Rivaroxaban or Apixaban?

Dear Readers,

Direct oral anticoagulants (DOACs) have revolutionized efforts to prevent strokes in patients with atrial fibrillation (AF). The few patients who continue to opt for warfarin do so either because they have a contraindication to the newer drugs, such as having a mechanical heart valve, or the patient is reluctant for cost reasons (currently, the average list price of a DOAC is $500/month). However, the compelling safety, efficacy, and convenience of DOACs relative to warfarin have resulted in a majority of patients who are now taking a DOAC. Most patients taking a DOAC (about 60%) are prescribed apixaban,¹ with the remainder being prescribed rivaroxaban. (Notice that the suffix for each drug’s generic name is because they both block factor Xa.)

For years, both patients and doctors have been asking which DOAC is better. A clear benefit of rivaroxaban is that it is dosed once per day compared to apixaban, which is twice per day. The importance of medication compliance has been a compelling argument to prescribe rivaroxaban. And because there has never been, and will likely never be, a head-to-head randomized comparison between the two DOACs, there remains controversy as to which drug is better. In the clinical trials that led to approval of each of the drugs, the DOACs were compared to warfarin. Although the studies evaluating apixaban seemed to demonstrate better outcomes relative to warfarin, the trials included patients with different risk profiles, had different study designs, and could not be directly compared. Meta-analyses and non-randomized cohort trials have found evidence that apixaban might be better, but these studies all have enough limitations that it has been difficult to declare a winner.

A new study was recently published in the Journal of the American Medical Association² that might be the most compelling evidence yet. Ray and colleagues performed a retrospective cohort study using computerized enrollment and claims files for nearly 600,000 U.S. Medicare beneficiaries 65 years or older with AF treated with either rivaroxaban or apixaban and followed for four years. As expected, the demographics of the patients in each group were very similar. Furthermore, the investigators went to great lengths to reduce confounding by adjusting for any differences in comorbidities, and reported adjusted risks. The results showed a significant difference between the two DOACs that is difficult to argue — treatment with rivaroxaban compared with apixaban was associated with a significantly increased risk of both major ischemic and hemorrhagic events. The outcomes are shown in percentages in the Figure.

Although one could argue that the differences between the two DOACs are trivially small and that the study is splitting hairs, the findings from this study do seem to make a very strong case that apixaban is actually “better” than rivaroxaban. Knowing that their paper will receive intense scrutiny and stir controversy, the authors wisely addressed several potential limitations of their study. For example, they acknowledged that residual confounding is still possible. However, significant confounding seems unlikely since there is no reason to think that the decision of which DOAC to prescribe is based on patient characteristics, a fact supported by the similarity in the patient demographics. Any other potential arguments for why the findings might be inaccurate would be based on the retrospective cohort study design, but the usual arguments seem unlikely in this case. For example, why would misclassification of outcomes or failure to identify discontinuation of the drug differ between groups?

Why would one DOAC be superior to another in both bleeding and embolic risks? A plausible explanation is that the daily dosing of rivaroxaban, compared to the twice-daily dosing of apixaban, results in greater fluctuations in the anticoagulant effect such that bleeding events are more likely when plasma levels peak and clot-based embolic events are more likely at low trough levels. One might speculate that a drug that causes more bleeding might be discontinued more often, and therefore, lead to more strokes. However, the authors argue against such a theory, stating that “when the cohort was restricted to patients with no more than 7 days between refills — a marker for adherence — findings were unchanged.”² One also might speculate that rivaroxaban is slightly less effective at preventing strokes and that the reason bleeding rates are higher is because rivaroxaban has a more direct effect on the gut lining, increasing the risk of gastrointestinal bleeding, the outcome that explains most of the differences in bleeding. However, intracranial bleeding was also more common with rivaroxaban.

These data are timely and newsworthy. As of late, some pharmacies are no longer including apixaban on their formulary. The new 2022 CVS Caremark formulary includes only two oral anticoagulants: warfarin and rivaroxaban.³ The reasons for this restriction are unclear, but given the high costs of both drugs, insurance coverage continues to be the most common determinant of which DOAC patients take. This should become less of a factor when these drugs become generic.⁴

Both rivaroxaban and apixaban are safe and effective in preventing strokes in patients with AF. However, although the differences between the performances of the two most popular DOACs are small, the Medicare data published by Ray and colleagues makes apixaban the gold medalist. 

Bradley P. Knight, MD, FACC, FHRS

@DrBradleyKnight
Editor-in-Chief, EP Lab Digest

References

1. Zhu  J, Alexander  GC, Nazarian  S, Segal  JB, Wu  AW.  Trends and variation in oral anticoagulant choice in patients with atrial fibrillation, 2010-2017. Pharmacotherapy. 2018;38(9):907-920. doi: 10.1002/phar.2158

2. Ray WA, Chung CP, Stein CM, et al. Association of rivaroxaban vs apixaban with major ischemic or hemorrhagic events in patients with atrial fibrillation. JAMA. 2021;326(23):2395-2404. doi: 10.1001/jama.2021.21222

3. CVS Caremark Value Formulary, Effective as of 01/01/2022. Accessed January 8, 2022. https://www.caremark.com/portal/asset/Value_Formulary.pdf

4. FDA approves first generics of Eliquis. US FDA. Published December 23, 2019. Accessed January 8, 2022. https://www.fda.gov/news-events/press-announcements/fda-approves-first-generics-eliquis

Disclosures: Dr. Knight reports that he has served as a consultant, speaker, investigator, and/or has received EP fellowship grant support from Abbott, AtriCure, Baylis Medical, Biosense Webster, Biotronik, Boston Scientific, CVRx, Medtronic, Philips, and Sanofi. He has no equity or ownership in any of these companies.