In patients with atrial fibrillation who have had an acute ischemic stroke, should direct oral anticoagulants be initiated earlier (0 to 4 days) or later (5 to 10 days) after the onset of symptoms?
For patients with atrial fibrillation and acute ischemic stroke who were not taking a DOAC at the time of the event, early initiation of a DOAC is at least as good as later initiation for patients without contraindications. Further study is needed in the subset of 121 patients who were undergoing thrombectomy, as those in the early initiation group had a nonsignificantly higher rate of the composite outcome (12.3% vs 3.6%). Studies regarding whether early initiation may be superior to late initiation are ongoing.
Plan de l'etude:
Randomized controlled trial (single-blinded)
Inpatient (any location)
The balance of benefits and harms for early versus late initiation of anticoagulation in patients with acute stroke is unclear, as the risk of both hemorrhage and recurrent ischemic stroke may be higher in the early phase (0 to 4 days). This Swedish study used a stroke registry to identify 888 patients with acute stroke (< 72 hours) and atrial fibrillation who had no contraindications to taking a direct oral anticoagulant (DOAC). Eligible patients had not been taking a DOAC prior to the stroke or had their DOAC therapy interrupted for 2 or more days prior to the stroke. At baseline, the patients' mean age was 78 years, 46% were women, the mean National Institutes of Health Stroke Scale score was 6 (moderate severity stroke), and approximately half had atrial fibrillation that was previously known. Groups were balanced at baseline and analysis was by intention to treat. Patients were randomized to receive early initiation of a DOAC at 0 to 4 days (mean initiation: day 3), or late initiation between 5 and 10 days (mean initiation: day 5). Physicians could choose the DOAC: 55% of patients received apixaban, 26% received dabigatran, and 13% received rivaroxaban. The primary composite of new ischemic stroke, symptomatic intracranial hemorrhage, or death occurred less often in the early start group (6.9% vs 8.7%) and, although noninferior, this difference was not significantly lower (superior). A numerical reduction was seen for both ischemic stroke and all-cause mortality, though these was also not statistically significant (there were no intracerebral hemorrhages in either group). An important omission is the lack of reporting of functional outcomes.
Mark H. Ebell, MD, MS
University of Georgia