Importance: The risk of ischemic stroke during periods of warfarin discontinuation for surgical procedures is recognized but not well characterized. Objective: The study aimed to quantitate the risk of ischemic stroke associated with high risk atrial fibrillation during periods of warfarin discontinuation. Design, Setting and Participants: A cohort of 4,060 patients (mean follow-up period of 3.5 ± 1.3 years) were randomized into the Atrial Fibrillation Follow-Up Investigation of Rhythm Management study. Patients enrolled in the study had atrial fibrillation plus at least one other risk factor for stroke or death: age ≥65 years', systemic hypertension, diabetes mellitus, congestive heart failure, transient ischemic attack, prior stroke, left atrium >50 mm, left ventricular fractional shortening <25% or left ventricular ejection fraction <40%. Exposure: Warfarin discontinuation for procedure. Main Outcome and Measures: The association of warfarin discontinuation with the incidence of ischemic stroke using pooled repeated measures and Cox proportional hazards analyses during follow-up after adjusting for age, gender, obesity, diabetes mellitus, hypercholesterolemia, cigarette smoking and study period. Results: Warfarin discontinuation for procedure occurred in 265 (0.4%) of the 71,355 person observations. Compared with those without warfarin discontinuation, the rate of ischemic stroke was higher among participants with surgery-related warfarin discontinuation (1.1% of 265 person observations vs. 0.2% of 71,090 person observations, p = 0.001). Warfarin discontinuation was associated with an increased risk for ischemic stroke (relative risk 5.8; 95% CI 1.8-18.4) after adjusting for potential confounders. The population-attributable risk associated with surgery-related warfarin discontinuation was estimated to be 23.1% (95% CI 15.2-30.9%) for ischemic stroke. Conclusions and Relevance: The 6-fold higher risk of ischemic stroke associated with discontinuation of warfarin for surgical procedures must be recognized in high risk atrial fibrillation patients and considered in the risk-benefit analysis of any procedure.