Postoperative atrial tachyarrhythmias (POAT) were a common complication in patients undergoing cardiac surgery and were associated with increased morbidity and hospital length of stay (LOS). Dofetilide is a class III antiarrhythmic that was never tested as a preventative therapy during the perioperative period. The aim of this study was to assess whether perioperative dofetilide is an effective and safe therapy for the prevention of POAT, with a decrease in LOS. A double-blind, randomized, placebo-controlled study of patients undergoing coronary artery bypass grafting with or without valve surgery was conducted. Patients were randomly assigned to receive either dofetilide or placebo. Medication was dosed and administered according to the manufacturer's guidelines. Daily review of cardiac monitoring was performed to assess for the occurrence of atrial fibrillation and atrial flutter. The primary outcome was defined as any atrial tachyarrhythmia lasting > or =5 minutes, whereas secondary outcomes included LOS and adverse events attributed to dofetilide. One hundred thirty-three patients were randomly assigned to dofetilide (n = 67) or placebo (n = 66). The incidence of POAT in the placebo arm was 24 (36%), and in the dofetilide arm, 12 (18%; absolute risk reduction 18%; p <0.017; number needed to treat = 5.4). Patients who developed POAT in the dofetilide arm had a 1.0-day decrease in mean LOS versus placebo (5.7 vs 6.7 days; p = NS). There was no incidence of torsades de pointes in either group. In conclusion, perioperative dofetilide can decrease both the incidence of POAT and LOS in a safe manner.