Combination therapy for cardiac arrhythmias

Development of an enzyme-linked immunosorbent assay for the quantification of amiodarone

A sensitive and specific enzyme-linked immunosorbent assay for an antiarrhythmic drug, amiodarone (AMI), was developed, which is capable of measuring levels as low as 16 ng/ml. Anti-AMI antibody was obtained by immunizing rabbits with an antigen conjugated with bovine serum albumin using diazotized 4-amino-1-(2-diethylaminoethoxy)-2,6-diiodobenzene. Enzyme labeling of AMI with beta-D-galactosidase was similarly performed using a diazotized 4-amino-1-(2-diethylaminoethoxy)-2,6-diiodobenzene. This enzyme-linked immunosorbent assay was specific for AMI and showed a very slight cross-reactivity (1.25%) with its major metabolite, mono-N-desethylamiodarone. The values of the AMI concentrations measured by this assay were in good correlation to those by HPLC. Its analytical applicability was demonstrated by a kinetic study with human liver microsomes. The enzyme-linked immunosorbent assay should be a valuable tool in therapeutic drug monitoring and pharmacokinetic studies of AMI.

High-performance liquid chromatographic assay for amiodarone N-deethylation activity in human liver microsomes using solid-phase extraction

A selective and sensitive assay for amiodarone N-deethylation activity in human liver microsomes by high-performance liquid chromatography (HPLC) with UV detection is reported. The extraction of desethylamiodarone from incubation samples was performed by means of an original solid-phase extraction (SPE) procedure using a polymeric reversed-phase sorbent (Oasis HLB). The method was validated for the determination of desethylamiodarone with respect to specificity, linearity, precision, accuracy, recovery, limit of quantitation and stability. Amiodarone N-deethylation activity from low to high substrate concentrations using human liver microsomes was precisely determined without a concentration step. This method is applicable to the study in vitro of the metabolism of amiodarone.

Antiarrhythmic agents: drug interactions of clinical significance

The management of cardiac arrhythmias has grown more complex in recent years. Despite the recent focus on nonpharmacological therapy, most clinical arrhythmias are treated with existing antiarrhythmics. Because of the narrow therapeutic index of antiarrhythmic agents, potential drug interactions with other medications are of major clinical importance. As most antiarrhythmics are metabolised via the cytochrome P450 enzyme system, pharmacokinetic interactions constitute the majority of clinically significant interactions seen with these agents. Antiarrhythmics may be substrates, inducers or inhibitors of cytochrome P450 enzymes, and many of these metabolic interactions have been characterised. However, many potential interactions have not, and knowledge of how antiarrhythmic agents are metabolised by the cytochrome P450 enzyme system may allow clinicians to predict potential interactions. Drug interactions with Vaughn-Williams Class II (beta-blockers) and Class IV (calcium antagonists) agents have previously been reviewed and are not discussed here. Class I agents, which primarily block fast sodium channels and slow conduction velocity, include quinidine, procainamide, disopyramide, lidocaine (lignocaine), mexiletine, flecainide and propafenone. All of these agents except procainamide are metabolised via the cytochrome P450 system and are involved in a number of drug-drug interactions, including over 20 different interactions with quinidine. Quinidine has been observed to inhibit the metabolism of digoxin, tricyclic antidepressants and codeine. Furthermore, cimetidine, azole antifungals and calcium antagonists can significantly inhibit the metabolism of quinidine. Procainamide is excreted via active tubular secretion, which may be inhibited by cimetidine and trimethoprim. Other Class I agents may affect the disposition of warfarin, theophylline and tricyclic antidepressants. Many of these interactions can significantly affect efficacy and/or toxicity. Of the Class III antiarrhythmics, amiodarone is involved in a significant number of interactions since it is a potent inhibitor of several cytochrome P450 enzymes. It can significantly impair the metabolism of digoxin, theophylline and warfarin. Dosages of digoxin and warfarin should empirically be decreased by one-half when amiodarone therapy is added. In addition to pharmacokinetic interactions, many reports describe the use of antiarrhythmic drug combinations for the treatment of arrhythmias. By combining antiarrhythmic drugs and utilising additive electrophysiological/pharmacodynamic effects, antiarrhythmic efficacy may be improved and toxicity reduced. As medication regimens grow more complex with the aging population, knowledge of existing and potential drug-drug interactions becomes vital for clinicians to optimise drug therapy for every patient.

