Abstract TMP116: Predictors of Readmissions Following Percutaneous Patent Foramen Ovale Closure in a Nationally Representative Dataset

Abstract

Introduction: The optimal management of patients with recurrent stroke and patent foramen ovale (PFO) is debated. While some are managed medically with antiplatelet/anticoagulant agents, others undergo percutaneous closure of PFO (PC PFO). Device-specific trials have demonstrated a risk of atrial fibrillation/flutter (AF) following closure. To date, nationally representative data has not yet been used to assess the rates and causes of readmission following PC PFO. Methods: The 2013 Nationwide Readmissions Database is a national database of readmissions for all payers and the uninsured with data on >14 million U.S. admissions. We used validated International Classification of Disease, Ninth Revision, Clinical Modification codes to identify PC PFO, ischemic stroke, and medical comorbidities and complications. We calculated readmission rate following PC PFO. We performed Kaplan-Meier analysis to determine the cumulative 1-year risk of readmission with AF and Cox regression modelling to calculate hazard ratios (HR) of AF readmission at 1 year. Results: Among 850 patients who underwent PC PFO, there were 151(17.7%) readmissions within 1 year and 52 within the first 30 days. 11.5% of 30-day readmissions were for atrial fibrillation and 3.9% for flutter. Kaplan-Meier cumulative risk for readmission with AF at 1 year was 2.17%. Readmission rate for ischemic stroke was <1% (0.38%, 0.69%, and 0.61% at 30, 60, and 90 days respectively), compared to just over 1% for AF (1.15%, 1.38%, and 1.37% at 30, 60, and 90 days respectively). Predictors of AF readmission included history of congestive heart failure (CHF) (HR 3.53, 95% CI 1.23-10.17, p = 0.02) and increasing age (HR 1.05 per year, 1.02-1.09, p = 0.002). Conclusion: 1 in 5 of those who undergo PC PFO are readmitted in the first year, and 1/3 of these readmissions occur within 30 days. PC PFO is associated with low readmission rates for stroke and a low risk of the independent vascular risk factor of AF. The hazard for AF is over 3-fold greater for those with CHF and increases with each year increase in age.