Background: The outcome of atrial fibrillation patients on warfarin partially depends on maintaining adequate time in therapeutic International Normalized Ratio range (TTR). Large differences in TTR have been reported between centers and countries. The association between warfarin dosing practice, TTR, and clinical outcomes was evaluated in Randomized Evaluation of Long-term Anticoagulation Therapy (RE-LY) trial patients receiving warfarin.
Methods and results: RE-LY provided an algorithm for warfarin dosing, recommending no change for in-range, and 10% to 15% weekly dose changes for out-of-range International Normalized Ratio values. We determined whether dose adjustments were consistent with algorithm recommendations but could not verify whether providers used the algorithm. Using multilevel regression models to adjust for patient, center, and country characteristics, we assessed whether algorithm-consistent warfarin dosing could predict patient TTR and the composite outcome of stroke, systemic embolism, or major hemorrhage. We included 6022 nonvalvular atrial fibrillation patients from 912 centers in 44 countries. We found a strong association between the proportion of algorithm-consistent warfarin doses and mean country TTR (R(2)=0.65). The degree of algorithm-consistency accounted for 87% of the between-center and 55% of the between-country TTR variation. Each 10% increase in center algorithm-consistent dosing independently predicted a 6.12% increase in TTR (95% confidence interval, 5.65-6.59) and an 8% decrease in rate of the composite clinical outcome (hazard ratio, 0.92; 95% confidence interval, 0.85-1.00).
Conclusions: Adherence, intentional or not, to a simple warfarin dosing algorithm predicts improved TTR and accounts for considerable TTR variation between centers and countries. Systems facilitating algorithm-based warfarin dosing could optimize anticoagulation quality and improve clinical outcomes in atrial fibrillation on a global scale.
Clinical trial registration: URL: http://www.clinicaltrials.gov. Unique identifier: NCT00262600.