The electrophysiological and haemodynamic effects of lidocaine (100 mg) and ajmaline (50 mg) were evaluated while attempting to interrupt sustained ventricular tachycardia. The study was performed as a prospective, non-blinded, randomized investigation in 61 patients. Lidocaine terminated ventricular tachycardia in four of 31 patients, ajmaline in 19 of 30 patients (P less than 0.001). QRS and RR intervals during ventricular tachycardia were prolonged by ajmaline from 164 +/- 28 ms to 214 +/- 49 ms and from 371 +/- 86 ms to 479 +/- 137 ms (P less than 0.001), respectively; lidocaine did not influence these parameters. The duration of the return cycles after termination of ventricular tachycardia did not differ between the two groups. Lidocaine did not change cardiac output during ventricular tachycardia whereas cardiac output increased significantly under ajmaline from 3.5 +/- 1.21.min-1 to 5.5 +/- 1.91.min-1 (P less than 0.001). It is concluded that anti-arrhythmic agents such as ajmaline, which slow conduction velocity and prolong refractoriness, are more effective than lidocaine in the medical treatment of haemodynamically stable, sustained ventricular tachycardia.