Amiodarone produces use-dependent block of cardiac sodium channels in vitro. This study assessed whether similar use-dependent block occurred in 19 patients with coronary artery disease and inducible, sustained, monomorphic ventricular tachycardia treated with amiodarone. Beat-to-beat measurements of ventricular paced QRS durations during 12-beat trains at cycle lengths of 700, 600, 400 and 300 ms were analyzed at a baseline antiarrhythmic drug-free study and after 2 and 10 weeks of amiodarone therapy. At the drug-free study, there were no significant changes in paced QRS durations within the 12-beat trains at any pacing cycle lengths. After 2 and 10 weeks of amiodarone therapy, progressive prolongation of paced QRS durations occurred over the 12-beat trains at pacing cycle lengths of 600, 400 and 300 ms (p less than 0.05). Significant changes in QRS duration were not observed at a pacing cycle length of 700 ms. This progressive prolongation in QRS duration can be fitted as a function of beat number to a monoexponential equation and occurred with an onset time constant of 1.02 +/- 0.41 beats (306 +/- 122 ms) at a pacing cycle length of 300 ms. The magnitude of QRS prolongation increased as the pacing cycle length was shortened. The magnitudes of QRS prolongation were similar after 2 and 10 weeks of amiodarone therapy. In conclusion, use-dependent prolongation in QRS duration occurs at rapid pacing cycle lengths in humans receiving amiodarone.