In a porcine model, we compared the efficacy of epinephrine, vasopressin, or the combination of epinephrine and vasopressin with that of saline placebo on the survival rate after bupivacaine-induced cardiac arrest. After the administration of 5 mg/kg of a 0.5% bupivacaine solution IV, ventilation was interrupted for 3 +/- 1 min (mean +/- SD) until asystole occurred. Cardiopulmonary resuscitation (CPR) was initiated after 1 min of cardiac arrest. After 2 min of CPR, 28 animals received, every 5 min, epinephrine; vasopressin; epinephrine combined with vasopressin; or placebo IV. Three minutes after each drug administration, up to 3 countershocks (3, 4, and 6 J/kg) were administered; all subsequent shocks were 6 J/kg. Blood was drawn throughout the experiment for the determination of plasma bupivacaine concentration. hi the vasopressin/epinephrine combination group, all pigs survived (P < 0.01 versus placebo); in the vasopressin group 5 of 7, in the epinephrine group 4 of 7, and in the placebo group none of 7 swine survived. The plasma concentration of total bupivacaine showed no significant difference among groups. In this model of bupivacaine-induced cardiac arrest, CPR with a combination of vasopressin and epinephrine resulted in significantly better survival rates than in the placebo group.