Impact of anticoagulation before stroke on stroke severity and long-term survival Karl Georg Haeusler 1,2 *, Maria Konieczny 1 , Matthias Endres 1,2 , Arno Villringer 3,4 , and Peter U. Heuschmann 2 Background Therapeutic anticoagulation by vitamin K antagonists is highly effective in reducing stroke risk in patients with atrial fibrillation. Vitamin K antagonist treat- ment before stroke reduces stroke severity and short-term mortality. Aims This study analyses vitamin K antagonists, used in patients with atrial fibrillation diagnosed before the index stroke. We also focus on the impact of preadmission anti- thrombotic medication on long-term survival. Methods We analyzed 2390 stroke patients consecutively admitted to the Department of Neurology, Charité Berlin, Germany between 2003 and 2004. Mean follow-up was 38 months (range 0–68). Using univariable and multivariable regression models, we identified factors for preadmission anticoagulation in patients with known atrial fibrillation and analyzed the impact of antithrombotic therapy preadmission on functional disability and long-term survival after stroke. Results Atrial fibrillation was diagnosed in 534 (22·3%) of the 2390 stroke patients. In 348 (65·2%) of all atrial fibrilla- tion patients, atrial fibrillation was already known before the index stroke. Three hundred twenty-five (93·4%) atrial fibrillation patients were amenable to anticoagulation, according to guidelines, 75 (23·1%) received vitamin K antagonists, and 20 (6·2%) had an international normalized ratio of 2–3 at the time of stroke onset. Males and younger patients were more likely to receive anticoagulation preadmission, while previous stroke had no significant impact on vitamin K antagonist prescription. Age (odds ratio 1·02 (95% confidence interval 1·00–1·04) per year), history of coronary artery disease (odds ratio 1·51 (95% confidence interval 1·01–2·26)), and therapeutic anticoagulation (odds ratio 0·28 (0·09–0·84)) were independent predictors of stroke severity. Age (hazard rates 3·11 (95% confidence interval 1·47–6·59), 4·65 (95% confidence interval 2·27–9·57), and 11·1 (95% confidence interval 4·90–25·1) for age categories 65–74, 75–84, and 85 years), preadmission antiplatelet therapy (hazard rate 1·85 (95% confidence interval 1·21– 2·82)), and stroke severity on admission (hazard rate 1·60 (95% confidence interval 1·03–2·46) and hazard rate 3·23 (95% confidence interval 1·88–5·55) for National Institutes of Health Stroke Scale categories 6–15 and >15 points) were associated with risk of death during follow-up. Conclusions In patients in which atrial fibrillation was diagnosed prior to the index stroke, about 23% received anticoagulation according to guideline recommendations. Therapeutic anticoagulation at stroke onset significantly decreased the risk of moderate to severe stroke on admis- sion but showed no significant association with long-term survival. Key words: anticoagulation, antithrombotic therapy, atrial fibrillation, ischemic stroke, long-term mortality, morbidity Introduction Atrial fibrillation (AF) is the most frequent cardiac arrhyth- mia, and its prevalence increases with age. As the population ages, it is anticipated that the number of AF patients will increase in coming years (1,2). Compared to patients without AF, those with AF have a fourfold to fivefold increased risk of ischemic stroke, almost independent of AF type (i.e. paro- xysmal, persistent, and permanent) (3–5). Patients with Correspondence: Karl Georg Haeusler*, Department of Neurology, Charité – University Medicine Berlin, Campus Benjamin Franklin, Hindenburgdamm, 30 D-12200 Berlin, Germany. Email: georg.haeusler@charite.de 1 Department of Neurology, Charité – University Medicine Berlin, Berlin, Germany 2 Center for Stroke Research, Charité, Berlin, Germany 3 Max-Planck Institute, University Hospital, Leipzig, Germany 4 Berlin School of Mind and Brain, Mind & Brain Institute, Humboldt-University, Berlin, Germany Conflict of interest: KGH received lecture fees by Sanofi-Aventis and Bayer Healthcare. ME has received grant support from AstraZeneca and Sanofi- Aventis, has participated in advisory board meetings of Boehringer Ingel- heim and Sanofi-Aventis, and has received honoraria from Novartis, Pfizer, EVER, Bayer Healthcare, AstraZeneca, Boehringer Ingelheim, Sanofi-Aventis, Trommsdorff, Berlin-Chemie, GlaxoSmithKline, and Bristol-Myers-Squibbs. AV reports lecture fees by Boehringer Ingelheim, Sanofi-Aventis, Ipsen, Bayer (Schering) and Berlin Chemie, and partici- pated in advisory board meetings of Boehringer Ingelheim and Bristol- Myers-Squibbs. Funding: The project has received funding from the Federal Ministry of Education and Research via the Competence Net Stroke and the grant Center for Stroke Research Berlin (01 EO 0801). ME receives support from the Volkswagen-Stiftung, Deutsche Forschungsgemeinschaft and the Federal Ministry of Education and Research. DOI: 10.1111/j.1747-4949.2011.00672.x Research © 2011 The Authors. International Journal of Stroke © 2011 World Stroke Organization Vol ••, September 2011, ••–•• 1