Abstract 3590: Defining a Threshold of Anticoagulation Quality that Leads to Net Clinical Benefit in Atrial Fibrillation: The ATRIA Study
Background. Warfarin is highly efficacious for stroke prevention in atrial fibrillation (AF), but achieving high quality anticoagulation is challenging. The threshold of time in therapeutic INR range 2–3 (%TTR) needed to maximize the net benefit of warfarin in AF is unknown.
Methods. We examined anticoagulation quality (%TTR) and outcomes in 13,559 patients with AF (1996 –2003). Longitudinal warfarin use was defined from prescriptions and INR tests obtained from pharmacy and lab databases. %TTR was based on INR results using linear interpolation. Thromboembolic (TE) and intracranial hemorrhage (ICH) events were identified from hospitalization databases and confirmed by chart review. We used multivariable Poisson regression to examine how %TTR affected TE and ICH rates. We then examined whether net outcomes associated with warfarin (net TE prevented+net ICH caused) varied by %TTR.
Results. During 64,415 person-years of follow-up, 8160 subjects received warfarin and we validated 1041 TE and 279 ICH events. Among warfarin users, the distribution of %TTR was: <30% (8.4%), 30 –39% (5.6%), 40 – 49% (10.1%), 50 –59% (17.2%), 60 – 69% (24.3%) and 70+% (34.4%). Compared with no anticoagulation, the adjusted benefit for preventing TE with warfarin increased with higher %TTR while the harm of ICH was lower with higher %TTR (Fig⇓). An overall net benefit (TE prevented+ICH caused) was observed only when the %TTR was 50% or higher (Fig⇓).
Conclusion. Anticoagulation quality affects net outcomes in AF, with a net benefit of warfarin that started at %TTR of 50% and rose sharply as %TTR increased. These results can help define outcome-based standards for delivering quality anticoagulation therapy.