Background: Clinical trial evidence supports patent foramen ovale (PFO) closure to reduce recurrent stroke risk in select patients. Guidelines recommend a multi-disciplinary approach to identify patients who may benefit from PFO closure. We assess factors associated with the decision to close PFO in a real-world cohort. Methods: We studied patients from University of Minnesota Medical Center and Yale-New Haven Hospital with TIA or ischemic stroke and PFOs evaluated by dedicated multidisciplinary teams of vascular neurologists and interventional cardiologists at each institution. Demographic, clinical, radiographic, and echocardiographic information was extracted by chart review. The study outcome was PFO closure. Descriptive analyses were performed comparing characteristics associated with PFO closure using Mann Whitney test and Chi-squared test. Multivariable logistic regression model was built with variables significant in descriptive analyses. Results: The analytic cohort included 173 patients, of whom 58 patients (33.5%) underwent closure (39.8% of Yale and 20% of UMMC patients were closed). Patients who were closed versus not were similar in age (median age 56 [IQR 21] versus 58 [IQR 21] years), sex, race, diagnosis of TIA or stroke, and co-morbidities except malignancy which was significantly less prevalent among closed patients (2.3% versus 12.1%). Closed patients had a higher RoPE score (median 6 versus 5, p=0.032), and were more likely to have a stroke or TIA of undetermined etiology (55.2 versus 49.6%) and high-risk PFO characteristics (p<0.001, Table). High-risk PFO characteristics (OR 6.4, 95% CI 2.1-19.7), RoPE score (OR 1.5, 95% CI 1.1-2.1), and undetermined etiology (OR 5.5, 95% CI 1.1-28.3) were independent predictors of PFO closure. Conclusions and Relevance: In this study of a real-world cohort, we identify patient characteristics that drive decision-making for PFO closure for secondary stroke prevention.