ORLANDO – Adherence to the novel oral anticoagulants (NOACs) is surprisingly poor in clinical practice, Xiaoxi Yao, Ph.D., reported at the American Heart Association scientific sessions.
Her retrospective study of nearly 65,000 patients with atrial fibrillation who initiated therapy with apixaban, dabigatran, rivaroxaban, or warfarin showed that during a median 1.1 years of follow-up fewer than half of all patients were treatment adherent, with adherence defined as possession of sufficient medication to cover at least 80% of days.
Bruce Jancin/Frontline Medical News
Xiaoxi Yao, Ph.D.
Adherence rates, while uniformly suboptimal, nevertheless varied considerably: lowest at 38.5% for dabigatran, followed by 40.2% for warfarin, 50.5% for rivaroxaban, and 61.9% for apixaban.
This poor adherence to NOACs in real-world clinical practice is surprising in light of the drugs’ greater convenience, with fewer drug interactions than warfarin and no need for laboratory monitoring, observed Dr. Yao of the Mayo Clinic in Rochester, Minn.
It’s possible, although speculative, that the NOACs’ greater convenience paradoxically contributes to the low adherence rates, since unlike warfarin, NOACs don’t require regular interactions with the health care system for INR monitoring. And then there is the hefty cost of the novel agents, she added.
The study population consisted of 3,900 patients with atrial fibrillation who initiated oral anticoagulation with apixaban (Eliquis), 10,235 who started on dabigatran (Pradaxa), 12,366 on rivaroxaban (Xarelto), and 38,190 on warfarin. The analysis utilized claims data from a large U.S. commercial insurance database.
Adherence rates were better among patients with greater stroke risk as reflected by their CHA2DS2-VASc scores. For example, at the high end of the adherence spectrum, the adherence rate for apixaban was 50% in patients with a CHA2DS2-VASc score of 0-1, rising to 62% with a score of 2-3 and 64% with a score of 4 or more. The corresponding adherence rates for dabigatran were 25% in patients with a CHA2DS2-VASc of 0-1, 40% among those with a score of 2-3, and 42% in patients with a score of 4 or higher.
Dr. Yao and coinvestigators were interested in whether lower adherence to oral anticoagulation was associated with worse outcomes. This proved to be the case with regard to stroke rate for patients with a CHA2DS2-VASc score of 2 or more, where a clear dose-response relationship was evident between the event rate and cumulative time off oral anticoagulation during follow-up.
Among patients with a CHA2DS2-VASc of 2 or 3, the stroke rate was nearly twice as high among those off oral anticoagulation for a total of 3-6 months and three times greater if off therapy for more than 6 months than in those with a total time off of less than 1 week. The stroke rate was even higher in patients with a CHA2DS2-VASc of 4 or more who had suboptimal adherence.
An unexpected finding, she continued, was that among patients with a CHA2DS2-VASc score of 2 or more there was no significant relationship between cumulative time off oral anticoagulation and the risk of major bleeding unless they were off treatment for a total of 6 months or more; only then was the major bleeding risk lower than in patients whose total time off therapy was less than a week. Also, one would expect that when patients are off oral anticoagulation they should be at significantly lower risk of intracranial hemorrhage than when on-therapy, but this proved not to be the case.
For patients at substantial stroke risk as indicated by a CHA2DS2-VASc score of at least 2, this finding about off-treatment bleeding risk actually constitutes a good argument for sticking to their medication, in Dr. Yao’s view.
“Physicians and patients often fear bleeding, especially intracranial hemorrhage, but we found that for patients at higher risk for stroke there is little difference in intracranial hemorrhage risk whether you’re on or off of oral anticoagulation. So higher-risk patients should definitely adhere to their medication because of the stroke prevention benefit. However, in low-risk patients with a CHA2DS2-VASc of 0-1, the benefits of oral anticoagulation may not always outweigh the harm,” she said.
Dr. Yao reported having no financial conflicts of interest regarding her study.