BACKGROUND: Fibrin D-dimer levels have been advocated as an useful clinical marker of thrombogenesis. HYPOTHESIS: We hypothesized that i) there is a hyperclotting state after the return of atrial fibrillation to sinus rhythm, ii) the measurement of plasma D-Dimer levels might be a good screening tool of this clotting status, and iii) the duration of arrhythmia influences the haemostasis measured by plasma D-Dimer levels. METHODS: Forty-two patients with atrial fibrillation undergoing cardioversion were divided into two groups: in Group A (n = 24,14 male, 56 +/- 11 years) the duration of atrial fibrillation was 72 hours or more (142.7 +/- 103.8 hours), in Group B (n = 18, 10 male, 61 +/- 13 years) the duration of atrial fibrillation was less than 72 hours (25 +/- 16 hours). Plasma fibrin D-dimer levels were measured by enzyme immunoassay before, and 36 hours after, cardioversion. The change of plasma D-dimer levels 36 hours after cardioversion was calculated as delta-D-dimer. RESULTS: There were no significant differences in demographic, clinical, and echocardiographic data, and the success of cardioversion between the two groups. Compared to the control, the baseline D-dimer levels were significantly higher in both groups. The delta D-dimer levels were significantly higher in Group A than in Group B (p < 0.005). Furthermore, plasma D-dimer levels 36 hours after cardioversion (r = 0.52, p = 0.0016) and delta-D-dimer levels (r = 0.73, p < 0.0001) showed significant correlations with the duration of atrial fibrillation. CONCLUSION: The longer duration of the atrial fibrillation episode could lead to a more prominent cardiovascular hyperclotting state after cardioversion, and the mean changes of plasma D-Dimer levels could be used as an useful clinical marker of the clotting state after atrial systole return.