Quality of individual INR control and the risk of stroke and bleeding events in atrial fibrillation patients: A nested case control analysis of the ACTIVE W study

INTRODUCTION: Time in therapeutic range (TTR) for international normalized ratio (INR) is an accepted quality measure of anticoagulation control in patient populations, but its usefulness for predicting stroke and bleeding in individuals is not well understood. MATERIALS AND METHODS: In a nested case control analysis among ACTIVE W study patients, cases with stroke and cases with bleeding were separately matched with controls. Several anticoagulation quality measures were compared, overall and in a time-dependent manner. RESULTS: 32 cases with ischemic stroke and 234 cases with bleeding in the analysis were matched in a 4:1 ratio to 122 and 865 controls, respectively. Follow-up duration was 257±154days for the stroke analysis and 222±146days for the bleeding analysis. Compared with their respective controls, the study mean TTR of both stroke cases (53.9%±25.1 vs 63.4%±24.8; p=0.055) and bleeding cases (56.2%±25.4 vs 63.4%±26.8; p<0.001) was lower. Time below range for stroke and time above range for bleeding were only greater in the last month leading up to the event, not over the entire study period. Rather, over the entire study period bleeding cases spent more time below range than controls (26.8%±25.9 vs 20.8%±24.0; p=0.001). CONCLUSIONS: TTR was lower in individual AF patients with stroke or bleeding compared with matched controls in ACTIVE W. Maintaining a high TTR, with equal importance to avoid low and high INRs, is a relevant goal of individual patient treatment to prevent stroke and bleeding.