Variability of INR and its relationship with mortality, stroke, bleeding and hospitalisations in patients with atrial fibrillation

Background - rationale for study: Atrial fibrillation is associated with an increased risk of stroke and mortality which is reduced by treatment with Warfarin. The most commonly used tool to assess the effectiveness of warfarin therapy is the time in therapeutic Range (TTR) of International Normalised Ratio (INR) 2.0-3.0. Our aim was to study whether INR variability, as assessed by the standard deviation of transformed INR (SDTINR) is more prognostically important than the TTR. Methods and Results: We studied 19,180 patients with atrial fibrillation on warfarin therapy to evaluate the association of TTR and that of SDTINR with all-cause mortality, stroke, bleeding and hospitalisation. The SDTINR was more prognostically important than the TTR. One standard deviation (SD) higher of SDTINR had a hazard ratio (HR) of 1.59 (95% CI 1.52-1.66) of mortality compared with 1.18 (95% CI 1.13-1.24) for one SD lower of TTR. For the other 3 events the HR was also higher for the SDTINR than for the TTR (stroke 1.30 (95% CI 1.22-1.39) vs. 1.06 (95% CI 1.00-1.13), bleeding 1.27 (95% CI 1.20-1.35) vs. 1.07 (95% CI 1.01-1.14), hospitalisation 1.47 (95% CI 1.45-1.49) vs. 1.13 (95% CI 1.10-1.15). When both metrics were included in the same analysis only the SDTINR was of significant predictive value. Conclusions: The SDTINR is a better predictor of mortality, stroke, bleeding and hospitalisation than the TTR in patients with atrial fibrillation receiving warfarin therapy. (C) 2011 Elsevier Ltd. All rights reserved.