Background: Post-operative (POP) atrial fibrillation (AF) is frequent in patients who undergo cardiac surgery. However, its prognostic impact in the long term remains unclear. Methods: We followed 1386 patients who underwent cardiac surgery for an average of 10 ± 3 years. According to clinical history of AF before and after surgery, four subgroups were identified: (1) patients with no history of AF and without episodes of AF during the first 30 days after surgery (control or Group 1, n = 726), (2) patients with no history of AF before surgery in whom new-onset POP AF was detected during the first 30 days after surgery (new-onset POP AF or Group 2, n = 452), (3) patients with a history of paroxysmal/persistent AF before cardiac surgery (Group 3, n = 125, including 87 POP AF patients and 38 who did not develop POP AF), and (4) patients with permanent AF at the time of cardiac surgery (Group 4, n = 83). All-cause mortality was the primary outcome of the study. We tested the associations of potential determinants with all-cause mortality using univariable and multivariable statistical analyses. Results: Overall, 473 patients (34%) died during follow-up. After adjustment for multiple confounders, new-onset POP AF (hazard ratio (HR) = 1.31, 95% confidence interval (CI): 0.90–1.89; p = 0.1609), history of paroxysmal/persistent AF before cardiac surgery (HR = 1.33, 95% CI: 0.71–2.49; p = 0.3736), and permanent AF (Group 4) (HR = 1.55, 95% CI 0.82–2.95; p = 0.1803) were not associated with a significantly increased risk of mortality when compared with Group 1 (patients with no history of AF and without episodes of AF during the first 30 days after surgery). In new-onset POP AF patients, oral anticoagulation was not associated with mortality (HR = 1.13, 95% CI: 0.83–1.54; p = 0.4299). Conclusions: In this cohort of patients who underwent different types of heart surgery, POP AF was not associated with an increased risk of mortality. In this setting, the role of long-term anticoagulation remains unclear.