A simple method to identify patients on long-term warfarin who may derive the most benefit from new oral anticoagulants

In many countries, new oral anticoagulants are only covered for patients with suboptimal anticoagulation control on vitamin K antagonists (VKAs). The quality of VKA management is often reported using the time in therapeutic range (TTR). We sought to predict a TTR 65% or less using a surrogate measure [number of changes in VKA dose and number of international normalized ratio (INR) tests] that could be easily determined by primary care physicians. This cross-sectional study included consecutive patients whose VKA therapy was managed in a specialized anticoagulation clinic. Patients were dichotomized according to their TTR in the past 6 months (TTR > or ≤ 65%). The ability of the number of INR tests and VKA dose changes to predict TTR group was assessed using receiver-operating characteristics (ROC) curve analysis. The analyses included 1381 patients with a median age of 63 years. The mean TTR was 81% (interquartile range 70-90) and 17.4% of patients had a TTR 65% or more. Based on the ROC curve, patients were stratified according to whether they had either 3 or more dose changes or 9 or more INR tests within the last 6 months. The sensitivity to identify patients with TTR 65% or less was 87% and the specificity was 63%. The number of dose changes and the number of INR tests might be used as indicators of TTR; they could offer a simple way for clinicians to identify patients who are good candidates for the new oral anticoagulants. However, external validation studies in different clinical settings are needed to confirm these findings.