Studies of isolated heart muscle and canine models of myocardial ischemia have demonstrated that lidocaine slows conduction in abnormal but not in normal tissues. To determine lidocaine's effects on intraventricular conduction (IVENT) in patients with coronary artery disease (CAD), we studied this agent in seven patients following left anterior descending coronary artery (LAD) bypass surgery. Epicardial electrodes were placed on the right atrium, left ventricle (LV) in the distribution of the LAD, and on the right ventricle (RV). On postoperative day 7, lidocaine was administered as 100 mg bolus followed by 4 mg/minute infusion for 2 hours. At constant atrially paced rate, bipolar electrograms were recorded from the LV and RV for the 2 hours of infusion and for 2 hours after discontinuation of infusion. Conduction intervals were measured from the earliest onset of QRS in three simultaneously recorded surface ECG leads to the major deflection of the electrogram from each ventricle. At peak effect, with mean lidocaine level of 2.7 +/- 0.5 mg/ml, lidocaine slowed LV conduction by a mean of 6 +/- 1 msec (14 +/- 2%) (p less than 0.001) and in the RV by 1 +/- 0.3 msec (4 +/- 1%) (p less than 0.01), QRS duration changed 1 +/- 1 msec) 1 +/- 1%) (NS). The values returned to baseline within 2 hours after discontinuation of lidocaine infusion. The difference in lidocaine's effect between the diseased LV and the normal RV was significant (p less than 0.001).