or without Aspirin in Stable Cardiovascular Disease

John W. Eikelboom, Stuart J. Connolly, Jackie Bosch, et al.

Bottom Line

The COMPASS trial demonstrated that in patients with stable coronary or peripheral artery disease, the dual pathway inhibition of rivaroxaban 2.5 mg twice daily plus aspirin 100 mg once daily significantly reduced the composite risk of cardiovascular death, myocardial infarction, or stroke compared to aspirin alone, despite a higher risk of major bleeding.

Key Findings

1. The primary composite outcome of cardiovascular death, myocardial infarction, or stroke occurred in 4.1% of patients in the rivaroxaban-plus-aspirin group compared to 5.4% in the aspirin-alone group (hazard ratio [HR], 0.76; 95% confidence interval [CI], 0.66–0.86; P<0.001), indicating a significant 24% relative risk reduction.
2. Major bleeding occurred in 3.1% of patients in the rivaroxaban-plus-aspirin group compared to 1.9% in the aspirin-alone group (HR, 1.70; 95% CI, 1.40–2.05; P<0.001), reflecting a substantial increase in bleeding risk.
3. Despite the increased risk of major bleeding, there was a significant reduction in all-cause mortality with the combination therapy compared to aspirin alone (3.4% vs 4.1%; HR, 0.82; 95% CI, 0.71–0.96; P=0.01).
4. Rivaroxaban monotherapy (5 mg twice daily) did not demonstrate superior efficacy compared to aspirin alone for the primary composite outcome.

Study Design

Design
RCT
Double-Blind
Sample
27,395
Patients
Duration
23 mo
Median
Setting
Multicenter, international
Population Patients with stable coronary artery disease or peripheral artery disease.
Intervention Rivaroxaban 2.5 mg twice daily plus aspirin 100 mg once daily
Comparator Aspirin 100 mg once daily
Outcome Composite of cardiovascular death, myocardial infarction, or stroke

Study Limitations

The trial was stopped early for superior efficacy, which may result in an overestimation of the treatment effect size.
The increased incidence of major bleeding remains a critical clinical concern, potentially limiting the adoption of this therapy in patients with a high baseline risk of bleeding.
The study population was restricted to patients with stable disease, limiting generalizability to those with acute coronary syndromes or recent surgical interventions where the benefit-risk profile might differ.

Clinical Significance

The results of the COMPASS trial support the use of dual pathway inhibition (low-dose rivaroxaban plus aspirin) in patients with chronic stable atherosclerotic vascular disease to reduce major adverse cardiovascular events. It suggests that moving beyond antiplatelet monotherapy is an effective strategy for patients at high ischemic risk, provided they are not at an excessively high risk of major bleeding.

Historical Context

Prior to COMPASS, aspirin monotherapy was the standard of care for secondary prevention in patients with stable coronary or peripheral artery disease, despite a significant residual risk of ischemic events. Previous studies combining vitamin K antagonists (like warfarin) with aspirin showed increased bleeding risks without clear mortality benefits. COMPASS provided evidence that using a selective factor Xa inhibitor at a low dose in combination with aspirin could overcome the limitations of prior antithrombotic strategies.

Guided Discussion

High-yield insights from every perspective

Med Student
Medical Student

Explain the concept of 'dual pathway inhibition' (DPI) as studied in COMPASS. Why would combining a low-dose Factor Xa inhibitor with an antiplatelet agent be theoretically superior to increasing the dose of the antiplatelet alone for preventing arterial thrombosis?

Key Response

Arterial thrombi (the cause of MI and stroke) are 'white clots' primarily driven by platelet activation, but thrombin (the final product of the coagulation cascade) is a potent activator of platelets via protease-activated receptors (PARs) and stabilizes the fibrin meshwork. While high-dose aspirin focuses on the COX-1/Thromboxane A2 pathway, it does not stop thrombin-mediated activation. Adding low-dose rivaroxaban targets the coagulation cascade (Factor Xa), providing a synergistic effect that addresses two distinct pathways of thrombus formation simultaneously.

Resident
Resident

In a patient with stable coronary artery disease and concomitant peripheral artery disease (PAD), the COMPASS trial showed a clear MACE benefit. However, the risk of major bleeding was significantly higher. How should you use the trial's inclusion and exclusion criteria to identify patients with the highest 'net clinical benefit' for dual pathway inhibition?

Key Response

The trial focused on 'high-risk' stable CAD (e.g., age >65, or <65 with polyvascular disease, diabetes, or renal dysfunction). The net clinical benefit was most pronounced in patients with polyvascular disease (CAD + PAD). Conversely, patients with a history of prior intracranial hemorrhage, recent stroke, or high baseline bleeding risk were excluded. Therefore, management decisions should prioritize patients with high ischemic burden while strictly avoiding those with factors that mirror the trial's exclusion criteria to minimize the 70% relative increase in major bleeding observed.

Fellow
Fellow

The COMPASS trial demonstrated a significant reduction in Major Adverse Limb Events (MALE) in the PAD subgroup. Critically analyze the impact of the 'dual pathway' approach on the natural history of chronic limb-threatening ischemia (CLTI) compared to traditional monotherapy or DAPT.

