New meta-analysis updates benefits of antithrombotics in AF

by Sue Hughes

San Antonio, TX - For patients with atrial fibrillation, taking adjusted-dose warfarin reduces the risk of stroke by approximately 60%, and antiplatelet agents reduce the risk by about 20%, a new meta-analysis estimates.

The authors conclude: "These results reaffirm recommendations for adjusted-dose warfarin for patients with atrial fibrillation who are deemed to be at high risk for stroke, with antiplatelet therapy for low-risk patients and for those who cannot safely receive warfarin." But they add that a substantial unmet need remains for antithrombotic agents that are more efficacious than aspirin and that are safer or more easily administered than adjusted-dose warfarin.

The meta-analysis, published in the June 19, 2007, issue of the Annals of Internal Medicine, was conducted by a team led by Dr Robert Hart (University of Texas Health Science Center, San Antonio). They note that nonvalvular atrial fibrillation is an important cause of disabling stroke, the incidence of which can be reduced by using antithrombotic prophylaxis. They previously conducted a meta-analysis of 16 randomized clinical trials testing antithrombotic therapies for stroke prevention in approximately 10 000 participants. But since publication of that meta-analysis, another 13 randomized trials in 18 140 additional patients have been reported. They therefore conducted a new meta-analysis of all currently available randomized trials.

They included 29 trials with a total of 28 044 participants (mean age 71 years; mean follow-up 1.5 years). Results showed that adjusted-dose warfarin is the most effective antithrombotic therapy at preventing stroke, being around 40% more efficacious than antiplatelet therapy.

Main results of meta-analysis of antithrombotic therapy in AF patients

Comparison	Trials (n)	Patients (n)	Reduction in stroke (%)	95% CI (%)
Warfarin vs control	6	2900	64	49-74
Antiplatelet agents vs control	8	4876	22	6-35
Warfarin vs antiplatelet agents	12	12 963	39	22-52

Other randomized comparisons were inconclusive. Absolute increases in major extracranial hemorrhage were small (less than 0.3% per year) and were less than the absolute reductions in stroke.

Hart et al note that they have confirmed the main findings of their previous meta-analysis but have improved the precision of the estimates of benefits and risks, because of the addition of 13 trials and the more than doubling of the total number of participants and strokes. "Major new evidence includes comparisons of adjusted-dose warfarin with the combination of clopidogrel plus aspirin (one trial with 6706 participants) and a more precise estimate of anticoagulation therapy compared with all antiplatelet therapies (six additional trials with 8911 participants)."

What is the best INR?

They note that the substantial relative risk reduction in stroke with adjusted-dose warfarin is consistent for all patient subgroups over a wide range of absolute stroke risks. Achieved INRs in randomized trials ranged from 2.0 to 2.9, and while direct randomized comparisons of different intensities of adjusted-dose warfarin are limited, the authors say that achieved INRs between 2.0 and 2.5 offer large relative risk reductions for primary stroke prevention and that the absolute increase in intracerebral hemorrhage during anticoagulation therapy remains relatively small if the INR is 3.5 or less.

They point out that the estimate of a 20% reduction in stroke with antiplatelet agents is less consistent among trials than the estimate for adjusted-dose warfarin, and another recent analysis suggested a larger estimate of the efficacy of aspirin, with a 35% risk reduction, but that aspirin seems to have a larger effect on nondisabling, noncardioembolic strokes than on cardioembolic infarctions.

Any role for clopidogrel?

Hart et al further report that the relative benefit of any specific antiplatelet agent or combinations of antiplatelet agents is unclear from the available data and that there is no evidence favoring any one dosage of aspirin over another. They add that the point estimate of the relative risk reduction with adjusted-dose warfarin was similar when compared with clopidogrel plus aspirin and with pooled results from eight trials of aspirin alone, suggesting that there is no major benefit of adding clopidogrel to aspirin, but these antiplatelet agents are being directly compared in the ongoing ACTIVE-A trial.

They further report that the two available randomized trials that added antiplatelet therapy to full-dose anticoagulant therapy were too small and had too few events to adequately characterize the efficacy and safety of such therapy.

Hart et al say the finding of only small increases in major extracranial hemorrhage during anticoagulation is reassuring, but the design of recent randomized trials may result in more favorable estimates of bleeding than those that can be anticipated in clinical practice for warfarin-naive patients and for older patients. However, they add that recent, large, practice- and population-based studies report reassuringly similar results.

But warfarin not best for everyone

The authors conclude by pointing out that the impressive relative risk reduction in stroke with adjusted-dose warfarin does not imply clinically important benefits for all patients who have atrial fibrillation. Noting that the intrinsic risk for stroke among such patients varies 20-fold, they advise that many low-risk patients do not benefit substantially from warfarin, whereas high-risk patients (particularly those with previous stroke or transient ischemic attack) have large absolute reductions in stroke rate with anticoagulation therapy.

They add that the choice of antithrombotic prophylaxis is best individualized and should consider the patient's inherent stroke risk and the best estimate of the absolute benefits, as well as bleeding risk, access to high-quality anticoagulation monitoring, and patient preferences.