Intravenous amiodarone

Intravenous amiodarone was approved in 1995 for the treatment of malignant and resistant ventricular arrhythmia. Although it is an "old drug," much has been learned recently about this complex drug and its application in a variety of cardiac arrhythmias. The objectives of this review were to summarize what is known about intravenous amiodarone, including its pharmacologic and electrophysiologic effects, to review its efficacy for the treatment of patients with highly malignant ventricular arrhythmia and to provide specific information about its clinical use for this and other indications. The studies that were reviewed were selected on the basis of time published (from 1983 to 1995) and the completeness of information provided regarding patient clinical characteristics, drug dosing and methods of evaluation, efficacy analyses, long-term follow-up and complications. The full data from the three controlled trials that formed the basis of the drug's approval are contained in published reports that were also extensively reviewed. Intravenous amiodarone has demonstrable efficacy for the treatment of frequently recurrent destabilizing ventricular tachycardia and ventricular fibrillation, with suppression rates of 63% to 91% in uncontrolled trials. The three pivotal trials confirmed these findings and demonstrated a dose-response relation, with at least comparable efficacy to bretylium, a drug with a similar indication. The safety profile has also been well described; cardiovascular adverse effects are the most frequent, especially hypotension. Intravenous amiodarone is a useful addition to the drugs available for the treatment of patients with very severe ventricular arrhythmia. Its use in patients with other rhythm disorders appears promising, but final recommendations must await development of definitive data from ongoing clinical trials.

Efficacy of antiarrhythmic drugs in patients with arrhythmogenic right ventricular disease. Results in patients with inducible and noninducible ventricular tachycardia

BACKGROUND

Ventricular tachyarrhythmias are the major clinical manifestation of arrhythmogenic right ventricular disease. Although antiarrhythmic therapy has been widely advocated, there is only limited information available on the efficacy of antiarrhythmic drugs in these patients.

METHODS AND RESULTS

The short- and long-term efficacies of various antiarrhythmic agents were retrospectively and prospectively analyzed in 81 patients (mean age, 39 +/- 14 years; range, 16-68 years; 61.7% males) with arrhythmogenic right ventricular disease. In 42 patients with inducible ventricular tachycardia during programmed ventricular stimulation, the following efficacy rates were obtained: class Ia and Ib drugs (n = 18), 5.6%; class Ic drugs (n = 25), 12%; beta-blockers (n = 8), 0%; sotalol (n = 38), 68.4%; amiodarone (n = 13), 15.4%; verapamil (n = 5), 0%; and drug combinations (n = 26), 15.4%. Only one of the 10 patients not responding to sotalol was treated effectively by amiodarone, whereas the remaining nine patients proved to be drug refractory toward all other drugs tested (3.8 +/- 2.3 drugs, including amiodarone in five cases) and underwent nonpharmacological therapy. During a follow-up of 34 +/- 25 months, three of the 31 patients (9.7%) discharged on pharmacological therapy had nonfatal recurrences of ventricular tachycardia after 0.5, 51, and 63 months, respectively. In 39 patients with noninducible ventricular tachycardia during programmed ventricular stimulation, the following efficacy rates were observed: class Ia and Ib drugs (n = 16), 0%; class Ic agents (n = 23), 17.4%; beta-blockers (n = 7), 28.6%; sotalol (n = 35), 82.8%; amiodarone (n = 4), 25%; verapamil (n = 24), 50%; and drug combinations (n = 11), 9.1%. During a follow-up of 14 +/- 13 months, four of 33 patients (12.1%) discharged on antiarrhythmic drugs had nonfatal relapses of their clinical ventricular arrhythmia.

CONCLUSIONS

Thus, in arrhythmogenic right ventricular disease, sotalol proved to be highly effective in patients with inducible as well as noninducible ventricular tachycardia. Patients with inducible ventricular tachycardia not responding to sotalol are likely to not respond to other antiarrhythmic drugs and should be considered for nonpharmacological therapy without further drug testing. Amiodarone did not prove to be more effective than sotalol and may not be an alternative because of frequent side effects during long-term therapy, especially in young patients. Verapamil and beta-blockers were effective in a considerable number of patients with noninducible ventricular tachycardia and may be a therapeutic alternative in this subgroup. Class I agents appear to be rarely effective in the treatment of both inducible and noninducible ventricular tachycardia in arrhythmogenic right ventricular disease.

