Effects of anticoagulation on infarct size and outcome have not been fully elucidated in patients with acute cardioembolic stroke, although the anticoagulation therapy reduces both occurrence and recurrence of ischemic stroke greatly. The authors retrospectively investigated the relationship of anticoagulation intensity to infarct size and outcome. In 104 consecutive patients (mean age 70.8 +/- 10.0 years) who had suffered acute supratentorial cardioembolic infarction or transient ischemic attacks, they analyzed risk factors for atherosclerosis, underlying heart diseases, the infarct size (maximal area) on brain computed tomography, and modified Rankin scale score upon discharge. They compared these clinical data between patients who had received warfarin before the ictus and those who had not. In addition, they investigated the effects of the international normalized ratio (INR) on infarct size and outcome in 19 patients who had been receiving anticoagulant therapy and had measurement of INR within 24 hours after stroke onset. There were 25 patients who had received anticoagulation before the stroke (A/C group) and 79 patients who had not (non-A/C group). The infarct size in the A/C group tended to be smaller than that in the non-A/C group (p = 0.081, Mann-Whitney U test). In the 19 patients who had prior anticoagulation and measurement of INR within 24 hours of stroke onset, large infarcts were seen in 6 of 13 patients with INR < 1.6 and in none of 6 patients with INR > or = 1.6. Poor clinical outcome was observed in 5 patients with INR < 1.6, but in none with INR > or = 1.6. In conclusion, anticoagulant therapy with INR > or = 1.6 appears to effectively prevent a large infarct and poor outcome, even when ischemic stroke dose occurs in patients with an emboligenic heart disease.