BackgroundIn patients with known atrial fibrillation (AF) different scores are utilized to estimate the risk of thromboembolic events and guide oral anticoagulation. Diagnosis of AF strongly depends on the duration of electrocardiogram monitoring. The aim of this study was to use established scores to predict the prevalence of AF.MethodsThe CHADS2- (Congestive Heart failure, hypertension, Age >75 years, Diabetes, Stroke [doubled]) and CHA2DS2VASc-score (Congestive Heart failure, hypertension, Age ≥75 years [doubled], Diabetes, Stroke [doubled], Vascular disease, Age 65–74 years, Sex category [female sex]) was calculated in 150,408 consecutive patients, referred to the University Hospital of Rostock between 2007 and 2012. All factors constituting these scores and a history of AF were prospectively documented with the ICD-10 admission codes.ResultsMean age of our study population was 67.6 ± 13.6 years with a mean CHADS2-score of 1.65 ± 0.92 and CHA2DS2VASc-score of 3.04 ± 1.42. AF was prevalent in 15.9% of the participants. The prevalence of AF increased significantly with every CHADS2- and CHA2DS2VASc-score point up to 54.2% in CHADS2-score of 6 and 71.4% in CHA2DS2VASc-score of 9 (P < 0.001).ConclusionThe prevalence of AF increases with increasing CHADS2- and CHA2DS2VASc-score. In intermediate scores intensified monitoring may be recommended. In high scores, thromboembolic complications occurred irrespective of the presence of AF and anticoagulant therapy may be initiated irrespective of documented AF.