Stroke Survivors With Afib Could Start DOACs Earlier

It was safe to start direct oral anticoagulant (DOAC) therapy without delay after acute ischemic stroke in people with atrial fibrillation (Afib), according to the OPTIMAS randomized trial.

Stroke survivors randomized to early or delayed DOAC initiation had the same 3.3% incidence of composite recurrent ischemic stroke, symptomatic intracranial hemorrhage, unclassifiable stroke, or systemic embolism incidence at 90 days (P=0.0003 for noninferiority), reported a group led by David Werring, PhD, of University College London in The Lancet.

Additionally, there was a “very low” incidence of symptomatic intracranial hemorrhage either way (0.6% early vs 0.7% delayed, P=0.78) — and the risk did not vary by stroke severity.

Consistent with other trials of early DOAC initiation, Werring’s group concluded OPTIMAS thus provides “reassurance for the safety of early anticoagulation with a DOAC and do not support the common current guideline-supported practice of delaying oral anticoagulation after acute ischemic stroke with atrial fibrillation for up to 14 days after moderate-to-severe acute stroke.”

“Early DOAC initiation also has the potential practical advantage of improving the proportion of patients who start secondary prevention treatment before hospital discharge, although this is not shown by our data and should be investigated in further studies,” study authors added.

In the absence of high-quality evidence, the current recommendations to wait on DOAC initiation stem from concerns about the risk of early intracranial hemorrhage in this setting. Yet evidence has been accruing, in recent trials like ELAN, that an earlier start to anticoagulation is safe and may in fact improve outcomes in people with Afib after a stroke.

OPTIMAS builds on this literature despite not finding evidence of superiority with early DOAC initiation (P=0.96 for superiority). The finding of noninferiority was nevertheless unchanged when the authors accounted for the competing risk of mortality (8.8% for early initiation vs 8.9% for delayed initiation).

Notably, unlike ELAN, OPTIMAS investigators considered early DOAC initiation to be within 4 days of stroke onset (vs 48 hours) and late initiation to be within 7-14 days (vs day 6-7).

Werring’s group also highlighted the relative broadness of the OPTIMAS population as it included people with severe stroke (for a trial-wide median NIH Stroke Scale score of 5) and those already taking an anticoagulant at the time of their stroke (32.2%).

The trial is ongoing, the authors noting that they are planning to conduct additional brain imaging analyses to see if there are any variables (e.g., hemorrhagic transformation, infarct volume, cerebral small vessel disease markers) that may alter the timing of DOAC therapy after ischemic stroke.

OPTIMAS was a multicenter open-label trial conducted at 100 U.K. hospitals from 2019 to 2024.

Participants were survivors of acute ischemic strokes likely related to their Afib. In the hospital stroke unit, they were randomized 1:1 to early or later DOAC initiation. Patients were not eligible if they had a coagulopathy or a high bleeding risk.

Ultimately, there were 3,621 patients (mean age 78.5 years, 45.3% women, 93.7% white) in the modified intention-to-treat analysis. Baseline characteristics were well balanced between groups, according to Werring and colleagues.

Patients had had their strokes treated with IV thrombolysis in 22.0% of cases, and endovascular therapy in 7.3%. After the stroke, 83.8% were put on antiplatelet therapy.

The DOAC initiated for the study was most commonly apixaban (Eliquis; 62.1%), followed by edoxaban (Savaysa; 28.9%).

The early DOAC group initiated on average 3.1 days after stroke onset; the later group initiated at 8.3 days.

Werring’s team acknowledged that one of the trial’s limitations was the lack of evaluation of DOAC initiation in the 4-7 days after stroke.

There were also few OPTIMAS participants with very severe strokes who actually made it to the trial, while the group of patients with parenchymal hematoma type 2 were excluded outright.