BACKGROUND Biphasic waveforms defibrillate more effectively than monophasic waveforms; however, the mechanism remains unknown. The "upper-limit-of-vulnerability" hypothesis of defibrillation suggests that unsuccessful defibrillation is due to reinduction of ventricular fibrillation (VF). Thus, VF induction mechanisms may be important for the understanding of defibrillation mechanisms. We therefore compared myocardial VF vulnerability for monophasic versus biphasic shocks. RESULTS In 10 Langendorff-perfused rabbit hearts, monophasic and biphasic T-wave shocks were randomly administered over a wide range of shock coupling intervals and shock strengths, and the two-dimensional coordinates within which VF was induced were used to calculate the area of vulnerability (AOV) for both shock waveforms. The arrhythmic response to biphasic shocks differed from that to monophasic shocks in three distinct ways: (1) the AOV was smaller (8.9 +/- 4.2 versus 13.9 +/- 6.0 area units, P < .02), (2) the transition zone between VF-inducing and nonarrhythmogenic shocks was narrower (14.7 +/- 4.8 versus 29.9 +/- 6.4 area units, P < .001), and (3) the entire AOV shifted toward longer coupling intervals (by 11.0 +/- 8.8 ms at the left border [P < .005] and 6.0 +/- 5.2 ms at the right border [P = .005] of the AOV). CONCLUSIONS Biphasic shocks encounter a smaller AOV than monophasic shocks, a narrower transition zone from VF to no arrhythmia induction, and a lesser effectiveness in inducing VF at short coupling intervals. In keeping with the upper-limit-of-vulnerability hypothesis, these waveform-dependent differences in VF inducibility might help explain the lower defibrillation threshold for biphasic shocks.