Summary To assess the response to encainide in patients with malignant arrhythmias, we treated 22 patients with recurrent ventricular tachycardia (VT) or fibrillation (VF) refractory to an average of 5.7 drugs. Thirteen patients (599%) showed a favorable in-hospital response, and 11 of 22 (50%) maintained a favorable antiarrhythmic response during a median follow-up of 13.5 months (range 0.3–25.5). Encainide was well tolerated. Six (46%) of 13 patients responded to intravenous encainide with suppression of VT by programmed ventricular stimulation or continuous monitoring: oral encainide has maintained arrhythmia control. In six of seven others, cycle length of VT was lengthened. Initial therapy with oral drug yielded five complete and two partial responses, assessed by monitoring, among nine other patients. Of three outpatient failures, only one was an inpatient responder. Proarrhythmic responses were also noted and included spontaneous increases in ectopy (three patients), easier induction of VT spontaneously or at programmed stimulation (three patients), facilitation of exercise VT (one patient), and syncope due to VT/VF (one patient). After intravenous therapy, plasma encainide concentration averaged 773 ng/ml (range 476–1.279. n = 4), with low or negligible metabolites. During chronic oral therapy, encainide concentrations were low and variable, averaging 67.5 ng/ml (< 10–585: n = 23, 10 patients): metabolite levels were higher and less variable: O-demethyl encainide, mean 198 ng/ml (61–882): 3-methoxy-o-demethyl encainide, mean 128 ng/ml (39–379). Active metabolite(s) may thus be more important than parent drug during chronic therapy. In summary, encainide may provide successful therapy in approximately half of patients with malignant, refractory ventricular arrhythmias. Inpatient response predicts outpatient results. Encainide, like other antiarrhythmies, has proarrhythmic potential, suggesting a carefully monitored initial approach.