Abstract TMP112: Left Atrial Enlargement is Associated With Anticoagulation Failure in Patients With Acute Ischemic Stroke and Atrial Fibrillation

• Published in: Stroke (1970) 2018-01, Vol.49 (Suppl_1)
• Main Authors: Dakay, Katarina, Chang, Andrew D, Hemendinger, Morgan, Cutting, Shawna M, McTaggart, Ryan A, Jayaraman, Mahesh V, Chu, Antony, Panda, Nikhil, Song, Christopher, Merkler, Alexander, Gialdini, Gino, Kummer, Benjamin, Lerario, Michael P, Kamel, Hooman, Elkind, Mitchell S, Furie, Karen L, Yaghi, Shadi
• Format: Article
• Language: English
• ISSN: 0039-2499; 1524-4628
• DOI: 10.1161/str.49.suppl_1.TMP112

Abstract only Background: Biomarkers to predict anticoagulation failure in patients with atrial fibrillation (AF) may help identify high-risk AF patients for further study. We hypothesize that left atrial enlargement (LAE) is more prevalent in AF patients admitted with ischemic stroke who are therapeutic, as opposed to non-therapeutic (NT), on anticoagulation. Methods: We included consecutive patients with AF admitted with a diagnosis of ischemic stroke between April 1 st , 2015 and December 31 st , 2016. Patients were divided into two groups based on whether they were therapeutic (warfarin with international normalized ratio ≥ 2.0 or non-vitamin K oral anticoagulant with uninterrupted use in the prior 2 weeks) vs. non-therapeutic on anticoagulation. We compared clinical and imaging characteristics, cardiac data (left atrial diameter on transthoracic echocardiogram), and outcomes (discharge modified Rankin Scale, discharge disposition, and symptomatic intracranial hemorrhage [sICH]) between the two groups. Multivariable logistic regression was used to estimate associations between therapeutic anticoagulation and variables, including CHADS 2 score and LAE (none/mild vs. moderate/severe). Results: We identified 225 patients; 52 (23.1%) were therapeutic on anticoagulation. The baseline demographics and risk factors were similar between the two groups. Patients therapeutic on anticoagulation were more likely to have a larger left atrial diameter (45.6 ± 9.2 vs. 42.3 ± 8.6 mm, p = 0.032), a higher CHADS2 score (2.9 ± 1.1 vs. 2.4 ± 1.1, p = 0.03), and sICH (13.5% vs. 3.5%, p = 0.013). After adjusting for the CHADS 2 score, patients who had stroke despite therapeutic anticoagulation were more likely to have moderate to severe LAE (odds ratio 2.05, 95% confidence interval 1.01-4.16). Results were unchanged when the CHA 2 DS 2 -VASc score was used. Conclusion: LAE is associated with anticoagulation failure in AF patients admitted with an ischemic stroke. This provides indirect evidence that LAE may portend failure of anticoagulation therapy in patients with AF; further studies are needed to delineate the significance of this association and improve stroke prevention strategies.