Background - Recently, interest has arisen in using biphasic waveforms for external defibrillation. Little work has been done, however, in measuring transthoracic defibrillation efficacy after long periods of ventricular fibrillation in protocol 1, we compared the efficacy of a quasi-sinusoidal biphasic waveform (QSBW), a truncated exponential biphasic waveform (TEBW), and a critically damped sinusoidal monophasic waveform (CDSMW) after 15 seconds of fibrillation. In protocol 2, we compared the efficacy of the more efficacious biphasic waveform from protocol 1, QSBW, with CDSMW after 15 seconds and 5 minutes of fibrillation. Methods and Results - In protocol 1, 50% success levels, ED50, were measured after 15 seconds of fibrillation for the 3 waveforms in 6 dogs. In protocol 2, defibrillation thresholds were measured for QSBW and CDSMW after 15 seconds of fibrillation and after 3 minutes of unsupported fibrillation followed by 2 minutes of fibrillation with femoral-femoral cross-circulation. In protocol 1, QSBW had a lower ED50, 16.0 ± 4.9 J, than TEBW, 20.3 ± 4.4 J, or CDSMW, 27.4 ± 6.0 J. In protocol 2, QSBW had a lower defibrillation threshold after 15 seconds, 38 ± 10 J, and after 5 minutes, 41.5 ± 5 J, than CDSMW after 15 seconds, 54 ± 19 J, and 5 minutes, 80 ± 30 J, of fibrillation. The defibrillation thresheld remained statistically the same for QSBW for the 2 fibrillation durations but rose significantly for CDSMW. Conclusions - In this animal model of sudden death and resuscitation, these 2 biphasic waveforms are more efficacious than the CDSMW at short durations of fibrillation. Furthermore, the QSBW is even more efficacious than the CDSMW at longer durations of fibrillation.