Background-Phase duration of biphasic shocks may be an important determinant of defibrillation success. The purpose of this study was to investigate the effect of changing phase duration of biphasic pulses delivered by 70-F capacitors on defibrillation energy requirements. This may be clinically relevant for the optimization of implantable cardioverterdefibrillator design and programming. Methods and Results-Defibrillation thresholds (DFTs) were determined for 13 waveforms in 13 pigs by application of a 70-F capacitance and a transvenous/submuscular lead system. In part I, phase-1 duration varied, preserving a phase-1/phase-2 duration ratio of 60%/40%. The phase-1 durations were 1, 2, 3, 4, 5, and 6 ms. The DFT was lowest (22.9Ϯ7 J) for phase 1ϭ3 ms compared with phase 1ϭ1 ms (36.4Ϯ7.5 J), 2 ms (25Ϯ6.5 J), 4 ms (25Ϯ7.6 J), 5 ms (30.7Ϯ7.3 J), or 6 ms (32.9Ϯ8.1 J) (PϽ.001). In part II, phase-1 duration was 3 ms but phase-2 duration varied: 0.7, 1.3, 2, 2.7, 3.3, 4, and 6 ms. Significant DFT minima were found at phase 2ϭ2 ms (22.5Ϯ4.2 J) and phase 2ϭ4 ms (22.5Ϯ4.2 J) compared with phase 2ϭ0.7 ms (31.7Ϯ9.3 J), phase 2ϭ3.3 ms (26.7Ϯ6.1 J), or phase 2ϭ6 ms (28.3Ϯ6.8 J) (PϽ.05). Conclusions-The strength-duration curve of biphasic defibrillation shocks demonstrates a single optimum for phase-1 duration. In contrast, two optima with minimal energy requirements were found for phase-2 duration. Optimization of both phases of low-capacitance biphasic shocks may reduce energy requirements for defibrillation. (Circulation. 1998;97:2073-2078.)