Cynthia A Jackevicius1,2,3,4, Lingyun Lu1, Zunera Ghaznavi5,6, Alberta L Warner5,7. 1. Department of Pharmacy (C.A.J., L.L.), Veterans Affairs Greater Los Angeles Healthcare System, CA. 2. Department of Pharmacy Practice and Administration, College of Pharmacy, Western University of Health Sciences, Pomona, CA (C.A.J). 3. ICES, Toronto, Canada (C.A.J). 4. Institute of Health Policy, Management and Evaluation, University of Toronto, Canada (C.A.J). 5. Department of Cardiology (A.L.W., Z.G.), Veterans Affairs Greater Los Angeles Healthcare System, CA. 6. Department of Preventative Medicine, University of Southern California (ZG). 7. David Geffen School of Medicine at UCLA (A.L.W.).
Abstract
BACKGROUND: Patients with heart failure and atrial fibrillation are an important atrial fibrillation subgroup in which direct oral anticoagulants (DOACs) have not been adequately studied in real-world settings. Since DOACs rely on renal elimination and renal dysfunction is prevalent in patients with heart failure, their use may increase bleeding risk, negating some of their advantage over warfarin. METHODS: We conducted a retrospective cohort study using linked Veterans Administration databases of patients with heart failure newly started on warfarin or DOACs for atrial fibrillation from October 2010 to August 2017 (23 635 warfarin, 25 823 DOAC). Outcomes included time to first bleeding, stroke, and death using Cox proportional hazards models with inverse probability of treatment weighting. RESULTS: Total bleeding (hazard ratio, 0.62 [95% CI, 0.56-0.68]), major bleeding (hazard ratio, 0.49 [95% CI, 0.40-0.61]), and death (hazard ratio, 0.74 [95% CI, 0.71-0.78]) were lower with DOAC than warfarin, and with apixaban and dabigatran, but not rivaroxaban. Moderate/severe chronic kidney disease was common (48.7%); moderate chronic kidney disease was associated with increased bleeding with DOACs but not warfarin. However, death and bleeding remained lower with DOACs than warfarin across all renal function levels and clinical subgroups. A >20% transient/persistent decline in renal function occurred in 53% of DOAC-treated patients at some point during follow-up, would have required dose reduction in 10.5% of patients, and was associated with increased bleeding. Dose adjustments were made more often, and bleeding and death were lower in patients seen by pharmacists or anticoagulation clinics. There were significant between-site variations in DOAC dosing. CONCLUSIONS: DOACs overall, apixaban, and dabigatran, but not rivaroxaban, were associated with less total bleeding and death than warfarin in patients with heart failure and atrial fibrillation at all levels of renal function. Renal function decline resulted in increased bleeding in patients with DOACs. DOAC dose adjustment was often indicated, associated with increased bleeding when not adjusted, emphasizing the need for closer monitoring in these patients.
BACKGROUND:Patients with heart failure and atrial fibrillation are an important atrial fibrillation subgroup in which direct oral anticoagulants (DOACs) have not been adequately studied in real-world settings. Since DOACs rely on renal elimination and renal dysfunction is prevalent in patients with heart failure, their use may increase bleeding risk, negating some of their advantage over warfarin. METHODS: We conducted a retrospective cohort study using linked Veterans Administration databases of patients with heart failure newly started on warfarin or DOACs for atrial fibrillation from October 2010 to August 2017 (23 635 warfarin, 25 823 DOAC). Outcomes included time to first bleeding, stroke, and death using Cox proportional hazards models with inverse probability of treatment weighting. RESULTS: Total bleeding (hazard ratio, 0.62 [95% CI, 0.56-0.68]), major bleeding (hazard ratio, 0.49 [95% CI, 0.40-0.61]), and death (hazard ratio, 0.74 [95% CI, 0.71-0.78]) were lower with DOAC than warfarin, and with apixaban and dabigatran, but not rivaroxaban. Moderate/severe chronic kidney disease was common (48.7%); moderate chronic kidney disease was associated with increased bleeding with DOACs but not warfarin. However, death and bleeding remained lower with DOACs than warfarin across all renal function levels and clinical subgroups. A >20% transient/persistent decline in renal function occurred in 53% of DOAC-treated patients at some point during follow-up, would have required dose reduction in 10.5% of patients, and was associated with increased bleeding. Dose adjustments were made more often, and bleeding and death were lower in patients seen by pharmacists or anticoagulation clinics. There were significant between-site variations in DOAC dosing. CONCLUSIONS:DOACs overall, apixaban, and dabigatran, but not rivaroxaban, were associated with less total bleeding and death than warfarin in patients with heart failure and atrial fibrillation at all levels of renal function. Renal function decline resulted in increased bleeding in patients with DOACs. DOAC dose adjustment was often indicated, associated with increased bleeding when not adjusted, emphasizing the need for closer monitoring in these patients.