Mechanism of cardiac defibrillation in open-chest dogs with unipolar DC-coupled simultaneous activation and shock potential recordings.

The automatic implantable cardioverter-defibrillator has been shown to dramatically improve survival. The future refinement of these devices requires a clear understanding of their mechanism of action. We performed the following study to test two hypotheses: 1) When defibrillation is successful, fibrillating activity must be annihilated in a critical mass of both ventricles; and 2) when defibrillation is unsuccessful, at least one area of the ventricular mass has been left fibrillating. Unipolar Ag/AgCl sintered electrodes were directly coupled from triangular arrays at 40 epicardial locations (total, 120 recording sites) that covered both right and left ventricular surfaces and were designed to measure the voltage gradient generated by the shock at each triangular array as well as the underlying myocardial electrical activity before and immediately after the shock. An algorithm was developed and tested that reliably scored whether a postshock activation was a continuation of the immediately previous fibrillating activity. This technique was applied to 203 defibrillation attempts in six open-chest dogs during electrically induced ventricular fibrillation. There were 139 successful defibrillation attempts and 64 unsuccessful attempts. Monophasic truncated exponential 10-msec defibrillation shocks (0.5-35 J) were delivered through an anodal patch on the right atrium and a cathodal patch on the left ventricular apex. In all cases of unsuccessful defibrillation, at least one ventricular site could be clearly identified that failed to be defibrillated. In cases of successful defibrillation two distinct patterns were observed: 1) complete annihilation of fibrillating activity at all sites or 2) nearly complete cessation of fibrillating activity with a single area of persistent fibrillation that subsequently self-extinguished within one to three activations. This single site in the second form of successful defibrillation was located in the region of minimum voltage gradient produced by the defibrillating waveform and was occasionally accompanied by dynamic encapsulation with refractory tissue as a result of a wavefront emanating from a region that had undergone successful defibrillation. These results support the hypothesis that a critical mass of myocardium must be affected for successful defibrillation and that unsuccessful defibrillation is always accompanied by residual fibrillating activity in at least one site. The results also demonstrate that the size of the critical mass required for successful defibrillation can be less than 100%.