Background Concomitant bipolar radiofrequency ablation and valve replacement in the elderly remains controversial. In the current study, we aimed to compare the outcomes of concomitant valve replacement and bipolar radiofrequency ablation with valve replacement alone in elderly patients with atrial fibrillation (AF). Methods This was a retrospective study of patients aged â„ 70 years who underwent valve replacement with or without bipolar radiofrequency ablation in a single-centre between January 2006 and March 2015. The early postoperative results and long-term clinical outcomes were compared after propensity score matching. Results A total of 34 pairs of patients (73.94 ± 2.64 years old; 34 in the AF with ablation group and 34 in the AF without ablation group) were enrolled in the propensity score matching analysis. There were no significant differences between the two matched groups in terms of surgical mortality (5.88% vs. 2.94%, P = 0.555) and major postoperative morbidity. KaplanâMeier analysis revealed a significantly better overall survival in the AF with ablation group compared to the AF without ablation group (P = 0.009). Cumulative incidence curves showed a lower incidence of cardiovascular death in the AF with ablation group (P = 0.025, Grayâs test). Patients in the AF with ablation group had a reduced incidence of stroke compared to patients in the AF with ablation group (P = 0.009, Grayâs test). The freedom from AF after 5 years was 58.0% in the AF with ablation group and 3.0% in the AF without ablation group. Conclusions The addition of bipolar radiofrequency ablation is a safe and feasible procedure, even in patients aged â„ 70 years, with a better long-term survival and a reduced incidence of stroke compared to valve replacement alone. These findings suggest that bipolar radiofrequency ablation should always be considered as a concomitant procedure for elderly patients with AF who require cardiac surgery. However, a large-scale, prospective, multi-centre, randomized study should be performed in the future to fully validate our findings.