For conversion of atrial fibrillation to sinus rhythm and management of ventricular arrhythmias, antiarrhythmic drugs were frequently used. However, the effects of antiarrhythmic drugs on exercise performance and on the variability of ventricular rate were not available. This study included 37 patients who had chronic atrial fibrillation complicated with symptomatic ventricular arrhythmias. The patients were divided into three groups and received sotalol, propafenone, and procainamide, respectively. Before and after taking the drugs for 14 days, these patients received treadmill exercise test, 24 h Holter electrocardiogram, and tilt table test for evaluation of the exercise performance and the variability of ventricular rate (including the mean RR intervals, mRR, the standard deviation of RR intervals, SDRR, and the root mean square of the difference in successive RR intervals, rMSSD). All these antiarrhythmic drugs could suppress ventricular arrhythmia but only sotalol could significantly increase the exercise duration (374+/-50 to 476+/-55 s, P=0.02), and reduce the maximal heart rate (186+/-23 to 136+/-16 beats/min, P=0.01) during exercise test. Furthermore, only sotalol increased the mRR (777+/-60 to 885+/-66 ms, P=0.02), SDRR (190+/-40 to 216+/-48 ms, P=0.04) and rMSSD (223+/-48 to 253+/-40 ms, P=0.03) during 24 h Holter electrocardiogram. With head-up tilt, the mRR, SDRR and rMSSD all decreased significantly before drug therapy, and these changes were still present only after propafenone therapy. Therefore, comparisons among sotalol, propafenone and procainamide showed that sotalol increased the exercise performance and the variability of ventricular rate in patients who had chronic atrial fibrillation complicated with symptomatic ventricular arrhythmias.