This study was undertaken to determine the ability of different markers extracted from single lead ECG recorded in sinus rhythm to identify patients prone to atrial fibrillation (AF). For this purpose, 5-minute ECGs recorded in sinus rhythm from two populations were compared: patients with a history of AF and healthy subjects without any history of AF. Several features based on P-waves and RR-intervals were extracted from the ECG. Among the extracted features, the most discriminative ones to identify the AF susceptibility were the P-wave duration, the standard deviation of the beat-to-beat Euclidean distance between successive P-waves and the sample entropy of the RR-intervals. The discriminative power of the aforementioned features was assessed using a classification tree approach. The results showed that the combination of P-wave duration, beat-to-beat Euclidean distance between P-waves and sample entropy could efficiently separate the two populations and therefore be used as an effective detection tool of patients at risk to develop AF.