Abstract 97: Pre-Existing Anticoagulation Does Not Affect Stroke Rates in Patients With Left Sided Infective Endocarditis

Abstract

Introduction: Stroke is a devastating and relatively frequent complication of infective endocarditis (IE), especially infection involving left-sided valves. Management of anticoagulation (AC) in left sided IE (LSIE) is controversial as it is unclear whether or not AC impacts stroke risk in this population. Current guidelines are equivocal on their recommendations regarding AC management. The objective of this study was to evaluate the effect of AC on neurologic outcomes in patients with LSIE. Methods: All patients admitted to a tertiary academic hospital in North Carolina, with a diagnosis of left sided IE between December 2011 and April 2018 were identified. Patients were stratified based on the receipt of therapeutic AC. The primary outcome measure was the rate of radiographically confirmed stroke at 10 weeks. Relevant data were analyzed with univariate and multivariate analysis. Results: Two-hundred and fifty-eight consecutive patients with LSIE were identified. Individuals receiving AC (n=50) were older (median age of 63 vs 52, p = 0.02) more likely to have a history of atrial fibrillation (34% vs 5%, p < 0.01), had more prosthetic valve IE (38% vs 10.1%, p <0.01) and had lower incidence of mitral valve IE (40% vs 62%, p < 0.01). There was no significant difference in the rate of stroke, cerebrovascular hemorrhage, or mortality at ten weeks between the two cohorts. Mitral valve involvement (OR 2.20, 95CI 1.25-3.988) and intravenous drug use within 3 months of presentation (OR 2.10, 95CI 1.11-3.95) were identified as independent risk factors for stroke. Conclusions: AC therapy did not appear to have an effect on stroke in patients with LSIE. Furthermore, use of AC did not result in a greater occurrence of cerebrovascular hemorrhage or increased mortality. Continuation of AC in patients with a pre-existing indication should be considered in patients with LSIE in the absence of other contraindications. More robust studies are warranted to substantiate this recommendation.