The effect of intravenous lidocaine, 2 mg/kg, and bretylium, 5 mg/kg, on defibrillation threshold (DFT) was investigated in alpha-chloralose anesthetized dogs undergoing conventional closed chest cardiopulmonary resuscitation (CPR) following induced ventricular fibrillation. Ventricular fibrillation was induced electrically and CPR was performed by a pneumatic device set to compress the chest 60 times and inflate the lung 12 times a minute. Defibrillation was achieved using underdamped sinusoidal current shocks from a special defibrillator which allowed determination of delivered energy. The DFT was defined as the peak current which defibrillated, but no more than 20% higher than a current which did not defibrillate. All DFTs were obtained within 5 min of CPR. The mean +/- SD current and energy thresholds required for defibrillation during lidocaine-CPR (seven dogs) were 17.0 +/- 8.9 A and 53.0 +/- 40.7 J as compared to 12.5 +/- 6.2 A and 34.3 +/- 30.7 J, respectively during control-CPR (P less than 0.05). The mean +/- SD current and energy thresholds during bretylium-CPR were 11.0 +/- 3.4 A and 24.1 +/- 1.3 J as compared to 11.8 +/- 1.7 A and 29.4 +/- 9.6 J, respectively, during control-CPR (NS). These results show that lidocaine acutely elevated defibrillation threshold whereas bretylium did not produce such an effect. The effect on DFT along with other pharmacologic properties should be considered when lidocaine or bretylium is used in the setting of cardiac arrest and CPR.