Resumption of Oral Anticoagulation After Warfarin-Associated Intracerebral Hemorrhage
A hematology/thrombosis consultation for a patient admitted with warfarin-associated intracerebral hemorrhage (ICH) should take into account the indication for anticoagulation, the treatment history with warfarin, any possible precipitating risk factors for the hemorrhage, the location of the hematoma, and the condition of the patient. Finally, the balance between the risk for ischemic stroke without anticoagulation and the risk of recurrent ICH with resumed anticoagulation should be assessed.
Until recently, expert opinion1 as well as systematic reviews2,3 were favoring early resumption, that is, after 3 to 14 days, of anticoagulation in stable patients. A retrospective study of 234 patients with warfarin-associated ICH concluded that the resumption should be delayed by 10 to 30 weeks to avoid the early high-risk period for recurrent hemorrhage, during which full anticoagulation appeared to increase this risk 5-fold.4 In risk modeling, it could be calculated from the data in this study that resumption of warfarin after 1 month would confer a 16% risk of recurrent ICH during the first year. Eckman et al used a decision analysis to evaluate whether warfarin should be restarted and they based this partly on data from a prospective cohort of 435 patients with intracerebral hemorrhage,5 not specifically warfarin-associated. From that material the authors calculated a 1-year risk of recurrence of 15% after lobar ICH. For deep hemispheric hemorrhage, they calculated a corresponding risk of only 2.1% per year.
In another retrospective review of 207 patients surviving an ICH, Viswanathan et al similarly found that the 2-year risk of recurrence was 22% after lobar and 4% after deep hemispheric hemorrhage.6 Unfortunately, lobar hemorrhage appears to be approximately 50% more common than deep hemispheric hemorrhage among the survivors.4,6 As for subdural hematoma, there was a trend to higher risk of recurrence compared with ICH in our study,4 although the long-term mortality was lower.
The 1-year risk of recurrent ICH of 15% to 16% is higher than the risk of ischemic stroke without warfarin in patients with mechanical valves as well as in patients with atrial fibrillation, except for those with the maximum CHADS2 score of 6.7 Provided that the burden of a recurrent intracranial hemorrhage is considered similar to that of an ischemic stroke, then we should not resume warfarin for the patients with lobar ICH and possibly also not for those with subdural hematoma. In other words, for the majority of patients with ICH, there seems to be an unfavorable risk/benefit ratio of resuming anticoagulation, at least for the first year. After that, our calculations become fraught with increasing uncertainty.
This reasoning is frustrating for many cardiologists, who are trained to do the most to protect their patients from thromboembolic complications. In my own experience, it results in repeated requests to assess the patient for resumption of anticoagulation. In the current case with a CHADS2 score of 3 and an annual risk of ischemic stroke of approximately 6%, a risk of recurrent intracranial bleeding that is more than twice as high should be convincing to abstain from resumption of warfarin. Moreover, this patient had the bleeding, whereas the international normalized ratio was within the therapeutic range and presumably with well-controlled blood pressure, so there is not much improvement in the management to be expected.
As a less effective stroke prophylaxis than warfarin but still better than no antithrombotic therapy, aspirin should be considered. Antiplatelet therapy after ICH was not associated with an increased risk of recurrent bleeding,6 although in this study, antiplatelet therapy was started late at a mean of 5 months after the event.
Finally, the question of prophylaxis against venous thromboembolism should be discussed. The risk of venous thromboembolism is high in an immobilized patient after any type of stroke and low-molecular-weight heparin at a prophylactic dose, started after 48 hours in patients with intracerebral hemorrhage appears to be safe.8
In summary, for this patient, warfarin should most likely not be restarted. However, if he belongs to the minority with deep hemispheric hemorrhage, resumption could be considered, although it is worrisome if no triggering factor for the index bleeding was identified. The timing of resumption should in that case be between 10 and 30 weeks after the hemorrhage to minimize the overall risk of recurrent hemorrhage or ischemic stroke.
The author has received consulting fees from Bayer HealthCare, Boehringer Ingelheim, GlaxoSmith-Kline, Octapharma, and Sanofi-Aventis; lecture fees from Sanofi-Aventis, and Boehringer Ingelheim; and grant support from Bayer HealthCare.
The opinions expressed in this article are not necessarily those of the editors or of the American Heart Association. This article is Part 2 of a 3-part article. Parts 1 and 3 appear on pages 3661 and 3665, respectively.
- Received June 14, 2011.
- Accepted August 19, 2011.
- © 2011 American Heart Association, Inc.
- Majeed A,
- Kim YK,
- Roberts RS,
- Holmstrom M,
- Schulman S
- Eckman MH,
- Rosand J,
- Knudsen KA,
- Singer DE,
- Greenberg SM
- Viswanathan A,
- Rakich SM,
- Engel C,
- Snider R,
- Rosand J,
- Greenberg SM,
- et al