CardioExchange Archive

A very large new meta-analysis finds a favorable risk-benefit for the new oral anticoagulant drugs in the setting of atrial fibrillation. The findings, published online in the Lancet, were remarkably consistent for all four of the new agents which have been fighting to replace warfarin, which was the only oral anticoagulant available for decades until the arrival of the new agents. Although warfarin is inexpensive, it has numerous interactions with other drugs and foods and requires regular monitoring and dose adjustments. The new agents can be taken once or twice a day and do not require dose changes.

Christian Ruff and colleagues combined data from the nearly 72,000 patients randomized in the four large mega-trials: RE-LY, which studied dabigatran (Pradaxa, Boehringer-Ingelheim); ROCKET AF, which studied rivaroxaban (Xarelto, Johnson & Johnson); ARISTOTLE, which studied apixaban (Eliquis, Pfizer and BristolMyers Squibb); and ENGAGE-AF-TIMI 48, which studied edoxaban (Daiichi Sankyo).

Compared to warfarin, the new agents were associated with a 19% reduction in stroke or systemic embolic events. The investigators said this finding was “mainly driven” by a significant 51% reduction in hemorrhagic stroke. In addition, they found a 10% reduction in mortality and a 52% reduction in intracranial hemorrhage. However, gastrointestinal bleeding was increased by 25%. There were only small, statistically insignificant reductions in myocardial infarction and ischemic stroke. Here are the relative risks and 95% confidence intervals:

• Stroke or systemic embolic events: 0.81 (0.73–0.91, p<0·0001)
• –Hemorrhagic stroke: 0.49 (0.38–0.64, p<0·0001)
• –Ischemic stroke: 0.92 (0.83-1.02, p=0.10)
• Mortality: 0·90 (0·85–0·95, p=0·0003)
• Intracranial hemorrhage: 0·48 (0·39–0·59, p<0·0001)
• Gastrointestinal bleeding: 1·25 (1·01–1·55, p=0·04)
• Myocardial infarction: 0.97 (0.78-1.20, p=0.77)

The results were consistent across all subgroups analyzed.The new agents were equally effective both in patients who had previously received warfarin and in patients who had not.

In an accompanying comment, Torben Bjerregaard Larsen and Gregory Lip point out that the paper assumes that there is a class effect for the new drugs or that they are “broadly equivalent,” though this has not been tested. The paper also “does not really answer the question of which novel oral anticoagulant is best, whether from an efficacy or safety perspective.” Warfarin, they write, is still a good choice for patients who are likely to achieve a high time in the therapeutic range, “because the main benefits of novel oral anticoagulants compared with warfarin might be only marginal in those with high times in therapeutic range, although the reduction in intracranial hemorrhage is still evident.”