The ultimate aim of defibrillation testing is to predict consistent defibrillation. This study tested the hypothesis that defibrillation success could be predicted from a single measurement of defibrillation threshold. We measured defibrillation threshold by using three patch electrodes and a standard protocol intraoperatively in 49 patients undergoing arrhythmia surgery. Each patient was then assigned to one of five energy subgroups (0.5, 1.0, 1.5, 2.0, or 2.5 times defibrillation threshold) for a single shock (followed by a rescue shock if necessary) for a subsequent ventricular fibrillation episode. A curve relating percent success to energy was then constructed for the group. Defibrillation threshold averaged 4.7 2.98 J for the group (mean + SD). There was a curvilinear relation between the energy of the defibrillation threshold ratio test shock and percent success: 33.3%, 58.3%, 81.8%, 91.7%, and 100% at mean defibrillation threshold ratios of 0.56 0.14, 1.02 0.07, 1.53 0.14, 1.88 0.09, and 2.60 0.14, respectively. We conclude that consistent defibrillation is predictable from a single measurement of defibrillation threshold. Furthermore, for an individual patient, a safety margin of 2.6 times defibrillation threshold should approximate 100% successful defibrillation for a single test shock.