ROCKET AF Reveals Higher GI Bleeding Rates With Rivaroxaban

Pam Harrison

November 30, 2015

WASHINGTON, DC — More gastrointestinal (GI) bleeding occurred in patients with nonvalvular atrial fibrillation (AF) treated with the novel oral anticoagulant (NOAC) rivaroxaban (Xarelto, Bayer Pharma/Janssen Pharmaceuticals) than in those who received warfarin, indicates a retrospective analysis of the ROCKET AF trial[x]. However, it also found that rates of severe and fatal bleeds were not significantly different between the two groups and event rates for both types of bleeds were rare.

The study included a multivariate analysis of potential clinical risk factors for GI bleeding in the trial; the strongest to emerge were anemia at baseline, a history of GI bleeding, long-term aspirin use, and the use of rivaroxaban itself.

"One of the important things we found was that patients with GI bleeding didn't have higher HAS-BLED scores compared with those who did not bleed, and they had the same CHA2DS2-VASc scores as well," said Dr Matthew Sherwood (Duke Clinical Research Institute, Durham, NC), referring to risk-assessment tools for bleeding and stroke, respectively. Sherwood is lead author on the study's publication in the December 1, 2015 issue of the Journal of the American College of Cardiology.

"So while I think risk stratification is important," he told heartwire from Medscape, "the more important message is that physicians need to address modifiable risk factors, such as the concomitant use of aspirin and antiplatelet agents, before starting an oral anticoagulant if patients have a higher risk for bleeding to start with—such as those who have a history of prior GI bleeding."

It does seem as if higher doses, higher plasma levels, and higher GI exposure of the factor Xa inhibitors are associated with an increased risk of GI bleeding compared with warfarin, observed Prof Lars Wallentin (Uppsala Clinical Research Center, Sweden) for heartwire .

"Concerning rivaroxaban, it would seem that testing a lower dose or perhaps dosing the drug twice a day or both might be a useful strategy to obtain a better balance between risk of stroke and risk of bleeding," proposed Wallentin, who was not part of the current study, in an email. He also proposed that use of the factor Xa inhibitor apixaban (Eliquis, Pfizer/Bristol-Myers Squibb) at the recommended dose of 5 mg, twice daily, might be another way to find "the sweet spot" when using a NOAC to reduce stroke risk, major bleeding, and mortality in patients with AF at risk for stroke.

Parity for Severe and Fatal Bleeds

ROCKET AF randomized patients to either rivaroxaban or to warfarin for the prevention of stroke and systemic embolism in patients with nonvalvular AF at intermediate to high risk for stroke, as heartwire previously reported. The primary findings were that rivaroxaban was noninferior to warfarin for the prevention of stroke and systemic embolism and that rates of major and nonmajor clinically relevant bleeding were similar between the two treatment groups.

The current analysis included 14,236 patients in the safety-end-point population of trial; it had been preliminarily reported at a meeting 3 years ago and covered then by heart wire .

As the article notes, GI bleeding rates were 42% higher for rivaroxaban-treated patients (hazard ratio [HR] 1.42, 95% CI 1.22–1.66), than for warfarin controls. Some 48% of GI bleeding events occurred in the upper-GI tract, 23% occurred in the lower-GI tract, and 29% occurred in the rectum.

"When separated by clinical severity, there were more major and nonmajor clinically relevant GI bleeding events in rivaroxaban-treated patients," the group writes. On the other hand, for the most severe bleeds—defined as bleeds requiring four or more units of whole blood or peripheral red blood cell transfusion—as well as fatal bleeds, there was a numerical balance between treatment groups, with no statistically significant difference between the two treatment arms.

Overall Rates of GI Bleeding by Treatment Arm
Outcomes Rivaroxaban events/100 patient-years Warfarin events/100 patient-years Rivaroxaban vs warfarin, adjusted P
Major or nonmajor clinical bleeding 3.61 2.60 1.42 <0.0001
Major bleeding 2.00 1.24 1.66 <0.0001
Transfusion >4 units 0.47 0.41 1.19 0.39
Fatal 0.01 (1 patient) 0.04 (5 patients) 0.21 0.15
Nonmajor clinical bleeding 1.75 1.39 1.28 0.023

"I think there is still benefit for using the direct-acting oral agents in patients with AF at intermediate to high risk for stroke, because these agents provide important advantages over warfarin—the first being that they have a lower risk for intracranial hemorrhage and the second that they do not require monitoring. So they provide a consistent level of anticoagulation and therapeutic effect," Sherwood said.

"In addition, they don't have as many interactions with drugs as do the vitamin-K antagonists, specifically warfarin. So those are all important benefits that can't be overlooked."

In an audio commentary accompanying the study's publication, JACC editor Dr Valentin Fuster (Icahn School of Medicine at Mount Sinai, New York, NY) also highlighted the low incidence of severe GI bleeds in both arms and how very rare fatal GI bleeds were across the entire ROCKET AF trial.

"When you take into account the fact that nearly 15,000 patients were included in the study, from my personal viewpoint, the difference between the two treatment groups is minute, and this has implications when it comes to making our own choice of which anticoagulant to use in patients with AF," he said.

NOACs Have "Changed the Stroke-Prevention Landscape"

Some other NOACs also may pose an excess risk of GI bleeds, including dabigatran etexilate (Pradaxa, Boehringer Ingelheim) 150 mg twice daily and edoxaban (Savaysa, Daiichi Sankyo) 60 mg, observe Drs Gregory Lip and Deirdre Lane (University of Birmingham, UK) in an accompanying editorial[2]. In contrast, they note, no excess GI bleeding was seen in the clinical trials with dabigatran 110 mg twice a day or apixaban 5 mg twice day.

When assessing bleeding vs stroke risk, Lip and Lane emphasize that the physicians should use scores specific to each end point, notably the HAS-BLED score to assess bleeding risk and the CHA2DS2-VASc score to assess the risk of stroke. However, "a high HAS-BLED score is not an excuse to withhold oral anticoagulation," they caution.

"Indeed, Sherwood et al emphasize the need for minimizing modifiable risk factors for GI bleeding in patients on oral anticoagulants, which is a very sensible approach," the editorialists writes. They also concur with the study authors' conclusion that, regardless of GI bleeding issues, "we should not forget that overall, the NOACs have changed the landscape for stroke prevention in AF," they state.

Sherwood discloses receiving consulting fees from Boehringer Ingelheim and educational grants from AstraZeneca and Gilead. Lip discloses serving as a consultant for Bayer, Merck, Sanofi, Bristol-Myers Squibb/Pfizer, Daiichi-Sankyo, Biotronik, Medtronic, Portola, and Boehringer Ingelheim and being on the speaker's bureau for Bayer, Bristol-Myers Squibb/Pfizer, Boehringer Ingelheim, Daiichi-Sankyo, Roche, and Medtronic. Lane discloses receiving investigator-initiated educational grants from Boehringer Ingelheim and Bristol-Myers Squibb and honoraria for educational lectures from Boehringer Ingelheim, Bayer Healthcare, and Bristol-Myers Squibb/Pfizer and being on the steering committee of a phase 4 apixaban study (AEGEAN). Wallentin reports receiving institutional research grants from Abbott, AstraZeneca, Bristol-Myers Squibb, Boehringer-Ingelheim, GlaxoSmithKline, InfraRedX, MedCo, Merck, and Roche Diagnostics.


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.