Background and Purpose- Although believed to be transient and self-limiting, new-onset perioperative/postoperative atrial fibrillation (POAF) might be a risk factor for stroke and mortality. We conducted a systematic review and meta-analysis to qualitatively and quantitatively evaluate the relationship of POAF with early and late risks of mortality and stroke. Methods- We searched Pubmed, EMBASE, and Cochrane Library (1966 through March 2018) to identify cohort studies that reported stroke and mortality associated with POAF. We computed a random-effects estimate based on the Mantel-Haenszel method. Odds ratios with 95% CI were used as a measure of the association between POAF and early (in-hospital or within 30 days of surgery) stroke and mortality, while hazard ratios (HR) were used for long-term outcomes. Results- Our analysis included 35 studies with 2 458 010 patients. Pooling the results from the random-effects model showed that POAF was associated with increased risks of early stroke (odds ratio, 1.62; 95% CI, 1.47-1.80), early mortality (odds ratios, 1.44; 95% CI, 1.11-1.88), long-term stroke (HR, 1.37; 95% CI, 1.07-1.77), and long-term mortality (HR, 1.37; 95% CI, 1.27-1.49). Analyses focusing on high-quality studies obtained similar results. In subgroup analyses, POAF was more strongly associated with stroke in patients undergoing noncardiac surgery (HR, 2.00; 95% CI, 1.70-2.35) than in patients undergoing cardiac surgery (HR, 1.20; 95% CI, 1.07-1.34). Conclusions- New-onset POAF is associated with an increased risk of stroke and mortality, both in the short-term and long-term. The best strategy to reduce stroke risk among these patients needs to be determined.