Intravenous lorcainide and lidocaine were administered to 25 patients with symptomatic ventricular tachyarrhythmias in a randomized single-blind crossover study. Prior to drug therapy, each patient underwent 48 hours of ambulatory monitoring and exercise testing on a motorized treadmill. At the completion of baseline studies, patients were randomized to receive either lidocaine or lorcainide intravenously. The dose of lidocaine was 1.0 mg/kg by bolus followed by an infusion of 2 to 3 mg/min. The dose of lorcainide was 2.0 mg/kg followed by a constant infusion of 200 to 300 mg over 24 hours. Two patients developed side effects on lidocaine and the infusion was discontinued prior to evaluation of efficacy. Of the remaining 23 patients, 11 (48%) had their arrhythmia controlled, defined as a greater than 90% reduction in repetitive forms (couplets and ventricular tachycardia) and a 50% reduction in ventricular premature beats. Lorcainide was effective in 8 of the 25 patients (32%). There was no correlation between the effect of lidocaine and the response to lorcainide (p = NS). Oral lorcainide therapy was administered to 17 patients who were free of side effects during the intravenous infusion. The oral drug was effective in nine patients (53%), five of whom had responded to the intravenous drug, and was ineffective in eight, seven of whom were also unresponsive to intravenous lorcainide. The intravenous drug predicted the response to the oral form in 71% of patients, but this was not statistically significant. Side effects occurred in 10 patients (59%) and were primarily neurologic.(ABSTRACT TRUNCATED AT 250 WORDS)