Background: The efficacy and safety of warfarin for stroke prevention in atrial fibrillation (AF) depend on the time in the therapeutic range (TTR) with an international normalised ratio (INR) of 2.0-3.0. This meta-analysis focused the relative efficacy and safety of non-VKA oral anticoagulants (NOAC) compared with warfarin at different thresholds of centre's TTR (cTTR).
Methods: We searched PubMed, Embase, CENTRAL and websites of regulatory agencies, limiting searches to randomized phase 3 trials. Primary outcomes were stroke or systemic embolism (SSE) and major or non-major clinically relevant (NMCR) bleeding. We used a random-effects model to pool effect on outcomes according to different thresholds of cTTR.
Results: Four TTR sub-studies with a total of 71,222 patients were included. The benefit of NOAC in reducing SSE compared with warfarin was significantly higher in patients at cTTR<60% (HR 0.79, 95% CI 0.68-0.90) and at 60% to <70% (0.82, 0.71-0.95) but not at ≥70% (1.00, 0.82-1.23) with a significant interaction for cTTR<70% or ≥70% (p=0.042). The risk of major or NMCR bleeding was significantly lower with NOAC as compared with warfarin in patients at all sub-groups (0.67, 0.54-0.83 for patients at cTTR<60% and 0.75, 0.63-0.89 at 60% to <70%) except for cTTR≥70% (HR 0.84, 0.64-1.11), but the interaction for cTTR<70% or ≥70% was not statistically significant (p=0.271).
Conclusions: The superiority in efficacy of NOAC compared with warfarin for stroke prevention is lost above a cTTR threshold of approximately 70%, but the relative safety appears to be less modified by the centre-based quality of INR control.
Keywords: Atrial fibrillation (AF); Effectiveness; Non-vitamin K antagonist (NOAC); Safety; Time in therapeutic range (TTR); Warfarin.
Copyright © 2017 Elsevier B.V. All rights reserved.