Study objective To evaluate the effect of anticoagulant therapy on the survival of patients with idiopathic pulmonary fibrosis (IPF). Design Prospective study. Setting Five hospitals located in the Miyagi prefecture in Japan, including a university hospital, a Red Cross hospital, two public general hospitals, and a municipal hospital. Patients Fifty-six patients with IPF (mean age, 69.4 years; range, 47 to 89) admitted to the hospitals from April 2001 to April 2004. Interventions Patients were assigned to receive prednisolone alone or prednisolone plus anticoagulant therapy. The anticoagulants included oral warfarin in an outpatient setting and low-molecular-weight heparin for rehospitalized patients with severely progressive respiratory failure. Measurements and results There was no difference in baseline characteristics, including age, gender, clinical condition, pulmonary function, and plasma d-dimer level between the nonanticoagulant group and the anticoagulant group. The overall survival and hospitalization-free periods were assessed. There was a significant difference between survival curves of the nonanticoagulant group and the anticoagulant group, with a 2.9 hazard ratio (p = 0.04, Cox regression model). There was no significant difference in the probability of a hospitalization-free period between groups. The major cause of clinical deterioration was acute exacerbation during follow-up in the present study. Therefore, the mortality and plasma d-dimer levels in patients with an acute exacerbation were also assessed. The mortality associated with acute exacerbations of IPF in the anticoagulant group was significantly reduced compared to that in the nonanticoagulant group (18% vs 71%, respectively; p = 0.008, Fisher Exact Test). Furthermore, the plasma d-dimer levels in patients who died were significantly higher than those in survivors during acute exacerbation of IPF (3.3 ± 2.3 μg/mL vs 0.9 ± 0.7 μg/mL, p < 0.0001). Histologic analysis performed in three patients who died due to an exacerbation of IPF in the nonanticoagulant group demonstrated the features of usual interstitial pneumonia and acute lung injury. Conclusions Our data suggested that plasma d-dimer levels are associated with mortality in patients with an acute exacerbation of IPF, and that anticoagulant therapy has a beneficial effect on survival in patients with IPF. To evaluate the effect of anticoagulant therapy on the survival of patients with idiopathic pulmonary fibrosis (IPF). Prospective study. Five hospitals located in the Miyagi prefecture in Japan, including a university hospital, a Red Cross hospital, two public general hospitals, and a municipal hospital. Fifty-six patients with IPF (mean age, 69.4 years; range, 47 to 89) admitted to the hospitals from April 2001 to April 2004. Patients were assigned to receive prednisolone alone or prednisolone plus anticoagulant therapy. The anticoagulants included oral warfarin in an outpatient setting and low-molecular-weight heparin for rehospitalized patients with severely progressive respiratory failure. There was no difference in baseline characteristics, including age, gender, clinical condition, pulmonary function, and plasma d-dimer level between the nonanticoagulant group and the anticoagulant group. The overall survival and hospitalization-free periods were assessed. There was a significant difference between survival curves of the nonanticoagulant group and the anticoagulant group, with a 2.9 hazard ratio (p = 0.04, Cox regression model). There was no significant difference in the probability of a hospitalization-free period between groups. The major cause of clinical deterioration was acute exacerbation during follow-up in the present study. Therefore, the mortality and plasma d-dimer levels in patients with an acute exacerbation were also assessed. The mortality associated with acute exacerbations of IPF in the anticoagulant group was significantly reduced compared to that in the nonanticoagulant group (18% vs 71%, respectively; p = 0.008, Fisher Exact Test). Furthermore, the plasma d-dimer levels in patients who died were significantly higher than those in survivors during acute exacerbation of IPF (3.3 ± 2.3 μg/mL vs 0.9 ± 0.7 μg/mL, p < 0.0001). Histologic analysis performed in three patients who died due to an exacerbation of IPF in the nonanticoagulant group demonstrated the features of usual interstitial pneumonia and acute lung injury. Our data suggested that plasma d-dimer levels are associated with mortality in patients with an acute exacerbation of IPF, and that anticoagulant therapy has a beneficial effect on survival in patients with IPF.