Cardiopulmonary resuscitation in patients with severe sepsis and septic shock is challenging and usually unsuccessful. The aim of the present study is to describe our swine model of cardiac arrest and resuscitation in severe sepsis and septic shock. In this prospective randomized animal study, 10 healthy female Landrace-Large White pigs with an average weight of 20 ± 1 kg (aged 19 - 21 weeks) were the study subjects. Septicemia was induced by an intravenous infusion of a bolus of 20-mL bacterial suspension in 2 min, followed by a continuous infusion during the rest of the experiment. After septic shock was confirmed, the animals were left untreated until cardiac arrest occurred. All animals developed pulseless electrical activity between the fifth and sixth hours of septicemia, whereas five (50%) of 10 animals were successfully resuscitated. Coronary perfusion pressure was statistically significantly different between surviving and nonsurviving animals. We found a statistically significant correlation between mean arterial pressure and unsuccessful resuscitation (P = 0.046), whereas there was no difference in end-tidal carbon dioxide (23.05 ± 1.73 vs. 23.56 ± 1.70; P = 0.735) between animals with return of spontaneous circulation and nonsurviving animals. During the 45-min postresuscitation monitoring, we noted a significant decrease in hemodynamic parameters, although oxygenation indices and lactate clearance were constantly increased (P = 0.001). This successful basic swine model was for the first time developed and may prove extremely useful in future studies on the periarrest period in severe sepsis and septic shock.