Summary The 2000 guidelines for cardiopulmonary resuscitation (CPR) recommend up to three sequential shocks for persistent ventricular fibrillation (VF). We hypothesized that the time consumed for repetitive rhythm analyses and recharging of the capacitor compromises the success of the second and third shock of each sequence. In 60 domestic pigs, VF was electrically induced and untreated for 7 min. After 1 min of CPR, which includes precordial compression and ventilation, up to three sequential shocks were delivered. All animals were resuscitated. For purposes of the present study we determined the outcomes of the first, the second, and the third shock of each sequence during attempted defibrillation. Our criterion of success was the restoration of spontaneous circulation (ROSC). Forty-eight of the 60 animals (80%) attained ROSC after the first shock, 9/60 (15%) after the second shock, and only 3/60 (5%) after the third shock. In confirmation of the earlier observations, ROSC was highly predictive when the coronary perfusion pressure (CPP) exceeded 12 mmHg and end-tidal CO 2 (ETCO 2) exceeded 11 mmHg. However, these criteria were never achieved after the second shock. The present study supports the rationale of delivering only a single shock, or at the most two shocks, prior to resuming chest compression, to re-establish the threshold levels of CPP and ETCO 2 before delivery of a subsequent electrical shock.