The Boston Area Anticoagulation Trial for Atrial Fibrillation (BAATAF) demonstrated that low-intensity warfarin anticoagulation can, with safety, sharply reduce the rate of stroke in patients with nonvalvular atrial fibrillation. The beneficial effect of warfarin was presumably related to a decrease in clot formation in the cardiac atria and subsequent embolization. To assess the effect of warfarin therapy on in vivo clotting in patients in the BAATAF, we measured the plasma level of prothrombin activation fragment F1+2. One sample was obtained from 125 patients from the BAATAF; 62 were taking warfarin and 63 were not taking warfarin (control group). The warfarin group had a 71% lower mean F1+2 level than the control group (mean F1+2 of 1.57 nmol/L in the control group compared with a mean of 0.46 nmol/L in the warfarin group; P < .001). F1+2 levels were higher in older subjects but were consistently lower in the warfarin group at all ages. Fifty-two percent of patients in the control group were taking chronic aspirin therapy at the time their F1+2 level was measured. Control patients taking aspirin had F1+2 levels very similar to control patients not taking aspirin (mean of 1.52 nmol/L for control patients on aspirin compared with 1.64 nmol/L for control patients off aspirin; P > .1). We conclude that prothrombin activation was significantly suppressed in vivo by warfarin but not aspirin among patients in the BAATAF. These findings correlate with the marked reduction in ischemic stroke noted among patients in the warfarin treatment group observed in the BAATAF.

Treatment with warfarin using a target International Normalized Ratio (INR) range of 1.7 to 2.5 is efficacious for many clinical indications, but the minimal intensity of anticoagulation required for antithrombotic protection has yet to be determined. To evaluate whether patients could be reliably monitored with a less intense regimen, we anticoagulated patients with warfarin for several months using a target INR range of 1.3 to 1.6 as determined by prothrombin time (PT) using a sensitive thromboplastin (Dade IS, International Sensitivity Index [ISI] = 1.3). Plasma measurements of F1+2, a marker of factor Xa action on prothrombin in vivo, were also obtained to determine the suppressive effect of warfarin on hemostatic system activity. Overall, 20 of 21 patients with a history of cerebrovascular events (mean age, 61 years) could be reliably regulated with warfarin in the target INR range. F1+2 levels were significantly suppressed from baseline in all patients, with a mean reduction of 49% (range, 28% to 78%). We found a significant relationship between the extent of suppression of prothrombin activation levels and the baseline measurements. A mean reduction of 65% was observed for those patients with baseline F1+2 greater than or equal to 1.5 nmol/L, but only 38% for baseline F1+2 less than or equal to 0.5 nmol/L. Overall, 68% of plasma samples obtained during stable anticoagulation were within the target INR range. PTs were also determined on all plasma samples with two thromboplastins of lower sensitivity (C+, ISI = 2.09; and automated simplastin, ISI = 2.10). Only 47% and 35% of PT determinations, respectively, were within the target range with these reagents. We conclude that prothrombin activation can be significantly suppressed in vivo with use of warfarin in an INR range of 1.3 to 1.6. This level of anticoagulation can be reliably achieved by monitoring PTs with a thromboplastin of high sensitivity.

Treatment with warfarin using a target International Normalized Ratio (INR) range of 1.7 to 2.5 is efficacious for many clinical indications, but the minimal intensity of anticoagulation required for antithrombotic protection has yet to be determined. To evaluate whether patients could be reliably monitored with a less intense regimen, we anticoagulated patients with warfarin for several months using a target INR range of 1.3 to 1.6 as determined by prothrombin time (PT) using a sensitive thromboplastin (Dade IS, International Sensitivity Index [ISI] = 1.3). Plasma measurements of F1+2, a marker of factor Xa action on prothrombin in vivo, were also obtained to determine the suppressive effect of warfarin on hemostatic system activity. Overall, 20 of 21 patients with a history of cerebrovascular events (mean age, 61 years) could be reliably regulated with warfarin in the target INR range. F1+2 levels were significantly suppressed from baseline in all patients, with a mean reduction of 49% (range, 28% to 78%). We found a significant relationship between the extent of suppression of prothrombin activation levels and the baseline measurements. A mean reduction of 65% was observed for those patients with baseline F1+2 greater than or equal to 1.5 nmol/L, but only 38% for baseline F1+2 less than or equal to 0.5 nmol/L. Overall, 68% of plasma samples obtained during stable anticoagulation were within the target INR range. PTs were also determined on all plasma samples with two thromboplastins of lower sensitivity (C+, ISI = 2.09; and automated simplastin, ISI = 2.10). Only 47% and 35% of PT determinations, respectively, were within the target range with these reagents. We conclude that prothrombin activation can be significantly suppressed in vivo with use of warfarin in an INR range of 1.3 to 1.6. This level of anticoagulation can be reliably achieved by monitoring PTs with a thromboplastin of high sensitivity.