Dabigatran Cuts ICH Mortality vs Warfarin: RE-LY Analysis

April 25, 2012

April 24, 2012 (Dallas, Texas) — Among patients stricken with intracranial hemorrhage (ICH) while on oral anticoagulation for nonvalvular atrial fibrillation (AF), the stroke is no more likely to be fatal for those on dabigatran etexilate (Pradaxa, Boehringer Ingelheim) than for those taking warfarin, suggests an analysis from the Randomized Evaluation of Long-Term Anticoagulant Therapy (RE-LY) trial [1].

Given that RE-LY had already shown significantly reduced ICH rates with the new drug vs the very old one (p<0.001), the current findings suggest that when patients with AF receive dabigatran, "their overall risk of dying from ICH is 70% lower" than it is with warfarin, lead author Dr Robert G Hart (McMaster University, Hamilton, ON) observed for heartwire .

The analysis was published online in Stroke April 5, 2012. In the trial's primary finding, released in 2009 and reported by heartwire at the time, treatment with dabigatran at 150 mg twice daily cut the composite end point of stroke or peripheral embolic events by 34% per year compared with warfarin (p<0.001).

The trial had randomized >18 000 patients with AF (plus one other stroke risk factor) in 44 countries to receive dabigatran at either 110 mg or 150 mg two times daily or to warfarin adjusted to an INR of 2.0 to 3.0.

There were 154 instances of ICH in 153 patients, of which 46% were intracerebral bleeds and 45% were subdural hematomas. There was no significant difference in ICH-associated mortality between the three groups.

Intracranial Hemorrhage Outcomes in the RE-LY Trial (Mean Follow-Up Two Years)

End point Dabigatran 150 mg bid (n=37) Dabigatran 110 mg bid (n=27) Warfarin (n=90)
ICH (%/y) 0.31 0.23 0.76a
ICH mortality (% of ICH) 35 41 33
Subdural hematoma (%/y) 0.20 0.08b 0.31

a. p<0.001 for warfarin vs either dabigatran dosage

b. p=0.02 for dabigatran 110 mg bid vs 150 mg bid, p<0.001 for dabigatran 110 mg bid vs warfarin

In multivariate analysis, ICH predictors included warfarin as anticoagulant therapy (relative risk [RR] 2.9, p<0.001), using aspirin (which was at the physician's discretion) (RR 1.6, p=0.01), age (RR per added year 1.1, p<0.001), and history of stroke or transient ischemic attack (RR 1.8, p<0.001).

Hart cautions that the reduced ICH mortality with dabigatran may not extend to centers that are proficient in managing warfarin-related ICH: there is an antidote for the older drug, but none is currently available for dabigatran.

But most patients with warfarin-related ICH are not managed at such centers. "In real life, you'd have to go to a tertiary hospital," he said. "If you go to Massachusetts General Hospital or to the Mayo Clinic Rochester, where they're used to taking care of warfarin-associated ICH, well, there maybe you can get activated [prothrombin complex concentrates] PCCs. So maybe these results don't apply where vigorous aggressive treatment of warfarin hemorrhage is available." But at other centers and throughout most of the world, such "modern and expensive treatments for warfarin aren't available."

Hart also questioned the clinical importance of dabigatran's lack of an antidote. Although it's "logical" that reversal of warfarin with activated PCC would also reduce ICH mortality, there's no solid evidence that it does. "We still don't know if the outcome is changed."

RE-LY was funded by Boehringer Ingelheim, for which Hart discloses serving as a consultant. Disclosures for the coauthors are listed in the paper; in particular, coauthor Dr Paul A Reilly is an employee of Boehringer Ingelheim.