Key Response

COMPASS showed that rivaroxaban 2.5 mg BID plus aspirin reduced MALE (including major amputation) by 46% in PAD patients. Unlike DAPT (aspirin + clopidogrel), which primarily targets platelet aggregation and has shown inconsistent results in PAD long-term, the addition of a thrombin-generation inhibitor addresses the high thrombotic milieu of diseased peripheral arteries. This suggests that for PAD, the coagulation pathway is a critical driver of local thrombotic occlusion, making DPI a superior strategy for limb salvage.

Attending
Attending

Given the COMPASS data, how do you clinically decide when to transition a patient from Dual Antiplatelet Therapy (DAPT) to Dual Pathway Inhibition (DPI) following a percutaneous coronary intervention (PCI) for stable ischemic heart disease?

Key Response

COMPASS generally enrolled patients at least one year post-PCI or those with stable disease. In the first 6–12 months post-PCI, the risk of stent thrombosis is the primary concern, requiring potent P2Y12 inhibition (DAPT). Once the stent is endothelized (typically after 1 year), the risk shifts toward 'spontaneous' atherosclerotic events in non-stented segments. At this transition point, replacing the P2Y12 inhibitor with low-dose rivaroxaban (DPI) leverages the long-term ischemic protection shown in COMPASS while moving away from the specific stent-focused protection of DAPT.

Scholarly Review

Critical appraisal through the lens of expert reviewers and guideline development

PhD
PhD

The COMPASS trial was terminated early for efficacy by the Data and Safety Monitoring Board (DSMB). Discuss the implications of 'stopping for benefit' on the potential inflation of effect sizes (the 'Winner’s Curse') and how this might impact the generalizability of the reported Number Needed to Treat (NNT).

Key Response

Trials stopped early for efficacy often capture the treatment effect at a random high point, potentially overestimating the magnitude of benefit and underestimating the risk of rare adverse events. While the p-value is robust, the NNT of 76 over 23 months might be slightly more optimistic than what would be observed had the trial reached its full planned follow-up. Researchers must account for this potential 'over-performance' when modeling the cost-effectiveness of rivaroxaban in broader healthcare systems.

Journal Editor
Journal Editor

The COMPASS trial utilized a 'run-in period' where patients were tested for tolerance and compliance before randomization. As a reviewer, how does this study design feature affect your interpretation of the reported safety outcomes and the 'real-world' applicability of the bleeding rates?

Key Response

A run-in phase excludes patients who experience early side effects or are non-compliant, effectively 'purifying' the study population. While this increases the study's internal validity and power to detect an efficacy difference, it creates a selection bias that likely underestimates the true incidence of bleeding and intolerance in the general population. A tough reviewer would flag that 'real-world' bleeding rates will likely be higher than the 3.1% reported in the trial.

Guideline Committee
Guideline Committee

The 2023 AHA/ACC Guidelines for Chronic Coronary Disease give a Class 2a recommendation for adding a second antithrombotic agent in high-risk patients. Should this evidence be upgraded to Class 1, and what specific clinical phenotypes should be highlighted to ensure the recommendation remains evidence-based?

Key Response

While COMPASS showed a mortality benefit (a rare feat in stable CAD trials), the increased risk of major bleeding prevents a universal Class 1 recommendation. To maintain rigor, guidelines should specify 'high-risk' phenotypes—specifically those with polyvascular disease (CAD + PAD) or diabetes plus a history of MI—as these groups derive the highest absolute risk reduction. The recommendation must emphasize that the benefit is contingent upon a low bleeding risk, mirroring the trial's exclusion of those with high hemorrhage potential.

Clinical Landscape

Noteworthy Related Trials

2002

WARIS II Trial

n = 3,630 · NEJM

Tested

Warfarin plus aspirin or warfarin alone

Population

Patients who had recently suffered an acute myocardial infarction

Comparator

Aspirin alone

Endpoint

Composite of death, nonfatal reinfarction, or thromboembolic stroke

Key result: The combination of warfarin and aspirin was more effective than aspirin alone in preventing major cardiovascular events after MI, though it increased bleeding risk.
2014

DAPT Study

n = 9,961 · NEJM

Tested

Dual antiplatelet therapy (thienopyridine + aspirin) for 30 months

Population

Patients who had undergone coronary stent implantation

Comparator

Aspirin monotherapy

Endpoint

Composite of death, MI, or stroke

Key result: Continued dual antiplatelet therapy beyond 1 year significantly reduced the risk of stent thrombosis and major adverse cardiovascular and cerebrovascular events.
2015

PEGASUS-TIMI 54 Trial

n = 21,162 · NEJM

Tested

Ticagrelor (60 mg or 90 mg) twice daily

Population

Patients with a prior myocardial infarction and additional risk factors

Comparator

Placebo

Endpoint

Composite of cardiovascular death, MI, or stroke

Key result: Long-term treatment with ticagrelor in patients with a history of MI significantly reduced the risk of major adverse cardiovascular events compared to placebo.

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