Introduction: It has recently been demonstrated in a meta-analysis that patients with a history of intracerebral hemorrhage (ICH) and atrial fibrillation have a reduced risk of ischemic stroke (IS) with the initiation or resumption of oral anticoagulation (OAC). It remains uncertain if there is a difference in clinical outcomes between direct oral anticoagulants (DOAC) compared to warfarin (WF) in this patient population. We aimed to compare these two populations using data from the Southern California Kaiser Permanente (SCKP) health care system.
Methods: Patient records with ICH and AF diagnosis were obtained from the SCKP Database, and divided into two cohorts: WF or DOAC. Subjects were included if the OAC was started after the index ICH. CT scans within 24 hours of symptom onset were examined for: ICH location, ICH size (Using the ABC/2 method), and leukoariosis. If available, an axial gradient echo sequence MRI scan taken within 180 days, either before or after the index ICH was examined for microbleeds. Differences in demographics, concurrent medications with an antithrombotic effect and statins, and traditional vascular risk factors were assessed using Kruskal-Wallis and Chi-Square tests.
Results: Eighty-one subjects were identified (53 WF and 28 DOAC). In the DOAC group, 18 patients were on dabigatran, 9 on apixaban and 1 on rivaroxaban. There was no difference between WF and DOAC in baseline demographics, concurrent medications and location of ICH. The mean ICH volume was 15.4cc in DOAC and 5.6cc in WF (P=0.09). Time to start drug from ICH was 15.0 and 24.5 months for DOAC and WF respectively (P=0.26). More patients died in WF (28.3%) vs DOAC (10.7%) (P=0.0478). There was no difference in time to IS, ICH and major hemorrhage between the two groups. The composite outcome of IS, ICH, Major Hemorrhage and death was 35.8 % in WF vs 25% in DOAC group (P=0.32).
Conclusion: Our results, while limited in size, suggest that patients with a previous ICH with AF, who are started on a DOAC have lower mortality compared to WF. There was a higher percentage of patients in the WF group who met the composite outcome of IS, ICH, major hemorrhage and death, but this did not reach statistical significance.