Should I resume anticoagulation in patients following an intracranial hemorrhage?

In this systematic review of observational studies, restarting anticoagulants after an intracranial hemorrhage (ICH) was associated with a lower rate of thromboembolic events without much of an increase in recurrent hemorrhage. Since other factors might account for these associations, randomized trials would be beneficial but difficult to conduct. In the meantime, clinicians who face this issue should actively engage the patient or their surrogate in the decision-making process. (LOE = 2a)

Murthy SB, Gupta A, Merkler AE, et al. Restarting anticoagulant therapy after intracranial hemorrhage: a systematic review and meta-analysis. Stroke 2017;48(6):1594-1600.  [PMID:28416626]

Meta-analysis (other)

Government

Various (meta-analysis)

These authors searched several databases to identify studies that compared outcomes of patients with nontraumatic ICH who were restarted on anticoagulants with those who were not. Two authors independently assessed which papers to include and assessed the risk of bias for the included studies. The authors included 8 studies with 5306 patients who were followed up for a median of 12 months to 43 months. None of the studies masked the investigators to the anticoagulant status and only 3 studies tried to adjust for confounding factors. Since none of the studies were randomized trials, the authors were unable to overcome channeling effects (ie, where clinicians treat patients differentially based on patient characteristics that lead the clinician to believe one intervention might be safer or more effective than another in a specific patient). Six of the studies provided data on thromboembolic events: 4 only included patients with parenchymal ICH, while the other 2 included patients with subdural hematomas and subarachnoid hemorrhages. In the studies that evaluated resumption of anticoagulants, most of the patients were taking them for atrial fibrillation, prosthetic valves, and so forth, but a few were taking them because of a prior ischemic stroke or myocardial infarction (for whom antiplatelet therapy might have been a reasonable option). The anticoagulants were restarted within 3 months of the ICH in slightly more than one third of the patients. Thromboembolic events occurred in 7% of those who restarted anticoagulants compared with 18% of those who did not restart. Recurrence of the ICH occurred in 9% of those who restarted compared with 8% of those who did not restart. Although the authors found some statistical heterogeneity, they found no evidence for publication bias.