Background Epidemiological studies on smoking and atrial fibrillation have been inconsistent, with some studies showing a positive association while others have found no association. It is also unclear whether there is a dose-response relationship between the number of cigarettes smoked or pack-years and the risk of atrial fibrillation. We conducted a systematic review and meta-analysis to clarify the association. Design Systematic review and meta-analysis. Methods We searched the PubMed and Embase databases for studies of smoking and atrial fibrillation up to 20 July 2017. Prospective studies and nested case-control studies within cohort studies reporting adjusted relative risk estimates and 95% confidence intervals (CIs) of atrial fibrillation associated with smoking were included. Summary relative risks (95% CIs) were estimated using a random effects model. Results Twenty nine prospective studies (22 publications) were included. The summary relative risk was 1.32 (95% CI 1.12-1.56, I2 = 84%, n = 11 studies) for current smokers, 1.09 (95% CI 1.00-1.18, I2 = 33%, n = 9) for former smokers and 1.21 (95% CI 1.12-1.31, I2 = 80%, n = 14) for ever smokers compared to never smokers. Comparing current versus non-current smokers the summary relative risk was 1.33 (95% CI 1.14-1.56, I2 = 78%, n = 10). The summary relative risk was 1.14 (95% CI 1.10-1.20, I2 = 0%, n = 3) per 10 cigarettes per day and 1.16 (95% CI 1.09-1.25, I2 = 49%, n = 2) per 10 pack-years and there was no evidence of a non-linear association for cigarettes per day, Pnon-linearity = 0.17. Conclusions The current meta-analysis suggests that smoking is associated with an increased risk of atrial fibrillation in a dose-dependent matter, but the association is weaker among former smokers compared to current smokers.