Evaluation of disopyramide and mexiletine used alone and in combination for ventricular arrhythmias in patients with and without overt heart disease

The efficacies and side effects of disopyramide and mexiletine used alone and in combination were assessed in 29 patients with chronic ventricular arrhythmias. In combination therapy, one half or two thirds of the conventional doses of each drug were administered. Each patient underwent Holter electrocardiographic monitoring during 4 different periods: baseline, disopyramide alone, mexiletine alone and combination of the two drugs. The mean baseline number of ventricular premature complex per hour was 783 +/- 521 (mean +/- SD), which was significantly reduced with all three therapies. Disopyramide alone significantly reduced the ventricular premature complex frequency in patients with organic heart disease (P less than 0.05), but did not significantly reduce the ventricular premature complex frequency in patients with no apparent heart disease. In contrast, mexiletine alone significantly decreased the ventricular premature complex frequency in no apparent heart disease patients (P less than 0.05), but did not significantly reduce the ventricular premature complex frequency in organic heart disease patients. With disopyramide alone, patients having a significant reduction in ventricular premature complexes (greater than or equal to 83% reduction in ventricular premature complexes) or elimination of ventricular tachycardias tended to be more frequently found in organic heart disease than in no apparent heart disease. The opposite was observed with mexiletine alone. QTc interval with disopyramide alone was significantly prolonged, and the prematurity index of ventricular premature complexes was significantly lowered as compared to mexiletine alone or combination therapy (P less than 0.01 for disopyramide versus mexiletine; P less than 0.05 for disopyramide versus combination therapy). During combination therapy, no patients withdrew from the study due to side effects. However, 3 patients receiving single drug therapy withdrew from the study due to severe side effects. Consequently, disopyramide is suggested to be more effective on ventricular premature complexes in organic heart disease than in no apparent heart disease patients, whereas the opposite was true for mexiletine. A combination of disopyramide and mexiletine in smaller doses may provide almost the same or enhanced antiarrhythmic effects, no aggravation of electrocardiographical parameters and less incidence of side effects when compared to the conventional dose of each drug alone.

Combined effects of different class I antiarrhythmic agents on maximum rate of depolarization (Vmax) of action potentials in guinea-pig papillary muscles

The combination of two different kinds of class I antiarrhythmic agents (class Ia, Ib, or Ic) was examined with regard to their effects on the maximum rate of depolarization (Vmax) of action potentials in guinea-pig papillary muscles. The combinations of disopyramide plus lidocaine, disopyramide plus mexiletine, mexiletine plus flecainide, and disopyramide plus flecainide were employed to study their effects on use-dependent block of Vmax. All the combinations increased the percent of use-dependent block at most of the frequencies employed (0.1-3.3 Hz) as compared to the effects of the single use of either drug, but no decrease in use-dependent block was found with any of the combinations. The time courses of the development of use-dependent block by disopyramide, lidocaine, and mexiletine were best expressed by two exponential functions, whereas those by flecainide were expressed by a single exponential function. Disopyramide plus lidocaine and disopyramide plus mexiletine produced increases in the time constant of the fast component of the block (tau f), the fast fraction of the block (Af), and the ratio of the fast to the slow fraction (Af/As). Mexiletine plus flecainide increased tau f, Af, and As; whereas disopyramide plus flecainide caused no changes in the kinetic parameters of use-dependent block. These results suggest that there may be diverse modes of interaction between the drug and the Na+ channel, and the combination of two different types of the drug may sometimes provide different effects on the fast and slow components of the use-dependent block of Vmax.

Effects of combined use of class I antiarrhythmic agents on Vmax of guinea-pig ventricular muscles

The effects of the combined use of class-I antiarrhythmic drugs on the resting potentials (RP), amplitude of action potential (AMP), and Vmax of the action potential were investigated in guinea-pig ventricular papillary muscles that were superfused with oxygenated Krebs-Ringer solution at 35 degrees C. Disopyramide (40 microM) reduced Vmax to 68.6 +/- 3.1% (mean +/- SE, n = 5) of the control with minimal changes in RP and AMP when preparations were stimulated at 1 Hz. The addition of mexiletine (20 microM) to the solution containing disopyramide (40 microM) caused a minimal reduction of Vmax (less than 5%) for the stimulation of 1 Hz, but a significant reduction of Vmax (13% p less than 0.05) when stimulation was increased to 2 Hz. This amount of the reduction is compatible with that obtained by mexiletine alone, suggesting a simple additive Na+ channel inhibition by this drug combination. This additive effect was also observed in the recovery process of Vmax from the use-dependent block induced by train stimuli at 1 Hz. Flecainide (5 microM) reduced Vmax to 58.6 +/- 13.3% (n = 5). The addition of mexiletine to the superfusate with flecainide produced a further depression of 14 +/- 2.6% of Vmax, even at 1 Hz. This depression was larger than that produced by mexiletine, suggesting a synergistic action of the two drugs on the Na+ channel. Such information about the interaction of the class I drug combinations with the Na+ channel may be clinically important.

