Study Summary

The COMPASS trial1 demonstrated that low-dose rivaroxaban with aspirin is associated with fewer adverse cardiovascular events, but more bleeding events, than aspirin alone in patients with stable atherosclerosis. Patients were randomly assigned to rivaroxaban 2.5 mg twice daily plus aspirin (n = 9,152), rivaroxaban 5 mg twice daily (n = 9,117), or aspirin (n = 9,126). Stable atherosclerosis was defined as atherosclerosis either in at least two vascular beds or with at least two additional cardiovascular risk factors. Exclusion criteria included recent stroke, advanced heart failure, advanced kidney disease, high risk of bleeding, or use of dual antiplatelet or anticoagulation therapies. Patients were followed up (mean duration 23 months) for the primary outcome of incidence of cardiovascular death, myocardial infarction, or stroke.

The primary outcome occurred in 4.1% of patients who received rivaroxaban with aspirin compared with 5.4% of those in the aspirin alone group (P <0.001). Rivaroxaban with aspirin also outperformed aspirin alone for the secondary outcomes of all-cause mortality (P = 0.01), all stroke (P <0.001), and ischemic stroke (P <0.001). However, major bleeding — primarily gastrointestinal bleeding — occurred more frequently in the rivaroxaban with aspirin group (3.1%) than in both the rivaroxaban alone group (2.8%) and the aspirin alone group (1.9%) (P <0.001 for all comparisons). Rivaroxaban 5 mg twice daily alone did not result in better cardiovascular outcomes than aspirin alone and resulted in more bleeding events.

The investigators also analyzed the lifetime benefit and risk of adding low-dose rivaroxaban to aspirin2 using patients from the COMPASS and SMART3 cohorts. Rivaroxaban with aspirin provided a predicted median gain of 16 months of life expectancy free from stroke or myocardial infarction, as well as a predicted median loss of 2 months of life expectancy from major bleeding. Lifetime benefit and risk were highly variable, depending on the individual. The models developed by the investigators (available online4) can be used by clinicians to predict risk and benefit on the basis of individual patient characteristics.


Study Strengths: This analysis benefitted from a large study population with diverse background characteristics. Additionally, the model incorporated competing risks (e.g. non-CVD mortality), thereby preventing overestimation of risk and treatment effects seen in traditional models.

Study Limitations: Patients were enrolled in the SMART cohort from 1996, when risk profiles and standards of care were different. The combination of the SMART and COMPASS cohorts, which have considerably different baseline risk profiles and event rates, may result in a cohort that is not representative of typical patients5. The c-indexes of the lifetime scores were modest (0.62–0.71), but comparable to other risk models for patients with known CVD. The models should be validated with prospective testing in a real-world cohort.

Next Steps/Clinical perspective: The COMPASS trial demonstrated a reduction in adverse cardiovascular events from adding low-dose rivaroxaban to aspirin in patients with stable atherosclerosis, but at the cost of increased major bleeding. This subsequent analysis enables clinicians to identify patients most likely to benefit from, and least likely to be harmed by, the addition of rivaroxaban. Clinicians can use the models to identify patients requiring close monitoring for bleeding after initiation of rivaroxaban. The use of a lifetime, rather than a 10-year, prediction model could shift the focus to younger patients with the largest potential gains. As patients differ in what they consider a meaningful benefit or risk, clinicians should use the models in the context of shared decision-making.


1. +
Eikelboom,  J. W. et al. Rivaroxaban with or without aspirin in stable cardiovascular disease. N. Engl. J. Med. 377: 1319–1330 (2017).
[PubMed: 28844192]
2. +
De Vries,  T. I. et al. Estimating individual lifetime benefit and bleeding risk of adding rivaroxaban to aspirin for patients with stable cardiovascular disease: results from the COMPASS trial. Eur. Heart J. doi:10.1093/eurheartj/ehz404.
3. +
Simons,  P.C.G. et al. Second manifestations of ARTerial disease (SMART) study: rationale and design. Eur. J. Epidemiol. 15: 773–781 (1999).
[PubMed: 10608355]
4. +
UMC Utrectht (2018). U-Prevent calculators. (Accessed November 23, 2019).
5. +
Koziel,  M., & Lip,  G.Y. Estimating individual lifetime benefit and bleeding risk of adding oral anticoagulation to aspirin for patients with stable cardiovascular disease: directions from COMPASS? Eur. Heart J. doi: 10.1093/eurheartj/ehz517
[PubMed: 31671800]