The incidence of proarrhythmic effect of antiarrhythmic drugs (AADs) in not well documented. The aim of the study was to assess the frequency od proarrhythmia in patients with ischemic heart disease (IHD) and ventricular premature beats (VPBs) in whom various class I, II and III AADs were tested by 24-h Holter ecg. All data were collected in a prospective manner. Our material consisted of 639 patients with IHD and VPBs (Lown's grade 2-5). The mean age was 53 years. 63% of patients had previously myocardial infarction. 15% and 3% had documented ventricular tachycardia (VT) or ventricular fibrillation (VF), (VF), respectively. Baseline Holter monitoring revealed repetitive VPBs or R on T phenomenon in 64% of cases. Plasma electrolytes level, renal and hepatic function were normal. Antiarrhythmic therapy was guided by repeated 24-h Holter ecg on a maintenance dosage of the drug. Propranolol was a drug of first choice. Disopyramide or mexiletine was added if propranolol alone was found to be ineffective in control Holter ecg. Amiodarone was a drug of a next choice. It was allowed modify the treatment in patients with contraindication to propranolol, clinical VT/VF or high grade VPBs. 794 drug tests were conducted. Number of tests/patient ranged 1-4. The following AADs were assessed: propranolol (352 tests), disopyramide (280 tests), mexiletine (73 tests), amiodarone (89 tests). Aggravation of arrhythmia was defined by modified criteria proposed by Velebit: 1) greater than or equal to 4-fold increase in VPBs, 2) greater than or equal to 10-fold increase in couplets or salvoes, 3) occurrence of VT. Proarrhythmia was recognized when at least one criterion was present.(ABSTRACT TRUNCATED AT 250 WORDS)