Efficacy of disopyramide and mexiletine used alone or in combination in the treatment of ventricular premature beats

The efficacy of oral disopyramide and mexiletine used alone or in combination was studied in 75 patients with frequent ventricular premature beats (VPBs). The efficacy was evaluated with 24-hour ambulatory ECG and greater than or equal to 75% reduction in the number of VPBs was defined as effective. When disopyramide or mexiletine were ineffective or not tolerated, the alternative drug was administered and the efficacy was again evaluated. If the single administration of neither drug was effective, the combination of disopyramide and mexiletine was then given. Either disopyramide or mexiletine was effective in 48 patients, and neither drug was effective in 19 patients. In 19 patients unresponsive to both drugs, combination therapy was effective in six patients (32%). Both drugs caused side effects or one drug caused side effects and another drug was ineffective in eight patients. In five out of those patients, we attempted combined therapy with a reduced dosage of those drugs that caused side effects. This therapy was effective in two patients without intolerable side effects. Thus, when the single use of neither disopyramide nor mexiletine single-drug therapy is effective, it is worthwhile to try combination therapy. Also, combination therapy with a reduced dosage of the drugs that caused side effects might be the therapy of choice in patients who have developed dose-dependent side effects.

A prospective comparison of class IA, B, and C antiarrhythmic agents in combination with amiodarone in patients with inducible, sustained ventricular tachycardia

BACKGROUND

Clinical experience suggests that combinations of antiarrhythmic agents provide more effective control of ventricular tachyarrhythmias than does therapy with single agents.

METHODS AND RESULTS

Antiarrhythmic and electrophysiological effects of three class I antiarrhythmic agents, one from each subclass A, B, and C, were assessed in single use and in combination with amiodarone in patients with inducible, sustained ventricular tachycardia that was not suppressed by monotherapy with these agents. Thirty-one patients underwent an electrophysiology test on four occasions: at baseline; after 2-4 days of treatment with quinidine, mexiletine, or encainide; after 2 weeks of treatment with 1,200 mg/day amiodarone; and last, after 2-4 days of treatment with both amiodarone and the previously tested class I agent. The combination of a class I agent and amiodarone prevented the induction of sustained ventricular tachycardia in only one of 31 (3%) patients. Ventricular tachycardia became hemodynamically stable in 11 of 31 (34%) patients because of a marked prolongation in the tachycardia cycle length. It increased from 323 +/- 39 to 423 +/- 84 msec (n = 11, p less than 0.01) by adding encainide to amiodarone therapy, and it showed a tendency to lengthen when quinidine was added to amiodarone (from 373 +/- 77 to 425 +/- 58 msec; n = 10, NS). Each class I agent increased amiodarone-induced depression in myocardial conduction, but the extent of the additional depression seemed to differ among the three subclasses. Ventricular refractoriness was increased by all class I agents when used in combination with amiodarone, although not by mexiletine or encainide when used alone.

CONCLUSIONS

Class I antiarrhythmic agents slow ventricular conduction and increase ventricular refractoriness when used in combination with amiodarone. When amiodarone and class I drugs by themselves do not suppress the induction of ventricular tachycardia, the combination of amiodarone and a class I agent seldom results in noninducibility; however, it often lengthens the ventricular tachycardia cycle length and may render the ventricular tachycardia hemodynamically stable.

The value of oral amiodarone in the treatment of ventricular arrhythmias in heart disease

