Subscribe to eTOC

Patients at High Risk for Stroke Benefit from Combination Rivaroxaban Plus Aspirin in International Trial

Article In Brief


Distribution of modified Rankin Scale scores of stroke (A) and ischemic stroke or uncertain stroke (B) with the different treatment options.

Rivaroxaban 2.5 mg twice daily with aspirin 100 mg prevented stroke better than aspirin alone in patients with coronary artery or periphery artery disease and no recent events.

A combination of the anticoagulant rivaroxaban and aspirin in people with stable coronary artery and peripheral artery disease, when compared with aspirin alone, led to the greatest benefit in those at greatest risk for stroke, according to a newly released, deeper probe of the data in the Cardiovascular Outcomes for People Using Anticoagulation Strategies (COMPASS) trial.

The new analysis, published in the February 26 online edition of Circulation, shows how the combination of 2.5 mg of rivaroxaban twice a day plus aspirin brings benefit without being canceled out by safety concerns, said Michael Sharma, MD, MSc, associate professor of neurology at McMaster University, and the lead author on the analysis.

The current report provides a more in-depth look at stroke outcomes in 33 countries and totaling more than 27,000 patients; it focused on the combined outcome of vascular death from different stroke subtypes and is an attempt to identify those at greatest risk for stroke, Dr. Sharma said.

“We felt that if we could help people do that, it would make it easier to target this therapy to patients most likely to benefit,” he said.

Study Design

Peripheral artery disease in COMPASS included carotid stenosis of more than 50 percent or previous carotid revascularization. Among those who had experienced a previous stroke and were taking only aspirin, the annual rate of stroke was 3.4 percent. But among those taking rivaroxaban with aspirin, the rate was 0.7 percent per year (p=.03).

This analysis shows how the benefits of the combination strategy were not offset by increases in intracerebral hemorrhage or hemorrhagic transformation. This five-fold increase in the aspirin alone group was evident despite good blood pressure control and use of statins. This finding was seen as particularly significant because a previous stroke was the strongest predictor of stroke in the trial, researchers found.

Among those who hadn't had a previous stroke, the benefit from the rivaroxaban-aspirin combination was greatest among the patients deemed at high-risk for ischemic stroke, who were identified based on age, blood pressure, hypertension, diabetes, and Asian ethnicity. Those in this no-prior-stroke, high-risk group who were taking aspirin alone had an annual stroke rate of 1.3 percent, compared to 0.6 percent for those taking the combination therapy (p=.001).

The new analysis was as notable for the effects the combination treatment had as it was for the effects it didn't have. Researchers didn't observe any significant increase in intracerebral hemorrhage among those on the combination treatment—even though it was slightly elevated in these patients—so the reduction in ischemic stroke risk was not offset by this complication, Dr. Sharma said.

Even more intriguing—and surprising—was that researchers actually observed a decrease in hemorrhagic transformation—or ischemia-related brain hemorrhage—among patients on the combination therapy, at an annual rate of 0.03 percent, compared with those on aspirin alone, at an annual rate of 0.08 percent.

“Generally, as a stroke neurologist, I would expect that an individual who has more intense antithrombotic therapy and experiences an ischemic stroke is more likely to have hemorrhagic transformation,” Dr. Sharma said. “In this analysis, we saw exactly the reverse. Patients who were on rivaroxaban and aspirin were less likely to have hemorrhagic transformation than those on aspirin.”

This finding, Dr. Sharma said, is “an enormous relief for us.”

“We're always treading the line between reducing ischemia and increasing hemorrhage with antithrombotic therapies, and that line becomes thinner and thinner in patients who are at highest risk for ischemic stroke,” he said. Here, investigators have found a way to have a significant effect of ischemia without significant hemorrhage risks.

One possible explanation is that the size of the stroke was smaller among those taking the combination, thereby reducing the risk of hemorrhagic transformation, Dr. Sharma said. But imaging wasn't collected in the trial, so investigators can't be sure of the reason, he said.

The combination not only reduced the risk of stroke, but also the risk of death. The annual rate of death within 30 days of stroke was 0.7 percent in the aspirin alone group and 0.6 percent in the rivaroxaban and aspirin group.

Usage of the rivaroxaban-aspirin combination is increasing, Dr. Sharma said, but it depends on insurance coverage and on the physician who is seeing a given patient.

“The population that will benefit...encounters different types of physicians,” he said. “I certainly see them, and stroke physicians and stroke neurologists will see patients like this. But they're also seen by cardiologists, primary care physicians, vascular surgeons—and part of the challenge has really been to describe to each group of practitioners what these patients look like in their population and what the specific benefit is for that group.”

People with a higher risk of bleeding were excluded from the study, so this population should be regarded with caution when considering the rivaroxaban-aspirin combination. But on an encouraging note, in an MRI sub-study, patients found to have lacunar infarct—despite this being an exclusion criterion—did not have an overall increase in intracerebral hemorrhage.

“I don't think we need to worry about lacunar infarct” in terms of the combination treatment and ICH, Dr. Sharma said.

He said he is hopeful for future research. “I think this opens up a whole new avenue of research in combining these two groups of anti-thrombotics,” he said.

He also said the COMPASS findings underscore the importance of phase 2 dose-finding trials: It's the only way the 2.5 mg twice daily dose was identified as the sweet spot for benefit while minimizing risk. A 5 mg twice-daily dose brought a significant increase in ICH, he noted.

“I think for us in the future in stroke, we're going to need to go back to phase 2 data and work very hard to establish a dose that's appropriate for stroke patients,” he said. “I tell you, for me, it's completely changed the way I'm thinking about our next trial.”

Expert Commentary

Hooman Kamel, MD, director of the clinical and translational neuroscience unit at the Brain and Mind Research Institute at Cornell University, called it “a rigorous trial with very interesting results” but said it was important to keep in mind that it enrolled a specific type of patient.

“For neurologists who typically focus on secondary stroke prevention, it may be challenging to interpret the trial and decide how its results should shape clinical care,” he said. “It involved a rather specific population since it was limited to patients with coronary artery or peripheral arterial disease, which applies to about one-quarter of stroke patients. Patients with atrial fibrillation were excluded since we already know they should be anticoagulated, and those with lacunar stroke were also excluded.”


“We're always treading the line between reducing ischemia and increasing hemorrhage with antithrombotic therapies, and that line becomes thinner and thinner in patients who are at highest risk for ischemic stroke.”


The finding that rivaroxaban led to fewer ischemic strokes than aspirin is at odds with several secondary stroke prevention trials, including NAVIGATE ESUS, published last year.

“Given this context, it remains to be seen whether anticoagulation will find a place for secondary prevention in stroke patients without atrial fibrillation,” he said.

A key question to be explored is the mechanism of the therapeutic effect, he said.

“Is it prevention of cardiac embolism or artery-to-artery embolism, or both?” It also remains to be seen if these results can be reconciled with other recent trials of novel oral anticoagulants.

Pooja Khatri, MD, MSc, professor of neurology at the University of Cincinnati, said she expects these findings to lead to more of a shift to using low-dose rivaroxaban with aspirin.

“Perhaps we finally have an approach that is less of that double-edged sword, in terms of bleeding and ischemia, although I would still like to see this confirmed in a second study to be certain,” she said.


Dr. Sharma reported grants and personal fees from Bayer, Bristol-Myers Squibb, Daiichi Sankyo, and Portola during the conduct of the study. Drs. Hooman and Khatri reported no competing interests.

Link Up for More Information

•. Sharma M, Hart RG, Connolly SJ, et al. Stroke outcomes in the COMPASS Trial Circulation 2019; Epub 2019 Feb 26.