The effects of procainamide and dofetilide (pure Class III antiarrhythmic agent) on the signal-averaged ECG (SAECG) were examined in relation to the results of programmed ventricular stimulation studies in 25 patients with inducible sustained monomorphic ventricular tachycardia. Procainamide prolonged significantly the total QRS and low amplitude signal durations (140 +/- 31 msec vs 166 +/- 48 msec, P < 0.0001; 50 +/- 25 msec vs 65 +/- 38 msec, P < 0.002, respectively) whereas the root mean square voltage of the last 40 msec of the QRS complex was significantly reduced (22 +/- 21 microV vs 13 +/- 12 microV, P < 0.006). Procainamide was effective (prevention of the inducibility of sustained ventricular tachycardia or prolongation of the cycle length of ventricular tachycardia by > 100 msec) in 15 of 27 drug trials. Of the procainamide induced SAECG changes, the fractional prolongation of the total QRS duration was the best parameter that identified effectively treated patients (24% +/- 16% in responders vs 10% +/- 11% in nonresponders, P < 0.014). A fractional prolongation of the total QRS duration by > 15% identified effectively treated patients with a sensitivity of 87%, specificity of 81%, and an overall predictive accuracy of 84%. Dofetilide did not change the SAECG, and no SAECG parameter predicted the results of programmed ventricular stimulation. The effects of both drugs on the spectral analysis (area ratios) and on the spectral temporal mapping (the values of normality factor) of the SAECG were not consistent. In conclusion, antiarrhythmic efficacy of procainamide can be predicted by the degree of drug induced prolongation of the signal-averaged QRS complex.(ABSTRACT TRUNCATED AT 250 WORDS)