This article is more than 5 years old. Information may no longer be current.
Apixaban may confer less stroke, systemic embolism, bleeding vs. rivaroxaban
In a retrospective cohort study, patients with atrial fibrillation who were treated with apixaban experienced lower incidence of ischemic stroke, systemic embolism and bleeding compared with patients who received rivaroxaban.
According to research published in the Annals of Internal Medicine, among patients with AF prescribed apixaban (Eliquis, Bristol-Myers Squibb/Pfizer), incidence of ischemic stroke or systemic embolism was 6.6 per 1,000 person-years, compared with eight per 1,000 person-years for patients prescribed rivaroxaban (Xarelto, Janssen/Bayer; HR = 0.82; 95% CI, 0.68-0.98; rate difference, 1.4 fewer events per 1,000 person-years; 95% CI, 0-2.7).
Moreover, patients prescribed apixaban also experienced a lower incidence of gastrointestinal bleeding or intracranial hemorrhage (12.9 per 1,000 person-years) compared with patients who were prescribed rivaroxaban (21.9 per 1,000 person-years; HR = 0.58; 95% CI, 0.52-0.66; rate difference, nine fewer events per 1,000 person-years; 95% CI, 6.9-11.1).
In addition, in a propensity-matched subanalysis of patients older than 70 years, patients prescribed apixaban had a stroke or systemic embolism incidence of 8.3 events per 1,000 person-years compared with 10.5 events per 1,000 person-years among those prescribed rivaroxaban (HR = 0.79; 95% CI, 0.63-0.99).
“In this study of more than 90,000 patients with atrial fibrillation in the United States, apixaban was associated with a lower rate of stroke or systemic embolism, as well as bleeding, compared with rivaroxaban,” Michael Fralick, MD, PhD, SM, research fellow in the division of pharmacoepidemiology and pharmacoeconomics at Brigham and Women’s Hospital and clinician scientist at Mount Sinai Hospital in Toronto, and colleagues wrote. “Findings were robust across several subgroup and sensitivity analyses, including a population restricted to patients older than 70 years.”
In other findings, there was no difference between the groups in hepatitis incidence or HF hospitalization.
Researchers assessed patient data (mean age, 69 years; 40% women; mean apixaban follow-up, 288 days; mean rivaroxaban follow-up, 291 days) from the U.S. nationwide commercial health care database with the aim of determining the safety and efficacy outcomes of apixaban compared with rivaroxaban.
“Apixaban may be safer and more effective than rivaroxaban for treating nonvalvular atrial fibrillation,” the researchers wrote. “Until head-to-head clinical trial data are available, the results of our study, which included a large sample of patients seen in routine care, provides updated evidence in support of apixaban for treating nonvalvular atrial fibrillation.” – by Scott Buzby
Disclosures: Fralick reports no relevant financial disclosures. Please see the study for all other authors’ relevant financial disclosures.
Perspective
Back to Top
I am disappointed in the Annals journal allowing such a clear conclusion of one drug being better than another. I do not see as such a clear finding and don’t think the data the investigators have can support that conclusion. Mainly, this is a nonrandomized comparison of the use of two drugs for the prevention of stroke and systemic embolism; a retrospective observational analysis. The investigators can’t determine why one agent was chosen by physician for one set of patients vs. other patients. When they just show the raw rates of observed stroke or systemic embolism in the paper, there is no difference. The rate of stroke with rivaroxaban is 7.27 per 1,000 person-years and with apixaban is 7.21 per 1,000 person-years. Yet, after propensity analysis and after a variety of statistical accounting for differences in patients, they then report 7.48 per 1,000 person-years vs. 6.38 per 1,000 person-years. This means, somehow, the stroke rate goes up for rivaroxaban vs. going down for apixaban after statistical adjustment — saying that the rivaroxaban patients were healthier than the apixaban patients? On face value, it doesn’t make sense to me, as critical data like renal function and hemoglobin labs are not present.
The researchers try to account for some of the difference or statistical co-linearity by using a “negative control,” the risk for HF, which they say should be the same between the groups. However, depending on the underlying condition (AF), this rate may not be the same and they don’t have underlying left ventricular ejection fraction.
Simply put, the researchers observed in a nonrandomized retrospective observational study that the rate of stroke in patients with AF getting rivaroxaban or apixaban is identical. They then perform a variety of statistical analyses to control for the differences they see in patients — they can’t control for the unmeasured differences that they don’t have like renal function, hemoglobin, LVEF — and conclude one is better than another. Unfortunately, this is not conclusive, and we need randomized trial data to conclude this. Until then, another conclusion could be that clinicians in practice are identifying the right patients to give the different agents to based on their clinical acumen (given the near-identical stroke rates observed before any statistical adjustment). I would suggest looking at the trial patients and using the therapies as studied in patients with risk similar to those in the randomized trials.