Introduction: For current implantable defibrillators, the nonthoracotomy approach to implantation fails in a substantial number of patients. In a prospective randomized cross-over study the defibrillation efficacy of a standard monophasic and a new biphasic waveform was compared for different lead configurations. Methods and Results: Intraoperatively, in 79 patients receiving nonthoracotomy defibrillation leads, the defibrillation threshold was determined in the initial lead configuration for the mono-and biphasic waveform. In each patient, both waveforms were used alternately with declining energies (20, 15, 10, 5 J) until failure of defibrillation occurred. Three different initial lead configurations were tested in different, consecutive, nonrandomized patients using a bipolar endocardial defibrillation lead alone (A; n = 36) or in combination with a subcutaneous defibrillation patch (B; n = 24) or array (C; n = 19) lead. The lowest successful defibrillation energy with the biphasic waveform was less than, equal to, or higher than with the monophasic waveform in 64%, 28%, and 8% of patients, respectively, and on average significantly lower with the biphasic waveform for all three lead configurations (A: 11.3 +/- 4.4 J vs 14.5 +/- 4.5 J; B: 9.7 +/- 4.7 J vs 15.1 +/- 4.5 J; C: 7.9 +/- 4.5 J vs 12.4 +/- 4.9 J). Defibrillation efficacy at 20 J was significantly improved by the biphasic waveform (91 % vs 76 %). Conclusion: In combination with nonthoracotomy defibrillation leads, the biphasic waveform of a new implantable cardioverter defibrillator showed superior defibrillation efficacy in comparison to the standard monophasic waveform. Defibrillation thresholds were improved for lead systems with and without a subcutaneous patch or array lead.