Cohen TJ, Chin MC, Oliver DG, Scheinman MM, Griffin JC. University of California, San Francisco. Pacing Clin Electrophysiol 1993 Jun;16(6):1285-92. Ventricular fibrillation (VF) that fails to respond to transthoracic defibrillation leaves the clinician with few alternatives. The purpose of this study was to develop a technique of rescue defibrillation by use of transesophageal electrodes. Fourteen anesthetized dogs (20-30 kg) were investigated in this study. Two electrodes (300 mm2) were mounted 8 cm apart on an esophageal probe and inserted approximately 40 cm from the mouth. VF was induced using AC current delivered to the myocardium. Defibrillation was then performed between the distal electrode (anode) and anterior skin patch (cathode). After 15 seconds of induced VF, transesophageal and transthoracic defibrillation thresholds (DFTs) were determined in random order. The esophageal DFT (90 +/- 15 joules) tended to be lower than the transthoracic DFT (115 +/- 35 joules), though this difference was not statistically significant. One dog could not be defibrillated by transthoracic defibrillation but responded to transesophageal defibrillation. Esophageal electrodes were also useful for arrhythmia discrimination and ventricular pacing (pacing threshold of 38 +/- 5 mA at a pulse duration of 2.5 msec). Following transesophageal DFT determination, in ten dogs (total energy of 600 +/- 150 joules), acute esophageal histopathology demonstrated mild to severe focal injury to the mucosa and/or muscular layers. However, esophagi in four chronic dogs (total energy of 470 +/- 110 joules) showed no gross evidence of mucosal damage, perforation, or stricture 4 weeks following defibrillation. Histopathology showed only focal myocyte atrophy and repair. As a last resort, transesophageal defibrillation was performed in the emergency room on four patients with out-of-hospital refractory VF who failed > 6 high energy transthoracic shocks. Transesophageal defibrillation successfully terminated VF in each patient in spite of > 50 mins of cardiac arrest, however none of the patients survived the initial resuscitation. In conclusion, transesophageal defibrillation is as effective as transthoracic defibrillation in a canine model and safe up to a total of 600 Joules. Preliminary clinical trials suggest that this method results in conversion from VF when transthroacic defibrillation fails.