This is the first evaluation of dabigatran, an oral direct thrombin inhibitor, in patients with atrial fibrillation (AF). Patients (n = 502) were randomized to receive blinded doses of 50-, 150-, or 300-mg dabigatran twice daily alone or combined with 81- or 325-mg aspirin or open-label warfarin administered to achieve an international normalized ratio of 2 to 3 for 12 weeks. Dabigatran plasma concentrations, activated partial thromboplastin time, d-dimer, urinary 11-dehydrothromboxane B₂ (DTB2), and liver function were measured at baseline and at 1, 2, 4, 8, and 12 weeks. Clinical end points were assessed according to the treatment received at the time of the event. Overall, 92% of patients completed the study. Major hemorrhages were limited to the group treated with 300-mg dabigatran plus aspirin (4 of 64), and the incidence was significant versus 300-mg dabigatran alone (0 of 105, p <0.02). Total bleeding events were more frequent in the 300-mg (39 of 169, 23%) and 150-mg (30 of 169, 18%) dabigatran groups compared with the 50-mg groups (7 of 107, 7%; p = 0.0002 and p = 0.01, respectively). Thromboembolic events were limited to the 50-mg dabigatran dose groups (2 of 107, 2%). The mean trough d-dimer measurements were suppressed for the 2 highest doses of dabigatran and warfarin (international normalized ratio of 2 to 3). Aminotransferase levels >3 times the upper limit of normal were observed in 0.9% of the dabigatran recipients and in none of the warfarin recipients. Two dabigatran recipients had aminotransferase levels >5 times the upper limit of normal as a result of gallstones, which resolved. Trough activated partial thromboplastin time values were 1.2, 1.5, and 1.8 times the baseline level for the 50-, 150-, and 300-mg dabigatran groups, respectively. DTB2 concentrations after 12 weeks of 50-, 150-, and 300-mg dabigatran treatment were increased by 31%, 17%, and 23%, respectively, versus baseline (p = 0.02, p = 0.03, and p = 0.0004). In conclusion, major bleeding events were limited to patients treated with dabigatran 300 mg plus aspirin and thromboembolic episodes were limited to the 50-mg dabigatran groups. The 2 highest doses of dabigatran suppress d-dimer concentrations. Serious liver toxicity was not seen. The significance of the increase of DTB2 concentrations in dabigatran-treated patients needs resolution.