Rivaroxaban in Addition to DAPT Lowers Death, MI, Stroke in ACS, Heart Failure Patients: Meta-analysis

Combined data from ATLAS ACS-2 and COMMANDER-HF suggest that another conversation about bleeding needs to occur, the study presenter says.

Rivaroxaban in Addition to DAPT Lowers Death, MI, Stroke in ACS, Heart Failure Patients: Meta-analysis

WASHINGTON, DC—Among post-ACS or heart failure patients on dual antiplatelet therapy (DAPT), the addition of rivaroxaban can reduce both the combination of death, MI, and stroke as well as all-cause mortality alone, according to a new meta-analysis. Additionally, this finding was maintained among ACS patients regardless of when the direct oral anticoagulant was started.

Notably, the US Food and Drug Administration has not approved rivaroxaban (Xarelto; Janssen Pharmaceuticals) for the indication of ACS, although it is currently endorsed for both CAD and PAD. However, C. Michael Gibson, MD (Beth Israel Deaconess Hospital, Boston, MA), who presented this data here today in a late-breaking clinical trial session at CRT 2019, said his team has had “active discussions” with the agency, which has reviewed these findings.

For the study, Gibson and colleagues combined data from 4,825 and 907 patients, respectively, from the ATLAS ACS-2 TIMI-51 and COMMANDER-HF trials who received the study drug at a twice-daily dose of 2.5 mg and compared them against those who received placebo. Ultimately, rivaroxaban reduced all-cause death and the composite endpoint of death, MI, and stroke compared with placebo, and there was a numerical decrease in MI but no difference in stroke.

Primary Efficacy Endpoints: Rivaroxaban vs Placebo 



95% CI

Death, MI, and Stroke



All-Cause Death










With regard to safety, there were no differences in fatal or critical organ bleeding, fatal bleeding, or intracranial bleeding between the rivaroxaban and placebo cohorts, but there was an increase in major bleeding with rivaroxaban (2.7 vs 1.52 events per 100 patient years; P < 0.01).

Results were maintained among the population of patients only with MI or unstable angina. In a preliminary exploratory analysis also including data from the COMPASS trial, rivaroxaban was effective in this cohort regardless of whether it was started within 1 month (HR 0.82), between 1 and 12 months (HR 0.77), or after 12 months (HR 0.74) after the MI or unstable angina event (for interaction = 0.68).

Bleeding Questions

In a press conference, Ron Waksman, MD (MedStar Washington Hospital, Washington, DC), commented that bleeding itself is associated with an increase in mortality. “How do you reconcile this [increase in] major bleeds versus reduction in overall mortality in this cohort versus other cohorts that show that bleeding is a hazard for mortality?” he asked.

“Bleeding tends to be seen in little old ladies,” Gibson replied. “It is a tremendous confounder. Is it the bleeding that’s killing you or all the confounders? I have to say, when we do this kind of analysis in multiple TIMI studies, we do not . . . see that major bleeding is related to long-term mortality. It's related to your risk within the 2 weeks surrounding the event, but beyond 2 weeks, we don't see it's related to mortality. So, I think in more carefully adjusted analysis and very large, massive data sets, that association is not born out, at least in the TIMI studies.”

At this point, Waksman followed up: despite the data from all three studies in this space, he said, “you see a huge lag in terms of practice adoption for rivaroxaban. So what needs to happen?”

“We have to have a conversation about bleeding, another conversation,” Gibson said. “No one comes to your office and says: ‘Thank you for saving my life or preventing the event that never happened.’ They don't know that they were going to die. They do come to your office and say: ‘I've had some gum bleeding or some nose bleeding.’ So there's this asymmetry between the complication of bleeding and the deaths and MIs and strokes that we're preventing, that if they don’t happen, you don’t know about them.”

The position of the FDA up until now seems to be that survival of a bleed is “not at same level as a stroke or a heart attack,” he continued. “So when they look at these events, they look at fatal and irreversible events, death, MI, stroke, or [intracranial hemorrhage], and there we do see a benefit in those fatal and irreversible events once you excluded the survivable things.”

  • Gibson CM. Rivaroxaban 2.5 mg BID combined with dual antiplatelet therapy for the prevention of death/MI/stroke: a patient level data meta-analysis of ATLAS-ACS-2 TIMI-51 and COMMANDER-HF. Presented at: CRT 2019. March 3, 2019. Washington, DC.

  • Gibson reports receiving research grant support and consulting fees in the past from all major manufacturers of antiplatelets and antithrombins.