Altered Fibrin Clot Properties Predict Stroke and Bleedings in Patients With Atrial Fibrillation on Rivaroxaban

Background and Purpose— We investigated whether clot permeability can predict clinically relevant outcomes in patients with atrial fibrillation (AF) treated with rivaroxaban. Methods— In the cohort study, we enrolled 232 consecutive patients with AF on rivaroxaban 20 mg/d (76.3%) or 15 mg/d (23.7%) for at least 3 months. Plasma clot permeability (K s ), a measure of fibrin network density, was determined 24 to 30 hours since the intake of rivaroxaban at undetectable drug’s levels. Ischemic cerebrovascular events and bleedings were recorded. Results— During a median follow-up of 48 months, patients with K s below median (6.8 cm 2 ·10 −9 ) had higher prevalence of stroke (5.84 versus 0.88% per year; P <0.0001) and relevant bleeding (7.06 versus 0.88% per year; P <0.0001) compared with those above median. The mortality rate was 1.53% per year and was not associated with K s . Lower K s predicted cerebrovascular ischemic events (hazard ratio, 6.64; 95% CI, 2.2–20.1) and relevant bleedings (hazard ratio, 7.38; 95% CI, 2.58–21.10). Minor bleeds (32.8% of patients) were observed more often in patients with K s above median (50.9 versus 14.7%; P <0.0001). Multivariate Cox regression analysis showed that in AF patients on rivaroxaban lower K s increased the risk of stroke (hazard ratio, 6.51; 95% CI, 2.14–19.75) and relevant bleedings (hazard ratio, 9.68; 95% CI, 3.21–29.18). Conclusions— Decreased clot permeability in AF patients can predict thromboembolic and clinically relevant bleeding events during therapy with rivaroxaban, while looser clot networks predispose to minor bleeds.