Introduction and Hypothesis: Previously we have reported that periodontal disease is associated with an increased risk of cardioembolic stroke. Atrial fibrillation (AF) is the most common reason for cardioembolic stroke. We assessed the hypothesis that periodontal disease is associated with incident AF. Methods: In the Dental Atherosclerosis in Communities Study (DARIC) study, participants were subjected to full-mouth clinical periodontal measurements (seven indices) collected at six sites per tooth from 6,501 subjects without prior AF. They were graded into seven distinct periodontal profile classes (PPC: A or periodontal health, F through G, based on increasing severity of periodontal disease). These patients were followed for AF adjudicated using electrocardiograms, hospital discharge codes, and death certificates, over the subsequent 17 years. Cox proportional hazards models adjusted for AF risk factors that can potentially confound the relationship between PPC and AF (Age, Gender, Race, Hypertension, Diabetes, Smoking, Coronary Artery Disease and Congestive Heart Failure). Results: Of 6,501 assessed on Visit 4 without prior history of atrial fibrillation, 883 were found to have AF over a 17-year follow-up period. Periodontal disease (PPC classes B-G) compared with periodontal health (PPC-A) was associated with higher rate of AF (HR 1.3 95% CI 1.1-1.5, Kaplan Meier plot, log rank p=0.0008). On univariate analysis PPC-D (HR 1.4 95% CI 1.1-1.7) PPC-E (HR 1.4 95% CI 1.2-1.8) PPC-F (HR 1.6 95% CI 1.3-2.0) and PPC-G (HR 1.4 (95% CI 1.04-1.8) were significantly associated with AF. On multivariate analysis only, PPC-F or severe tooth loss (adjusted HR 1.3 95% CI 1.01-1.6) was significantly associated with AF. Conclusions: We report a significant association periodontal disease and AF. Among the PPC, PPC-F or severe tooth loss is independently associated with AF. The association may explain the cardioembolic stroke risk noted with periodontal disease.