How good is time-in-therapeutic-range defined warfarin treatment compared to standard dose direct oral anticoagulants? Incidence of stroke in atrial fibrillation in a nationwide registry study

Abstract

Abstract Funding Acknowledgements Type of funding sources: Public hospital(s). Main funding source(s): The Finnish Foundation for Cardiovascular Research, and Helsinki and Uusimaa Hospital District research fund. Background and aims Direct oral anticoagulants (DOACs) have largely replaced warfarin in patients with atrial fibrillation (AF) to reduce risk of stroke. However, it is not known whether DOACs are more efficacious in preventing stroke or safer than good-quality warfarin treatment with high time in therapeutic range (TTR). Methods The Finnish AntiCoagulation in Atrial Fibrillation (FinACAF) is a nationwide study of AF patients that combines data from several Finnish health care registers. We analyzed all new-onset AF patients in Finland from January 2011 to December 2018 with laboratory data available. Comorbidities, concomitant medications and CHA2DS2-VASc -scores were recorded until date of AF diagnosis. OAC purchases from diagnosis of AF to the end of study were analyzed. ICD-10 diagnoses for stroke (ischemic or hemorrhagic) or transient ischemic attack during OAC use were analyzed until maximum follow-up of 730 days. Confounding factors by baseline characteristics as well as medications were taken in account by use of an inverse probability of treatment weighted analysis. For warfarin users, time in therapeutic range (TTR) was calculated and patient quartiles by TTR were assigned. Rate of stroke in different TTR quartiles for warfarin users were compared to standard dose dabigatran (150 mg bd.), apixaban (5mg bd.) and rivaroxaban (20mg od.). Results Altogether, 74 008 new-onset AF patients with OAC therapy (50.4 % male, mean age 73.0 years, 43 549 on warfarin, with mean TTR 66% and median TTR 72%) were followed for 109 143 patient years (py). Rate of stroke among patients with standard dose direct oral anticoagulants was 6.0/100 py for dabigatran (n=4545), 6.2/100 py for apixaban (n=12 426) and 4.0/100 py for rivaroxaban (n=12 950). In the TTR quartiles of patients on warfarin, rates of stroke were 9.3, 7.2, 5.5 and 4.3/100 py from the lowest (mean TTR 32%) to the highest quartile (mean TTR 90%), respectively. The weighted rates of stroke were 6.3, 6.1 and 4.5/100 py for dabigatran, apixaban and rivaroxaban, respectively, and for warfarin from the lowest to the highest quartile: 8.8, 7.0, 5.5 and 4.2/100 py, respectively. The Figure provides survival probabilities (crude rates) of stroke with confidence intervals of the analysed OAC groups. Conclusion Rates of stroke were evidently the highest among patients on warfarin in the two lowest TTR quartiles, while the differences between high TTR groups and standard dose DOACs were only modest.