BACKGROUND: Unintended intravenous injection of bupivacaine causes severe cardiovascular complication, which is known for its difficulty in resuscitation. This study was performed to evaluate the effects of pretreatment with midazolam and droperidol in the cardiac toxicity caused by intravenous infusion of bupivacaine. METHODS: Thirty rabbits were divided into three groups; saline- as a control, midazolam, and droperidol pretreated group. We observed the time intervals for the arrhythmia, 25% and 50% reduction in baseline mean arterial blood pressure, and arrest. We also checked the dose of infused bupivacaine to be required for arrest during continuous intravenous infusion of bupivacaine at the rate of 1 mg/kg/min. RESULTS: The onset of dysrhythmia and the time to 50% reduction in baseline mean arterial blood pressure and arrest were significantly more delayed in the midazolam group than the control group (P < 0.05). With respect to the time to 25%, 50% reduction in baseline mean arterial blood pressure and arrest, the data of the droperidol group was significantly shorter than that of the control group (P < 0.05). CONCLUSIONS: Droperidol pretreatment hastened bupivacaine induced cardiac arrest in rabbits. Midazolam pretreatment exerted protective effects on arrhythmia and cardiac arrest. Thus midazolam would be a preferable agent as a supplement for regional anesthesia using bupivacaine.