Objective: Adjusted-dose warfarin therapy can prevent stroke in patients with atrial fibrillation. However, the quality of the warfarin control may be considered to be important for elderly patients. methods: We followed 188 patients (age > or =70 years) with atrial fibrillation (warfarin, 120 patients; non-warfarin, 68 patients) for 2 years. Their warfarin control was assessed by time in therapeutic range (TTR) for an international normalized ratio of prothrombin time of 1.6-2.6, based on the Japanese guidelines of anticoagulation for elderly patients with atrial fibrillation.
Results: Stroke occurred in 23 patients (12.2%). In warfarin-treated patients, receiver-operator characteristic (ROC) curves suggested that patients with TTR >68% had anticoagulation benefit. In the ROC curves for prediction of stroke, the area under the curve of TTR was 0.709 (95% confidence interval, 0.585 to 0.834; p=0.02). The sensitivity and specificity of TTR < or =68% were 91.7% and 54.0%, respectively. Kaplan-Meier curves showed that the event-free ratio of stroke was significantly higher in patients who achieved this cut-off of TTR.
Conclusion: The results suggest that the quality of warfarin control is directly associated with the incidence of stroke in elderly patients.