A randomized, prospective comparison of the defibrillation efficacy of various shock waveforms and nonthoracotomy lead configurations was performed in five distinct patient groups undergoing implantation of a cardioverter defibrillator. In the first group using a bidirectional lead configuration, there was no significant difference in the mean defibrillation threshold (DFT) between simultaneous and sequential monophasic shocks (17.8 +/- 5.8 joules versus 17.3 +/- 2.7 joules). In the second group using a bidirectional lead configuration, the mean DFT was 21.9 +/- 7.3 joules with monophasic shocks and 14.9 +/- 5.0 joules with biphasic shocks (p < 0.001). In the third group using a unidirectional lead configuration, the mean DFT was significantly higher (p < 0.001) with monophasic shocks (22.1 +/- 4.2 joules) compared with biphasic shocks (15.0 +/- 5.4 joules). In the fourth group, an intraindividual comparison with monophasic shock waveforms showed no significant differences in DFT using either a bidirectional (21.3 +/- 5.8 joules) or a unidirectional (21.7 +/- 2.6 joules) lead configuration. In the fifth group, a simplified unipolar transvenous defibrillation lead system ("active can") demonstrated significant lower DFTs (9.7 +/- 3.8 joules) compared with a standardized unidirectional lead configuration (18.0 +/- 6.8 joules). It is concluded that: (1) there seems to be no significant difference in the DFT between simultaneous and sequential monophasic shocks; (2) biphasic waveforms require significantly less energy for defibrillation than their corresponding monophasic waveforms; and (3) the unipolar single-electrode defibrillation system is easy to implant and provides DFTs at energies comparable with epicardial lead systems.