The antiarrhythmic properties of amiodarone at the ventricular level were discovered in the early 1970s. The unanimously recognised efficacy of amiodarone includes a weak negative inotropic effect and compensatory vasodilatory properties, making amiodarone particularly suitable for treating the potentially malignant arrhythmias associated with organic disease. In a review of 611 hospitalised patients on amiodarone, and 353 patients in whom the drug had been prescribed, over a 52-month period in our 60-bed department, we noted that amiodarone was prescribed in 53% of patients for arrhythmias and in 47% of patients for coronary insufficiency. Ventricular arrhythmias represented 13% of the rhythmic indications. These indications differ from those in the USA. The efficacy (70 to 90%) of amiodarone in ventricular extrasystoles has been shown in open studies. In coronary patients, the antiarrhythmic activity of amiodarone is superior to that of propranolol. However, there has been no controlled study because the need for a loading dosage, and the electrocardiographic effects render such studies difficult. After myocardial infarction, ventricular arrhythmias constitute a significant risk factor independently of prognosis; amiodarone may be useful in this indication, and studies of the European Myocardial Infarction Amiodarone Trial (EMIAT) type will examine its value here. Since 1973, it has been recognised that amiodarone can prevent ventricular tachycardia in 55 to 89% of patients in the clinical situation. After a long-standing controversy, the positive predictive value of programmed stimulation has finally been agreed on. In hypertrophic cardiomyopathy, retrospective studies suggest a reduction in mortality in patients treated with amiodarone. By contrast, the value of amiodarone in dilated cardiomyopathy requires more intensive investigation. We consider amiodarone to be indicated in ventricular arrhythmic complexes, particularly if they are associated with an ejection fraction of less than 35% and/or atrial fibrillation. The value of amiodarone in arrhythmias associated with heart failure needs to be evaluated. In conclusion, amiodarone is a powerful antiarrhythmic agent but, because of the possibility of dose- and duration-dependent side effects, evaluation of the risk: benefit ratio in each indication is needed.

Combination of flecainide and mexiletine for the treatment of ventricular tachyarrhythmias

A combination of oral flecainide and mexiletine was given to 11 patients in whom monotherapy with one of these drugs was ineffective for the suppression of inducible ventricular tachycardia or fibrillation. In eight of 11 studies, combination therapy prevented inducibility of a sustained ventricular tachycardia or resulted in induction of only nonsustained tachycardia (P = 0.0003, when compared to monotherapy). In one patient, a slow ventricular tachycardia was induced. During exercise testing ventricular tachycardia occurred in two of these nine patients, and ventricular fibrillation in another patient. Seven patients received combination on the long term, for a mean of 18 months. One patient had recurrences of ventricular tachycardia which was well tolerated. Another patient had a recurrent episode of ventricular fibrillation, but was successfully resuscitated. Severe congestive heart failure occurred in two patients. ACE inhibitors were given to them and to another four patients. No other important unwanted effects occurred. The combination of mexiletine and flecainide is very effective in suppressing inducible sustained ventricular tachycardia. The efficacy of this combination to prevent recurrences of ventricular tachyarrhythmias is acceptable. Exercise testing is of importance to unmask proarrhythmic effects before discharge from hospital.

Are antiarrhythmic drugs safe?

The indications for antiarrhythmic therapy are far from clearly defined and the choice of treatment is usually based on empiric strategies. Antiarrhythmic agents can have serious side effects. Systemic adverse effects are usually use-related and reversible with withdrawal of the drug. Impairment of left ventricular function is considerable in patients with heart failure. The most important, life-threatening side effect of antiarrhythmic drugs is their proarrhythmic tendency which gives rise to certain concern about their clinical use. Aggravation of arrhythmia often occurs without symptoms, goes unrecognized by the patient, and is exposed only by monitoring, exercise testing, or invasive electrophysiological testing. Patient monitoring with electrolyte measurement, Holter recording, and electrophysiological reassessment can reveal or reduce the proarrhythmic risk but cannot eliminate the problem completely. The institution of antiarrhythmic therapy should be considered in highly symptomatic or life-threatening arrhythmias after careful consideration of the benefit-risk ratio.

The combined electrophysiological effects of lignocaine and sotalol in canine isolated cardiac Purkinje fibres are rate-dependent

1. The frequency-dependent electrophysiological effects of lignocaine, sotalol, and their combination were studied in dog isolated cardiac Purkinje fibres, both at various constant rates of stimulation and following abrupt changes in pacing cycle length. 2. The combined effect of 18 microM lignocaine and 30 microM sotalol selectively lengthened duration of premature action potentials evoked at a diastolic interval of 40 ms (from 172.2 +/- 5.4 to 201.7 +/- 4.9 ms, n = 6, P less than 0.01) without significantly changing the durations of action potentials evoked at the basic cycle length of 500 ms (259.1 +/- 7.7 vs 251.9 +/- 3.9 ms, n = 11). 3. The combination of lignocaine with sotalol, like lignocaine alone, displayed a use-dependent depression of Vmax and revealed a slow component for a recovery of Vmax (tau = 173.5 +/- 16.0 ms, n = 5). 4. The kinetics for restitution of action potential duration were also slowed by the combination of the two dwo drugs (tau f = 173.6 +/- 16.7, before, vs 268.5 +/- 8.5 ms, after; n = 5, P less than 0.01), while the maximum action potential duration attained in this relation was not increased as it was by sotalol alone. 5. Lignocaine, therefore, appeared to inhibit the sotalol-induced lengthening of action potential duration at slow pacing rates and at long diastolic intervals. The combination of lignocaine with sotalol also completely abolished the occurrence of sotalol-induced early after depolarizations. 6. Finally, sotalol alone moderately increased the range of premature action potential durations, while the combination of the two drugs significantly decreased this parameter. 7. These findings indicate that the combination of lignocaine with sotalol may provide important, and unique, beneficial electrophysiological alterations that might be expected to provide enhanced antiarrhythmic efficacy in patients.

