The antiarrhythmic efficacy and pharmacokinetics of N-acetylprocainamide (NAPA), the major metabolite of procainamide, were investigated in 23 patients with chronic, high frequency ventricular ectopic depolarizations. An extensive trial design incorporated the approaches of (1) generation of dose-response relations, (2) randomized crossover, and (3) prolonged electrocardiographic monitoring. Seven patients with reproducible suppression of arrhythmias (70 percent or greater reduction in frequency) were thus identified. The mean plasma concentration of acecainide associated with efficacy was 14.3 micrograms/ml (range 9.4 to 19.5) and with side effects (primarily gastrointestinal) was 22.5 micrograms/ml (10.6 to 37.9). The antiarrhythmic response to procainamide did not predict response to acecainide; this finding implies that estimates of the antiarrhythmic contribution of acecainide concentrations achieved during long-term procainamide therapy are unlikely to be meaningful in a given person. The mean half-life of elimination after a single 500 mg dose of acecainide was 7.5 hours; this had prolonged significantly (p < 0.05) to 10.3 hours after higher dosages. No variable examined (including acetylator phenotype) was found to be a predictor of responsiveness to acecainide. Outpatient therapy (2 to 20 months) was not associated with the development of antinculear antibodies or the lupus syndrome; one patient's procainamide-induced arthritis resolved during therapy. Acecainide, unlike procainamide, is an agent whose pharmacokinetics allow long-term therapy on a practical schedule. It is effective in a subset of patients with ventricular arrhythmias yet appears much less likely to induce the lupus syndrome seen with the parent compound.