NEW ORLEANS – Real-world use of rivaroxaban produced comparatively low rates of cardiovascular and major bleeding events when used for stroke prevention in unselected patients with atrial fibrillation.
The centrally adjudicated annual stroke rate was 1.4% among 1,194 patients given once-daily rivaroxaban (Xarelto) by 230 physicians from private practices and community hospitals across Saxony, Germany, in the prospective Dresden Novel Oral Anticoagulant (NOAC) registry.
This result falls within the annual stroke rate of 1.7% seen in the ROCKET-AF trial, which served as the basis for rivaroxaban’s November 2011 U.S. approval in this setting, Dr. Jan Beyer-Westendorf reported at the annual meeting of the American Society of Hematology.
Nonmajor, clinically relevant bleeding occurred in 21.6% of patients per year in the NOAC registry, while major bleeding was 3.4% per year.
Patrice Wendling/Frontline Medical News
Dr. Jan Beyer-Westendorf
Again, this is within the range of the phase III ROCKET-AF results (3.6% per year) and lower than the rate of up to 8% seen in the daily care of patients on vitamin K antagonists, he said. Gastrointestinal bleeding, a known side effect of rivaroxaban, was the most common major bleed. Intracranial bleeds were rare at three events.
Adherence to the oral direct factor Xa inhibitor was high at 12 months (13.7%), compared with up to 25% of patients who discontinue vitamin K antagonist therapy in the first year.
"In our population, the discontinuation rate was 14%, so basically we don’t have any concern that we put these patients at more risk in a daily care situation by treating them with rivaroxaban," said Dr. Beyer-Westendorf of the University Hospital Carl Gustav Carus, Dresden, Germany.
According to the analysis, 2,345 patients have been prospectively enrolled in the NOAC registry since October 2011; 1,194 with atrial fibrillation have been treated with anticoagulation to prevent venous thromboembolism (VTE) or stroke. No patients have been lost to follow-up, which now stands at 2,313 patient-years.
NOAC patients were slightly older at baseline than were those in ROCKET-AF (74.8 years vs. 73 years), less likely to have had prior vitamin K antagonists (37% vs. 62%), and had lower CHADS2 (Cardiac failure, Hypertension, Age, Diabetes Stroke system) scores (mean, 2.4 vs. 3.48). Higher CHADS2 scores are associated with worse outcomes, including higher rates of major bleeding.
During follow-up through October 2013, 53 NOAC patients (4.4%) experienced a major vascular event including 22 strokes, transient ischemic attacks (TIA), systemic emboli; 15 acute coronary syndromes; and four VTEs, he said. In all, 56 patients died (4.7%).
Patients on rivaroxaban 20 mg/day vs. 15 mg/day had a significantly lower annual stroke rate (0.9% vs. 2.3%; P = .052), defined as a new stroke, TIA, or systemic embolism.
Compared with those on rivaroxaban 20 mg, patients on the lower dose had higher baseline CHADS2 scores (mean, 2.8 vs. 2.2), were older (79 years vs. 73 years), more likely to have had a prior stroke (17% vs. 12.5%) or prior vitamin K antagonist therapy (40% vs. 36%), and took more concomitant drugs (mean, 6.4 vs. 5.4).
"It’s no surprise that these patients get a dose reduction because they have more comorbidities; they are at high risk of bleeding, and so it’s not a surprise that they have a slightly higher event rate," said Dr. Beyer-Westendorf, who noted that the lower dose did not reduce the risk of bleeding.
The NOAC registry is supported by scientific grants from Bayer Healthcare, Boehringer Ingelheim, and Pfizer. Dr. Beyer-Westendorf reported research funding from and serving as a speaker for these firms.
Bleeding complications detailed
In a separate presentation at ASH, Dr. Beyer-Westendorf detailed the pattern and management of bleeding complications in NOAC patients treated for stroke prevention in atrial fibrillation or VTE with rivaroxaban, dabigatran (Pradaxa), or apixaban (Eliquis).
These complications are a major concern for practitioners because there isn’t an emergency lab test available or rescue medications.
Of the 1,241 bleeding events that have occurred so far in 879 patients, 742 were minor (60%), 425 were nonmajor clinically relevant (34.2%), and 74 were major (6%).
Major bleeds per year of therapy were reported with rivaroxaban in 3.4% of patients with atrial fibrillation and in 4.4% with VTE. The major bleeding event rate with dabigatran in patients with atrial fibrillation was 2.6/100 patient-years at the 110-mg dose and 2.0/100 patient-years at the 150-mg dose. Short-term follow-up and low numbers of dabigatran and apixaban patients did not allow for sound event-rate calculations, according to Dr. Beyer-Westendorf, who stressed that no direct comparisons should be made between event rates for the different agents since the patients were in different cohorts and not in a randomized trial.