Tocainide and metoprolol: an efficacious therapeutic combination in the treatment of premature ventricular beats

A double-blind crossover study was performed in 20 patients to verify the efficacy of tocainide plus metoprolol in patients with premature ventricular contractions (PVCs) class Lown greater than or equal to 2 (mean frequency greater than or equal to 30/h) judged as being "stable" by at least three basal 24-h Holter ECGs with PVC variation of less than +/- 25%. All 20 patients were submitted to a placebo period; and all were subsequently randomized to therapy with tocainide 1800 mg/day or metoprolol 200 mg/day for 15 days and then to tocainide 1800 mg + metoprolol 200 mg/day or tocainide 1200 mg + metoprolol 200 mg/day for 15 days, followed by a crossover of the two combination treatments. At steady state in every stage we controlled for plasma levels of the drugs, a 24-h Holter recording, and a 12-lead ECG. A modified Lown score was evaluated together with the Lown class. Tocainide (mean plasma level 3.3 +/- 0.7 micrograms/ml) was efficacious in 3 of 8 patients, the modified Lown score decreased from 63 +/- 32 (placebo period) to 42 +/- 27 (p less than 0.01) and Lown 4B arrhythmias were abolished in 3 of 4 patients. Metoprolol (mean plasma level 97.4 +/- 89.6 ng/ml) was efficacious in 2 of 10 patients; the modified Lown score and Lown classes did not change significantly. Administration of tocainide 1200 mg + metoprolol 200 mg obtained a positive response in 9 of 12 patients, the modified Lown score decreased significantly compared with placebo (from 53 +/- 31 to 32 +/- 30, p less than 0.01) and Lown 4B arrhythmias were abolished in 2 of 5 cases. Tocainide 1800 mg plus metoprolol 200 mg was scarcely tolerated owing to neurologic and gastroenteric side effects, and only three patients completed this stage with no better antiarrhythmic results compared to the lower dose. In conclusion, the combination of tocainide at 1200 mg and metoprolol 200 mg is well tolerated, efficacious in a high percentage of patients, and superior to single drug therapy in patients with stable PVCs.

Combination antiarrhythmic therapy for management of malignant ventricular arrhythmia

The efficacy of combination drug therapy in the suppression of ambient ventricular arrhythmia was retrospectively evaluated in a study of 49 patients discharged from the hospital taking 2 membrane-active antiarrhythmic agents. Thirty-one patients (63%) had ischemic heart disease, 15 had miscellaneous cardiac disorders and 3 were free of ostensible heart disease. Therapy in all patients had previously been unsuccessful with an average of 3.7 single membrane-active drugs. Antiarrhythmic agents were discontinued for at least 48 hours to determine baseline arrhythmia levels by Holter monitoring and maximal exercise treadmill testing. Ventricular premature beats were evaluated according to the grading system of Lown and Wolf. Data on ventricular ectopic activity were obtained during Holter monitoring and exercise testing for both a control ("drug-free") period and for a period of combination therapy. During the control period, ventricular tachycardia was recorded during 23% of monitored hours, and the level was nearly twofold greater during stress testing. After institution of combined therapy, the percent of monitored hours of arrhythmia were reduced during Holter monitoring, with a greater reduction in couplets and ventricular tachycardia than in single ventricular premature beats. Ventricular tachycardia was more difficult to provoke by exercise testing in patients taking combination therapy than in control subjects. These data indicate that combination therapy can significantly reduce the density of ventricular ectopic activity in patients refractory to monotherapy. During an average follow-up of 26 months, 23 patients (47%) were able to receive decreased drug dosages, affording diminished adverse effects and improved tolerance to long-term use.