Abstract Transient entrainment was used to test the hypotheses that 1) procainamide prolongs the cycle length of ventricular tachycardia in patients with coronary artery disease because it has a preferential effect on the reentrant tachycardia circuit, and 2) regions of slow conduction in the reentrant circuit are more susceptible to the effect of procainamide than are other areas of the ventricles. In five patients with prior myocardial infarction, sustained ventricular tachycardia with identical QRS configuration was inducible before and after intravenous infusion of procainamide. Transient entrainment of ventricular tachycardia was demonstrated at two or more cycle lengths by rapid pacing in the baseline state and after procainamide. Rapid pacing was performed from the same site during sinus rhythm at the cycle lengths that demonstrated transient entrainment of ventricular tachycardia. The conduction interval to the transiently entrained site during ventricular tachycardia (orthodromic interval) was compared with the conduction interval to the same site during pacing in sinus rhythm (antidromic interval). The mean tachycardia cycle length increased by 27% after procainamide administration (p = 0.002). The antidromic conduction intervals were prolonged by 9 % (p = 0.06) compared with a 28% increase in the mean orthodromic conduction interval (p = 0.002). The difference between the orthodromic and antidromic conduction intervals increased by 40% (p = 0.003). Prolongation of the tachycardia cycle length after procainamide administration correlated positively with increases in the orthodromic conduction intervals (r = 0.94, p = 0.02) but not with changes in the antidromic intervals (r = −0.08, p = NS). The effect of procainamide on the difference between orthodromic and antidromic conduction intervals correlated strongly with changes in the cycle length of venetricular tachycardia (r = 0.97 p = 0.006). Thus, procainamide has a preferential effect on the reentrant circuit of ventricular tachycardia compared with other areas of ventricular myocardium. In addition, these results are compatible with the hypothesis that the effect of procainamide is more marked in regions of conduction delay in the reentrant circuit.