The efficacy of lidocaine during myocardial reperfusion in coronary artery bypass surgery was evaluated in 20 patients randomly assigned to a control group (n = 10) or to receive lidocaine, 1 mg/kg intravenously 5 min before aortic unclamping and cardiac reperfusion, followed by infusion at 40 micrograms X kg-1 X min-1 (n = 10). We recorded ECG leads II and V5 continuously, and number, energy, and current of direct current (DC) shocks starting at 1 joule. The number of low energy DC shocks to sustained defibrillation (5.5 +/- 2.0 vs 3.5 +/- 2.0, mean +/- SD, P less than 0.05) decreased significantly with lidocaine infusion. The energy (11.0 +/- 6.3 vs 5.6 +/- 3.9 joules, P less than 0.05) and current (12.7 +/- 4.2 vs 8.9 +/- 4.7 amperes, not significant) likewise decreased with lidocaine infusion. Energy and current for the first successful shock, although lower in the lidocaine group, were not statistically significantly lower than in the control group. Initial reperfusion rhythm was not influenced by lidocaine. Plasma electrolyte levels, arterial blood gas tensions, myocardial temperature, and surgical technique--factors known to influence defibrillation--were similar in all patients. Administration of lidocaine during myocardial reperfusion allows defibrillation with fewer DC shocks of lower energy and current.