Heart failure (HF) is one of the most common and leading cause of death worldwide. Clinical trials provide evidence that the development of atrial fibrillation (AF) is a marker of poor prognosis in patients with HF. Furthermore, elevated D-dimer level is associated with increased cardiovascular mortality independent of AF in HF patients. We investigated whether plasma D-dimer levels in patients with hospitalized systolic HF could predict development of AF. A total of 150 consecutive patients with sinus rhythm who admitted to the emergency department with hospitalized systolic HF were evaluated. All hospitalized patients were obtained D-dimer levels within the first 24 h following admission. Atrial fibrillation developed in 31 (20.7%) patients during follow-up period of 6.3 +/- A 5 months. Patients who developed atrial fibrillation had significantly increased levels of D-dimer [608 (339-1,022) ng/ml versus 1,100 (608-2,599) ng/ml, P = 0.001]. Optimal cut-off level of D-dimer to predict development of AF was found to be > 792 ng/ml. D-dimer > 792 ng/ml, right ventricular dilatation, age, systolic pulmonary pressure, left atrium size, moderate to severe tricuspid regurgitation, and beta blocker usage were found to have prognostic significance in univariate analysis. In multivariate Cox proportional-hazards model, D-dimer levels > 792 ng/ml (HR = 3.019, P = 0.006), and right ventricular dilatation (HR = 8.676, P = 0.003) were associated with an increased risk of new-onset AF. In conclusion, D-dimer could predict development of AF in patients with hospitalized systolic HF.