The purpose of this study was to assess the effect of azimilide, a Class III antiarrhythmic drug, on ECG conduction intervals recorded during sinus rhythm in patients with a history of symptomatic atrial fibrillation or atrial flutter. In Phase I clinical pharmacology studies, azimilide was associated with prolongation of the QT and QTc intervals on electrocardiograms recorded during sinus rhythm in normal subjects, but the effect of azimilide on the target population of patients with atrial fibrillation has not been reported in detail previously. Patients with a history of atrial fibrillation, atrial flutter, or both were randomly assigned to receive placebo or azimilide twice daily for 3 days and then qd thereafter. Azimilide doses of 50 mg, 100 mg, or 125 mg were tested. The RR, PR, QRS, QT, QTc(Bazett), and QTc(Fridericia) intervals were measured from electrocardiograms recorded at baseline and on Day 4 of test therapy. Increasing azimilide doses were associated with significant increases in the RR, QT, QTc(Bazett), and QTc(Fridericia) compared with changes in the placebo group based on the F-test for differences among treatment groups and the test for a dose response. In the azimilide 125 mg dose group, the mean change in RR was significantlygreater than the mean change in the placebo group (+61.4 ms in the azimilide 125 mg group vs. -14.1 ms in the placebo group). The mean change in QT was significantly greater in the azimilide 125 mg group than the mean change in the placebo group (+44.2 ms in the azimilide 125 mg group vs. -1.0 ms in the placebo group). The mean change in QTc using both correction methods was significantly greater in the azimilide 125 mggroup than the mean change in the respective placebo group: QTc(Bazett) +31.6 ms in the azimilide 125 mg group versus +2.1 ms in the placebo group and QTc(Fridericia) +35.8 in the azimilide 125 mg group versus +1.0 ms in the placebo group. It was concluded that in patients with a history of atrial fibrillation or flutter, azimilide was associated with statistically significant increases in RR, QT, QTc(Bazett), and QTc(Fridericia) when patients were in sinus rhythm.