Abstract 16540: Impact of Baseline Stroke Risk and Bleeding Risk on INR Control among Patients with Atrial Fibrillation on Warfarin: the TREAT-AF Study
Background: In atrial fibrillation (AF), time in INR therapeutic range of 2.0-3.0 (TTR) mediates stroke reduction and bleeding risk with warfarin. Comorbidities may affect ability to achieve adequate TTR. We investigated whether TTR varies by baseline stroke (CHADS2 score) and bleeding risk (ATRIA score).
Methods: The Retrospective Evaluation and Assessment of Therapies in AF (TREAT-AF) study is a retrospective cohort study of patients with newly diagnosed AF treated in the Veterans Health Administration. National claims and lab data were used to identify patients with newly diagnosed AF between 10/1/04 - 9/30/08. We identified warfarin use from drug prescriptions data up to 90 days after first AF diagnosis (index date). Baseline CHADS2 (stroke) and ATRIA (bleeding) risk scores were calculated from validated ICD9 algorithms and laboratory data. We calculated first-year and long-term TTRs and outpatient INR monitoring rates (INRMR: % of months covered with an INR check) during warfarin treatment.
Results: From 123,188 patients with newly diagnosed AF, 50,527 patients (41.1%) had VA-prescribed warfarin and INR testing (mean age 72.1±10.3 y; 1.6% women). First-year outpatient INRMRs were higher than long-term (71.9% vs 64.6%, p<0.0001), but first-year TTRs were lower than long-term (47.1% vs 61.3%,p<0.0001). Although mean first-year and long-term INRMRs increased across strata of CHADS2 and ATRIA risk, the TTRs decreased (Table 1). After multivariate adjustment for age, sex, race, comorbidities, medications, VA service eligibility, and distance to VA facilities, CHADS2 stroke risk and ATRIA bleeding risk scores were independently associated with declines in TTR (p<0.0001 for all).
Conclusion: Increased baseline stroke or bleeding risk is associated with poor INR control, which may paradoxically limit warfarin’s safety and effectiveness among AF patients most vulnerable to over- or underanticoagulation.
- © 2012 by American Heart Association, Inc.