Potential Embolic Sources and Outcomes in Embolic Stroke of Undetermined Source in the NAVIGATE-ESUS Trial

Background and Purpose— Emboli in embolic stroke of undetermined source (ESUS) may originate from various potential embolic sources (PES), some of which may respond better to anticoagulation, whereas others to antiplatelets. We analyzed whether rivaroxaban is associated with reduction of recurrent stroke compared with aspirin in patients with ESUS across different PES and by number of PES. Methods— We assessed the presence/absence of each PES (atrial cardiopathy, atrial fibrillation, arterial atherosclerosis, left ventricular dysfunction, cardiac valvulopathy, patent foramen ovale, cancer) in NAVIGATE-ESUS (New Approach Rivaroxaban Inhibition of Factor Xa in a Global Trial Versus ASA to Prevent Embolism in Embolic Stroke of Undetermined Source) participants. Prevalence of each PES, as well as treatment effect and risk of event for each PES were determined. Results by number of PES were also determined. The outcomes were ischemic stroke, all-cause mortality, cardiovascular mortality, and myocardial infarction. Results— In 7213 patients (38% women, mean age 67years) followed for a median of 11 months, the 3 most prevalent PES were atrial cardiopathy (37%), left ventricular disease (36%), and arterial atherosclerosis (29%). Forty-one percent of all patients had multiple PES, with 15% having ≥3 PES. None or a single PES was present in 23% and 36%, respectively. Recurrent ischemic stroke risk was similar for rivaroxaban- and aspirin-assigned patients for each PES, except for those with cardiac valvular disease which was marginally higher in rivaroxaban-assigned patients (hazard ratio, 1.8 [95% CI, 1.0–3.0]). All-cause mortality risks were similar across treatment groups for each PES while too few myocardial infarctions and cardiovascular deaths occurred for meaningful assessment. Increasing number of PES was not associated with increased stroke recurrence nor all-cause mortality, and outcomes did not vary between rivaroxaban- and aspirin-assigned patients by number of PES. Conclusions— A large proportion of patients with ESUS had multiple PES which could explain the neutral results of NAVIGATE-ESUS. Recurrence rates between rivaroxaban- and aspirin-assigned patients were similar across the spectrum of PES. Registration— URL: https://www.clinicaltrials.gov ; Unique identifier: NCT02313909.