icle Navigation EDITOR'S CHOICE Should we abandon the common practice of withholding oral anticoagulation in paroxysmal atrial fibrillation? Robby Nieuwlaat, Trang Dinh, S. Bertil Olsson, A. John Camm, Alessandro Capucci, Robert G. Tieleman, Gregory Y.H. Lip, Harry J.G.M. Crijns on behalf of the Euro Heart Survey Investigators European Heart Journal, Volume 29, Issue 7, April 2008, Pages 915–922, https://doi.org/10.1093/eurheartj/ehn101 Published: 10 March 2008 Article history Received: 31 July 2007 Revision received: 15 January 2008 Accepted: 16 January 2008 Published: 10 March 2008 A correction has been published: European Heart Journal, Volume 29, Issue 7, April 2008, Pages 840–842, https://doi.org/10.1093/eurheartj/ehm594 pdfPDF Views Cite Permissions Icon Permissions Share Abstract Aims To assess the relation between the atrial fibrillation (AF) subtype and thrombo-embolic events. Methods and results The observational Euro Heart Survey on AF (2003–04) enrolled 1509 paroxysmal, 1109 persistent, and 1515 permanent AF patients, according to the 2001 American College of Cardiology, American Heart Association, and the European Society of Cardiology guidelines definitions. A 1 year follow-up was performed. Permanent AF patients had at baseline a worse stroke risk profile than paroxysmal and persistent AF patients. In paroxysmal AF, the risk for stroke, any thrombo-embolism, major bleeding and the combined endpoint of cardiovascular mortality, any thrombo-embolism, and major bleeding was comparable with persistent and permanent AF, in both univariable and multivariable analyses. Compared with AF patients without stroke, patients suffering from a stroke had a comparable frequency and duration of AF attacks, but tended to have a worse stroke risk profile at baseline. During 1 year following cardioversion, paroxysmal AF patients had a higher risk for stroke (P = 0.029) and any thrombo-embolism (P = 0.001) than persistent AF patients. Conclusion In the Euro Heart Survey, paroxysmal AF had a comparable risk for thrombo-embolic events as persistent and permanent AF. This observation strengthens the guideline recommendation not to consider the clinical AF subtype when deciding on anticoagulation.