Key points are not available for this paper at this time.
Purpose: Discontinuation rates for vitamin K antagonists are high in patients with atrial fibrillation (AF). The aim of the current study was to assess the impact of real-life treatment persistence on the incidence of ischaemic strokes in patients with AF treated with a vitamin K antagonist or one of the newer oral anticoagulants (NOACs: rivaroxaban 20 mg, dabigatran 110 mg or dabigatran 150 mg). Methods: Real-life data on persistence from a US claims database and a German registry were combined to estimate the persistence rates for warfarin, dabigatran and rivaroxaban over 6 months and to calculate the mean duration on treatment. A model, accounting for switching patterns, was developed, combining these data with published ischaemic stroke rates for each treatment. The model assumed an annual risk of ischaemic stroke of 1.65% and 4.59% for patients receiving warfarin or no treatment, respectively. To obtain the risk of stroke for patients receiving NOACs, published relative risks were applied to the warfarin risk, giving an annual probability of stroke between 1.29% and 1.78%. Results: In the first 6 months, patients starting on warfarin stayed on treatment for a mean of 131 days before switching to another NOAC or stopping treatment completely. Patients starting on a NOAC had a longer duration on treatment (dabigatran combined dose: 149 days; rivaroxaban 20 mg: 168 days), before switching to warfarin or stopping treatment completely. The total ischaemic stroke risk at 6 months for each strategy was estimated to be 1.19% (warfarin), 1.12% (dabigatran 110 mg), 0.91% (dabigatran 150 mg) and 0.87% (rivaroxaban 20 mg). Compared with patients starting on warfarin, these risks translated into 7, 28 and 32 strokes avoided with dabigatran 110 mg, dabigatran 150 mg and rivaroxaban 20 mg, respectively, in a hypothetical cohort of 10,000 patients with AF. Conclusions: The model indicates that starting patients on a NOAC may decrease the total number of ischaemic strokes relative to warfarin therapy. Moreover, persistence rates should be considered in addition to treatment effects for the assessment of the overall value of oral anticoagulants, as better persistence may outweigh differences in risk reduction.
Evers et al. (Fri,) studied this question.