This study was designed to test the comparative efficacy of burst pacing, autodecremental (ramp) pacing, and universal (steep ramp) pacing for termination of ventricular tachycardia. A prospective, randomized sequence cross-over design was used to achieve comparisons of the pacing modalities that were matched for patient, day, and ventricular tachycardia characteristics. Thirty eight patients were enrolled, whose ventricular tachycardia was well-enough tolerated to be reinduced, and tested with 3 pacing modalities. There were 27 series 1 patients in which the pacing modalities were nonsynchronized burst pacing, synchronized burst pacing, and ramp pacing. The 11 patients in series 2 were tested with synchronized burst pacing, ramp pacing, and universal pacing. All pacing methods proved to be comparable in their ability to terminate ventricular tachycardia (p = NS). The 2 burst methods required the fewest number of attempts (significant vs ramp pacing). Universal pacing required the fewest number of stimuli. The mean paced cycle length was similar will all methods. The shortest paced cycle lengths were found with the autodecremental and universal methods because of their ramp patterns. It is concluded that burst, ramp, and universal pacing are of similar efficacy, although ramps were least efficient. Choice of a modality depends on operator preference, and individual patient response.