Asian Patients With A-fib Don’t Gain Anything From Low-Dose Rivaroxaban
Avoidance of the standard dose did not lessen bleeding and carried a higher risk of MI.
Use of lower-than-recommended doses of oral anticoagulants due to concerns about bleeding is common in Asia, but a study out of Taiwan suggests there’s little to gain—and perhaps some harm—from the practice.
Outside experts who weighed in with their interpretation of the findings, however, did not necessarily agree on the implications for patients and practitioners.
According to research led by Yi-Cheng Lin, PharmD (Taipei Medical University Hospital, Taiwan), patients taking a low dose of rivaroxaban (Xarelto; Bayer/Janssen) off-label had similar safety and effectiveness outcomes as patients receiving the standard dose, except for a higher risk of MI (subdistribution HR 2.26; 95% CI 1.13-4.52).
There was no hint of a decrease in major or nonmajor clinically relevant bleeding with the low dose, and the difference in systemic embolism—though not statistically significant—suggested that avoidance of the standard dose could be increasing risk (subdistribution HR 2.04; 95% CI 0.95-4.36).
Lin and colleagues reported their findings online ahead of the July 31, 2018, issue of the Journal of the American College of Cardiology.
Consistent with prior research, they write, “our study indicated that low-dose rivaroxaban may be less effective than standard-dose rivaroxaban without lowering the adverse effect of bleeding in an Asian population.”
Commenting for TCTMD, Peter Brønnum Nielsen, PhD (Aalborg University Hospital, Denmark), lead author of an accompanying editorial, said in an email: “Lin and colleagues have made an interesting contribution of observational data for an inherently difficult patient population [in whom] to safely manage oral anticoagulant treatment.”
He said, though, that because of the lack of information on how well the prescribed doses adhered to label-recommended dosing, “the research question is difficult to translate into a practical clinical question.”
“As highlighted in the editorial, causal inference from comparative effectiveness studies requires careful consideration of study design,” Nielsen said. “[Relating] the proposed research question with an everyday clinical question would ease the interpretation of the study results, and could thereby guide clinical practice for the Asian AF population receiving rivaroxaban for stroke prevention.”
Low Dose Used in More Than One-Third
The standard dose of rivaroxaban for patients with A-fib is 20 mg once daily, with a reduced 15-mg dose used for patients with renal impairment. However, a dose of 10 mg, which is available for indications outside of stroke prevention in A-fib, is used extensively in Asia for patients with the arrhythmia, despite the lack of trial data supporting the safety and efficacy of this approach.
To provide insight into the question, Lin et al examined data from the National Health Insurance Research Database in Taiwan for patients diagnosed with atrial fibrillation or flutter who filled a prescription for rivaroxaban between May 2014 and September 2015. The analysis included 6,558 patients; 36% were taking the low 10-mg dose, with the rest were receiving a standard dose of 15 or 20 mg.
The lack of differences in most safety and clinical outcomes between dosing regimens was observed after the researchers used inverse probability of treatment weighting to adjust for variation in baseline characteristics. For MI, the increased risk associated with the low dose of rivaroxaban was driven by NSTEMI (subdistribution HR 2.54; 95% CI 1.17-5.49), with no difference in STEMI.
Is the Low-Dose Strategy a Winning One?
In their editorial, Nielsen and his colleagues say the results of this analysis are probably not strong enough to change treatment recommendations for Asian patients with A-fib, but do provide some insights.
“Perhaps the most accurate interpretation of the results is that the routine practice for choosing the optimal dosage of rivaroxaban has been safe in this patient population, and that rivaroxaban has an acceptable safety profile and effectively reduces stroke risk,” they write.
But Jonathan Piccini, MD (Duke University, Durham, NC), disputed that view. “I don’t think the article in any way, shape, or form supports the use of a lower dose,” he said to TCTMD, pointing out that there are prior analyses suggesting that patients who receive lower doses of the non-vitamin K antagonist oral anticoagulants (NOACs) have worse outcomes.
Although the study by Lin et al is valuable, Piccini said, it lacks a key piece of information—the exact level of renal function for the patients—that would be necessary to assess the appropriateness of the doses used.
Both the increased risk of MI and the trend toward more systemic embolism in the low-dose group should give clinicians and patients pause, he added.
“The reason we give these medications is to prevent ischemic stroke and systemic embolism, and I think some of the associations that suggest a potential higher frequency of thrombotic events or embolic events is concerning,” Piccini said. “Maybe the lower dose is associated with worse efficacy and increased harm, and it doesn’t appear to be associated with less bleeding, so it’s not clear to me how a lower dose would be a winning strategy for a given patient.”
When asked about his interpretation of the study results, Nielsen noted that absolute risks of both systemic embolism and MI were low and that information on potentially important confounders—ie, smoking, prior MI, and PAD—was missing.
“The low risk of systemic embolism and the very low risk MI should warrant very careful interpretation of the study results if one would like to understand the study outcomes as cause and effect (under strong assumptions). . . . Personally, I wouldn’t put too much emphasis on events occurring 10 times less frequently than stroke and major bleeding in an AF population—disregarding hazard ratios and P-values,” he said.
Lin Y-C, Chien S-C, Hsieh Y-C, et al. Effectiveness and safety of standard- and low-dose rivaroxaban in Asians with atrial fibrillation. J Am Coll Cardiol. 2018;72:477-485.
Nielsen BP, Skjøth F, Søgaard M. Causal inference from real-world data: a request for asking the proper research question. J Am Coll Cardiol. 2018;72:486-488.
- The study was supported by Taipei Medical University and the National Science Council in Taiwan.
- Lin reports no relevant conflicts of interest.
- Piccini reports receiving funding for clinical research from Abbott Medical, ARCA biopharma, Boston Scientific, Gilead, Janssen Pharmaceuticals, the National Heart, Lung, and Blood Institute, and Verily; and serving as a consultant to Allergan, Bayer, Johnson & Johnson, Medtronic, Sanofi, and Philips.