Introduction: Atrial fibrillation (afib) patients are often prescribed anticoagulants to reduce stroke risk. Recently, NOACs have begun to supplant warfarin because of faster onset/offset of effects and purportedly lower hemorrhage risk. This study compares the efficacy of NOACs vs. warfarin in preventing ischemic stroke and hemorrhages.
Hypothesis: We hypothesize a lower incidence of both ischemic stroke and hemorrhaging among afib patients taking NOACs than among those on warfarin because of the simpler NOAC dosing.
Methods: Records of 5593 afib patients on a NOAC and 5796 patients on warfarin were obtained from St. David’s HealthCare hospitals (Austin, TX). The relative incidence of both ischemic and hemorrhagic strokes (as well as all other non-traumatic hemorrhages) among these patients was compared using Fisher’s tests. In order to establish that patients prescribed warfarin vs. NOACs do not differ according to confounding risk factors, demographic characteristics and medical histories were compared using t-tests.
Results: There is significantly higher (Odds Ratio = 2.43, p <0.001) ischemic stroke incidence among patients taking NOACs. In contrast, hemorrhages generally (OR = 0.844, p=0.003) and hemorrhagic strokes in particular (OR = 0.21, p < 0.001) are much less frequent in patients taking NOACs. There was no significant difference between patients on warfarin and NOACs with respect to age. There is a higher proportion (54%) of male patients on NOACs, but no gender bias in the warfarin population. Patients on NOACs have a higher frequency of hypertension (OR = 6.6, p < 0.001) and of diabetes (OR = 2.3, p < 0.001).
Conclusions: NOAC use is associated with a much lower risk of hemorrhaging (especially hemorrhagic stroke), but with a significantly higher ischemic stroke risk than warfarin. The results indicate that simpler dose control with NOACs can mitigate the risk of hemorrhages, at the cost of being less effective at preventing embolisms among afib patients. These findings suggest that closer monitoring of patients taking NOACs for ischemic stroke risk may be warranted.