Stroke Rates Lower and Anticoagulation Times Improve with Warfarin Treatment

The use of warfarin for stroke prophylaxis in patients with nonvalvular atrial fibrillation (AF) was associated with a rate of stroke or systemic embolism of 1.66% per year, according to results of a recent meta-analysis [Arch Intern Med. 2012;172(8):623-631].  A previous meta-analysis had estimated the rate to be 2.09%. Improvements in the amount of time spent in therapeutic anticoagulation were also noted.

Nonvalvular AF is associated with a 5-fold increase in the risk of ischemic stroke and is believed to be a factor in 15% to 20% of all ischemic strokes. Warfarin has been used for stroke prophylaxis for decades, but during the past few years, alternative agents have been introduced. These new agents have advantages over warfarin such as ease in administration, consistency in effect, and a lack of interaction with other medications. However, in comparison with warfarin, these newer agents are costly, and consequently many physicians are likely to continue prescribing warfarin.

The researchers conducted a meta-analysis of randomized controlled trials within the past 10 years that compared warfarin with an alternative antithrombotic therapy for primary or secondary stroke prevention in patients with nonvalvular AF. Their objective was to provide physicians with the best current estimate of the risks and benefits associated with warfarin. Trials had to be of high quality (Jadad score of ≥3) and include at least 400 patients in the warfarin cohort for study inclusion.

The primary efficacy outcome was the occurrence of ischemic or hemorrhagic stroke or non-central nervous system (CNS) embolism. Secondary efficacy end points included myocardial infarction (MI), all-cause mortality, and composite adverse vascular events. Safety outcomes included major bleeding, intracranial hemorrhage, clinically relevant nonmajor bleeding, and minor bleeding. Pooled event rates for the safety and efficacy outcomes in the warfarin arm were calculated using standard methods.

The meta-analysis included 8 trials: ROCKET-AF, ARISTOTLE, Amadeus, ACTIVE W, RE-LY, SPORTIF III, BAFTA, and SPORTIF V. Alternative therapeutic agents in the comparison groups consisted of ximelagatran, idraparinux, aspirin, aspirin plus clopidogrel, dabigatran, rivaroxaban, and apixaban.

Results showed the rate of stroke or non-CNS embolism varied from 1.2% to 2.3% per year, with a pooled event rate per year of 1.66% (95% confidence interval [CI], 1.41%-1.91%). Rates of MI, all-cause mortality, and composite outcomes ranged from 0.53% to 1.4%, 2.21% to 8%, and 3.93% to 5.90% per year, respectively. Pooled event rates were calculated to be 0.76% (95% CI, 0.57%-0.96%), 3.83% (95% CI, 3.07%-4.58%), and 4.80% (95% CI, 4.22%-5.38%) per year, respectively.

The researchers found episodes of major bleeding ranged from 1.40% to 3.40% per year, but because the definitions of major and minor bleeding were not consistent across studies, a pooled estimate could not be determined. The annual rate of intracranial hemorrhage varied from 0.33% to 0.80% per year, with a pooled event rate of 0.61% per year (95% CI, 0.48%-0.73%). The cumulative adverse event rate (major vascular events reported, death, or major bleeding) ranged from 3.00% per year (BAFTA) to 7.64% per year (RE-LY).

Results also showed patients who were ≥75 years of age had a significantly higher incidence of stroke or non-CNS embolism (2.27% per year) compared with patients <75 years of age (1.62% per year; P<.001). Women also experienced a significantly higher pooled incidence of stroke or non-CNS embolism compared with men (P <.001), as did patients with a history of stroke or transient ischemic attack compared with those who did not have any history of cerebrovascular events (P =.001).

The relative risk of stroke or non-CNS embolism was significantly greater among patients having an intermediate-risk CHADS₂ score of 2 (relative risk, 1.46; 95% CI, 1.13-1.89; P =.004) and high-risk CHADS₂ score of ≥3 (relative risk, 2.89; 95% CI, 2.28-3.66; P <.001), compared with patients having the lowest CHADS₂ score (≤1).

The researchers also found that in 7 of the 8 trials, patients were therapeutically anticoagulated (INR of 2.0 to 3.0) more than 60% of the time, an improvement over an earlier meta-analysis.