P4312Effect of prior treatment with non-vitamin K antagonist oral anticoagulants on severity and in-hospital outcome of acute ischemic stroke in patients with atrial fibrillation

Background/Introduction: Non-vitamin K antagonist oral anticoagulants (NOACs) are effective and safe alternatives to vitamin K antagonists (VKAs) in patients with atrial fibrillation (AF). However, the effects of prior treatment with NOACs on acute ischemic stroke (IS) severity and in-hospital outcome are unknown. Purpose: To compare the effects of prior treatment with NOACs, VKAs and antiplatelets on IS severity and in-hospital outcome in patients with AF. Methods: We prospectively studied 921 consecutive patients admitted with acute IS [42% males, median age (interquartile range) 80 (9) years]. In patients who were receiving acenocoumarol prior to stroke, the international normalized ratio (INR) was determined at the emergency department. The outcomes of the study were severity of IS at admission [assessed with the National Institutes of Health Stroke Scale (NIHSS)], dependency at discharge [modified Rankin Scale (mRS) between 2 and 5] and in-hospital mortality. Results: At admission, 291 patients had a history of AF and among them 66, 98, 17, 69, 22 and 13 were receiving no antithrombotic treatment, single antiplatelet treatment, dual antiplatelet treatment, acenocoumarol with INR<2, acenocoumarol with INR 2–3 and NOACs, respectively. The patients on acenocoumarol with INR>3 (n=6) were excluded from the analyses. The median NIHSS at admission did not differ between the 6 treatment groups [8.5 (13), 10 (16), 4 (13), 9 (11), 4 (12) and 7 (12), respectively; p=0.200]. Patients on NOACs had comparable rates of dependency with patients on no antithrombotic treatment, single antiplatelet treatment and acenocoumarol with INR<2 whereas patients on acenocoumarol with INR 2–3 and on dual antiplatelet treatment had a tendency for lower dependency rates at discharge (72.7, 70.8, 71.6, 73.7, 41.2 and 42.9%, respectively; p=0.056). Independent predictors of dependency at discharge were age, NIHSS at admission and weight. In-hospital mortality rates were numerically but non-significantly lower in patients on NOACS and on acenocoumarol with INR 2–3 than in patients on no antithrombotic treatment, on single or dual antiplatelet treatment and on acenocoumarol with INR<2 (7.7, 9.1, 21.2, 17.3, 11.8 and 13.0%, respectively; p=0.720). Independent predictors of in-hospital mortality were type 2 diabetes mellitus and NIHSS, diastolic blood pressure and serum glucose levels at admission.