ObjectiveSurgical ablation of atrial fibrillation (AF) is indicated both in patients with AF undergoing concomitant cardiac surgery and in those who have not responded to medical and/or catheter-based ablation therapy. This study examined our long-term outcomes following the Cox-Maze IV procedure (CMP-IV). MethodsBetween May 2003 and March 2018, 853 patients underwent either biatrial CMP-IV (n = 765) or a left-sided CMP-IV (n = 88) lesion set with complete isolation of the posterior left atrium. Freedom from atrial tachyarrhythmia (ATA) was assessed for up to 10 years. Rhythm outcomes were compared in multiple subgroups. Predictors of recurrence were determined using Fine–Gray regression, allowing for death as the competing risk. ResultsThe majority of patients (513/853, 60%) had nonparoxysmal AF. Twenty-four percent of patients (201/853) had not responded to at least 1 catheter-based ablation. Prolonged monitoring was used in 76% (647/853) of patients during their follow-up. Freedom from ATA was 92% (552/598), 84% (213/253), and 77% (67/87) at 1, 5, and 10 years, respectively. By competing risk analysis, incidence of first ATA recurrence was 11%, 23%, and 35% at 1, 5, and 10 years, respectively. On Fine–Gray regression, age, peripheral vascular disease, nonparoxysmal AF, left atrial size, early postoperative ATAs, and absence of sinus rhythm at discharge were the predictors of first ATA recurrence over 10 years of follow-up. ConclusionsThe CMP-IV had an excellent long-term efficacy at maintaining sinus rhythm. At late follow-up, the results of the CMP-IV remained superior to those reported for catheter ablation and other forms of surgical ablation for AF. Age, left atrial size, and nonparoxysmal AF were the most relevant predictors of late recurrence.