More Excess Bleeding Seen With Rivaroxaban vs Other NOACs and Aspirin

Two of three commonly prescribed non-vitamin K oral anticoagulants (NOACs)—dabigatran and apixaban—have similar rates of major bleeding compared with aspirin, a large systematic review and meta-analysis suggests.

Rivaroxaban, on the other hand, is associated with an increased signal of harm.

“What our analysis shows is that within the limitations of the studies that have been done, it does appear that the risk of bleeding with rivaroxaban is higher versus aspirin. We don't see that same kind of signal with the other two NOAC drugs,” Michael Ke Wang, MD (Population Health Research Institute, McMaster University, Hamilton, Canada), told TCTMD.

While smaller, more limited analyses have suggested that the risk of bleeding may be similar between NOACs and single antiplatelet therapy, many clinicians remain unsure, especially for their patients who are at higher bleeding risk, said Wang, who led the new study, which was published last week in the Annals of Internal Medicine.

“All else being equal, I think there is a consideration in the study that there may be a preference for apixaban or dabigatran when it comes to bleeding risk, but there are always important patient-specific factors to consider as well,” Wang said. “If a clinician has other reasons or has comfort level with rivaroxaban, it remains a reasonable choice.”

For example, there may be practical considerations in some patients, he added, including the need for a once-daily medication like rivaroxaban to foster compliance.

Adds to Prior Observations

For the analysis, Wang and colleagues included nine RCTs in which bleeding risk of NOACs was compared with single-agent therapy in 26,224 patients (mean age 67 years; 42% female) treated with anticoagulation for a variety of reasons.

The population consisted of patients with a recent history of stroke, clinical atrial fibrillation (AF) and intracranial hemorrhage, AF considered unsuitable for vitamin K antagonist therapy, subclinical AF detected on pacemakers and defibrillators, and venous thromboembolism after six to twelve months of initial anticoagulation. The mean follow-up duration was 20 months.

In the NOAC group, apixaban (Eliquis; Bristol Myers Squibb/Pfizer) was used in four trials, rivaroxaban (Xarelto; Janssen/Bayer) in two trials, dabigatran (Pradaxa; Boehringer Ingelheim) in two trials, and either apixaban or dabigatran in a single trial. Once-daily aspirin was used in all trials as the comparator. The two rivaroxaban studies included in the meta-analysis were NAVIGATE ESUS, which showed an excess risk of bleeding with rivaroxaban versus aspirin, and EINSTEIN CHOICE, which did not.

Overall, bleeding incidence was low in all the studies, with rates of major bleeding and intracranial hemorrhage of 2.16% and 0.66%, respectively, out to nearly 2 years.

There was no difference between aspirin and apixaban for the endpoint of major bleeding, which included fatal bleeding, symptomatic bleeding in a critical area or organ, or bleeding causing a decrease in hemoglobin level of ≥ 20 g/L or leading to transfusion of ≥ 2 units of whole blood or red blood cells.

Similarly, for dabigatran versus aspirin, there was no significant difference in major bleeding. With intracranial hemorrhage, there was no significant difference in risk between apixaban and aspirin and no difference between dabigatran and aspirin.

For rivaroxaban versus aspirin, however, the absolute risk difference was 0.9 percentage points higher (95% CI -0.1 to 3.7 percentage points) for major bleeding and 0.3 percentage points higher (95% CI -0.1 to 79.7 percentage points) for intracranial hemorrhage. Neither difference was significant, however, and Wang said that given the wide confidence intervals, too, a null effect could not be ruled out.

Wang and colleagues say their data build on prior studies, including a network meta-analysis of AF trials that showed excess major bleeding with rivaroxaban compared with apixaban and edoxaban (Savaysa; Daiichi Sankyo) as well as several large observational studies. The latter included one that showed a 46% greater risk of GI bleeding with rivaroxaban compared with apixaban and another that found lower risks of ischemic and bleeding risks in AF patients newly started on apixaban versus rivaroxaban.

To TCTMD, Wang said the addition of more recent trial data and a varied population of anticoagulant users should help reassure clinicians on the safety of apixaban and dabigatran compared with rivaroxaban, particularly when counseling patients about options and weighing the risks and benefits of switching patients from aspirin to a NOAC.

“These were all large, high-quality, randomized trials all looking at long-term bleeding risks using very well-defined definitions. In many ways, they did measure things similarly, so despite the limitations of a meta-analyses, I do think that it is [helpful in] considering the evidence as a whole,” he added.