Apixaban Tops Rivaroxaban in AF Patients With Valve Disease, Claims Data Suggest

The use of apixaban (Eliquis; Bristol-Myers Squibb) over rivaroxaban (Xarelto; Bayer/Janssen) in patients with atrial fibrillation (AF) and valvular heart disease is associated with better clinical outcomes and improved safety, according to a new observational study.

In roughly 20,000 new users of the two factor Xa inhibitors, apixaban was linked with a lower risk of ischemic stroke and pulmonary embolism and a lower risk of bleeding, particularly gastrointestinal bleeding.  

The study, which was published this week in the Annals of Internal Medicine, addresses a significant evidence gap because head-to-head comparisons of direct oral anticoagulants (DOACs) are lacking. “Until evidence from randomized controlled trials becomes available, we believe clinicians should consider our findings when selecting anticoagulants in patients with AF and valvular heart disease,” Ghadeer K. Dawwas, PhD, MBA (University of Pennsylvania Perelman School of Medicine, Philadelphia), and colleagues conclude.

To date, most studies evaluating the safety and efficacy of oral anticoagulants, such as ARISTOTLE and ROCKET AF, which compared the new DOACs against warfarin, enrolled few patients with AF and valvular heart disease, Dawwas told TCTMD. Despite that, valvular heart disease is very common in patients with AF—the prevalence can range from 30% to nearly 70%—and rivaroxaban and apixaban are the two most commonly prescribed drugs in this setting.

“The question then is if apixaban and rivaroxaban are the two most commonly used, how do you choose between the two agents in clinical practice?” said Dawwas.

The new data add to a growing body of evidence showing that apixaban may be a safer alternative to rivaroxaban when used in patients with AF. Researchers led by Neena Abraham, MD (Mayo Clinic, Phoenix, AZ), investigated the safety profile of various DOACs and found that apixaban was the safest agent when it came to the risk of gastrointestinal bleeding while rivaroxaban had the least favorable safety profile. An observational study from Iceland confirmed those findings, with researchers reporting that gastrointestinal bleeding was nearly 50% more common with rivaroxaban than with apixaban.

To TCTMD, Abraham said the latest results, along with other published studies, make the decision relatively straightforward when it comes to deciding which DOAC to use. “Apixaban should be considered the first-choice DOAC in patients because of its safety and comparative effectiveness profiles,” she said. “This paper in Annals of Internal Medicine is just the latest of many to show a superior safety profile of apixaban.” 

Abraham pointed that the safer gastrointestinal bleeding profile with apixaban is preserved even in very elderly patients who are at the highest risk of bleeding.

More Efficacious, Better Safety

The 2019 guidelines from the American Heart Association, the American College of Cardiology, and the Heart Rhythm Society recommend a non-vitamin K antagonist (NOAC) over warfarin for patients with nonvalvular AF and an elevated CHA₂DS₂-VASc score (class I, level of evidence A), but do not give preference to any of the available drugs. In this setting, which excludes patients with moderate-to-severe mitral stenosis or a mechanical heart valve, options include dabigatran (Pradaxa; Boehringer Ingelheim), edoxaban (Savaysa; Daiichi Sankyo), apixaban, and rivaroxaban.

Valvular AF typically refers to patients with moderate-to-severe mitral stenosis, which potentially requires surgery, or AF in the presence of a mechanical valve. For these patients, long-term oral anticoagulation with warfarin is recommended. Nonvalvular AF, on the other hand, does not mean the absence of valvular heart disease, but rather is used to define patients without moderate-to-severe mitral stenosis or a mechanical valve.

To look at relative safety and efficacy of apixaban and rivaroxaban, the researchers analyzed data from a commercial health insurance database that included 33,907 patients with AF and valvular heart disease who started a new prescription of either DOAC between 2013 and 2020. In the propensity-matched analysis of 19,894 patients, there was no difference in any of the covariates, including prescribing provider, type of valvular heart disease, baseline comorbidities, medication use, or baseline HAS-BLED and CHA₂DS₂-VASc scores. Median follow-up was 183 and 155 days in the apixaban- and rivaroxaban-treatment arms, respectively.

In the matched analysis, use of apixaban was associated with a 43% lower risk of ischemic stroke or systemic embolism (HR 0.57; 95% CI 0.40-0.80). Separately, apixaban was associated with a 42% lower risk of ischemic stroke, but the reduction in systemic embolism was not statistically significant. Apixaban was associated with a 49% lower risk of bleeding (HR 0.51; 95% CI 0.41-0.62), including a 49% lower risk of gastrointestinal bleeding. There was a trend toward less intracranial hemorrhage, but it was not statistically significant. There was no difference in all-cause mortality between the two DOACs.

Maybe Not Ready for Guideline Changes Yet

In an array of sensitivity and subgroup analyses, investigators say the results were largely consistent, with no evidence of an effect modified age, sex, presence of chronic kidney disease, or risk scores. The lower risk of bleeding was also seen in patients regardless of the type of valvular heart disease, including mitral, aortic, and tricuspid valve disease. However, Dawwas pointed out that valvular heart disease was captured using ICD codes and there is no validated definition of the different subtypes of valve disease using claims data, which makes it challenging to draw conclusions about these findings.  

Overall, Dawwas said their results aren’t particularly surprising given the existing data. In a study published earlier this year, for example, they compared the safety and efficacy of apixaban and rivaroxaban in patients with venous thromboembolism (VTE) using the same database. Again, apixaban-treated patients had lower rates of recurrent VTE and less bleeding. 

Still, when it comes to the guidelines, Dawwas is uncertain they should be changed to recommend apixaban over other DOACs.

“There are many factors that need to be taken into consideration, such as, for example, cost, patient preference, and adherence,” she said, noting that apixaban is administered twice daily whereas rivaroxaban is taken just once per day. “Also, the study was conducted using a commercial database and we know the results from this study might not be generalizable to other patient populations. We still need to conduct further studies before taking this and saying the guidelines should change. I don’t see that happening at this point.”

There is a randomized, controlled trial currently ongoing: COBRRA AF. In that study, which will include more than 3,000 patients, Canadian researchers are looking at the relative safety of apixaban and rivaroxaban in AF patients, with the primary endpoint defined as the risk of clinically adjudicated relevant bleeding at 12 months. A similar study, known as COBRRA, is being conducted by the same researchers in 2,700 patients with VTE.

Dawwas said observational studies like theirs are valuable because they represent a broader, less-selected patient population than those in randomized trials, but that both types of studies will help inform future guidelines.