The aim of the present study was to explore substitution of intake of saturated fatty acids (FAs) with monounsaturated and polyunsaturated FAs and incident atrial fibrillation (AF) in men and women. A total of 57,053 Danish participants aged 50 to 64 years were enrolled in the Diet, Cancer and Health cohort study in 1993 to 1997 and completed a semiquantitative food frequency questionnaire at baseline. Follow-up was registry-based and data were analyzed using Cox proportional hazards regression. The statistical model was formulated in such a way that 1?g/day of saturated FAs was replaced with 1?g/day of monounsaturated or polyunsaturated FAs while keeping total fat intake, total energy intake, and energy intake from protein and carbohydrates constant. During a median follow-up of 17 years, 5,175 incident cases of AF occurred. In men, there was a higher hazard of AF when total n-3 polyunsaturated FAs replaced dietary saturated FAs-hazard ratio per 1-g substitution of FAs of 1.08 (95% confidence interval 1.02 to 1.14) in a model adjusted for lifestyle factors. For other substitutions of FAs (monounsaturated, total or n-6 polyunsaturated FAs), no consistent nor statistically significant associations were found. In conclusion, we found a moderately higher risk of AF in men, but not in women, when total n-3 polyunsaturated FAs replaced dietary saturated FAs. Substitution of saturated FAs with monounsaturated, total or n-6 polyunsaturated FAs was not associated with the risk of AF.