A model analysis for competitive binding of mexiletine and aprindine to the cardiac sodium channel.

A simulation model was developed to predict complex interaction between antiarrhythmic drugs and cardiac sodium channels. This model has four assumptions: (1) Vmax of the action potential is a linear indicator of available sodium channel conductance; (2) antiarrhythmic drugs block the channel by binding to a single common receptor site associated with the channel; (3) binding and dissociation rate constants differ for the three channel states: activated, inactivated and resting, and (4) both drug-free and drug-bound channels change states far more rapidly than binding and dissociation processes. Binding and dissociation rate constants for the three channel states were calculated from single cell experiments using guinea pig hearts. Vmax changes reflecting tonic and use-dependent sodium channel block in the presence of mexiletine and aprindine were simulated and compared with those obtained in the single cell experiments. The model predicted that 'tonic' Vmax inhibition would be enhanced, whereas 'use-dependent' ones would be attenuated after admixture of mexiletine with aprindine. The mechanisms would involve competitive interaction at the common receptor site. Single-cell experiments supported this prediction. We conclude that our simple two-drug binding model provides a useful tool to predict pharmacological interaction between class I antiarrhythmic drugs given in combination.     

Effect of drugs on defibrillation capacity

Over 300,000 people die of sudden cardiac death (SCD) in the US annually. Implantable cardioverter-defibrillators (ICDs) have been shown to be more effective than antiarrhythmic drugs for the prevention of SCD in specific susceptible populations. Many patients in whom ICDs have been implanted receive concomitant therapy with antiarrhythmic drugs, for the purpose of reducing the frequency of appropriate and inappropriate defibrillation shocks. Drugs may influence defibrillation capacity and therefore influence the function of ICDs. The objective of this article is to review and update the literature regarding the effects of drugs on defibrillation capacity.A literature search was performed using PubMed (1966 to December 2007) to identify clinical studies, case reports and animal studies describing the effects of drugs on defibrillation capacity. Search terms included: antiarrhythmic drugs; cardiovascular drugs; amiodarone; sotalol; flecainide; propafenone; dofetilide; ibutilide; beta-blockers; lidocaine; procainamide; N-acetylprocainamide; mexiletine; disopyramide; moricizine; calcium channel blockers; defibrillation threshold; defibrillation energy requirements; defibrillation energy changes; defibrillation efficacy; implantable cardioverter defibrillators; and external defibrillators.Evidence from clinical studies indicates that amiodarone may increase defibrillation threshold (DFT). In addition, some data indicate that drugs including lidocaine, mexiletine, moracizine (moricizine), verapamil, venlafaxine and anaesthetic agents may increase DFT. In contrast, agents including sotalol, dofetilide and beta-adrenergic receptor antagonists (beta-blockers) may reduce DFT. Propafenone and procainamide appear to have minimal effect on DFT. For those antiarrhythmic drugs with both sodium and potassium channel blockade (e.g. amiodarone), the effect of sodium channel blockade predominates, resulting in an increase in DFT.Numerous drugs may affect defibrillation capacity. These effects must be considered when managing patients who have an ICD and require concomitant pharmacotherapy.     

Effects of altered calcium homeostasis on the expression of glutathione S-transferase isozymes in primary cultured rat hepatocytes.

The effects of altered Ca2+ homeostasis on glutathione S-transferase (GST) isozyme expression in cultured primary rat hepatocytes were examined. Isolated hepatocytes were cultured on Vitrogen substratum in serum-free modified Chee's essential medium and treated with Ca2+ ionophore A23187 at 120 hr post-plating. GST activity increased slightly, albeit significantly, in a concentration-dependent manner in A23187-treated hepatocytes relative to untreated controls. Western blot analysis using GST class alpha and mu specific antibodies showed an approximately 1.6- and 1.5-fold increase in the class alpha, Ya and Yc subunits, respectively, whereas no significant increase (approximately 1.2-fold) in class mu GST expression was observed following A23187 treatment. Northern blot analysis revealed an approximately 5-fold increase in GST class alpha and an approximately 7-fold increase in class mu GST mRNA levels in ionophore-treated hepatocytes compared to untreated cells. Results of the Western and Northern blot analyses of the ionophore-treated hepatocytes were compared with those obtained for tert-butyl hydroperoxide-treated cells. Immunoblot analysis showed a significant increase in the expression of GST class alpha, Ya and Yc subunits, approximately 1.8- and 1.7-fold, respectively, for tert-butyl hydroperoxide-treated hepatocytes as compared to controls, with little or no increase in class mu GSTs. Northern blot analysis showed approximately 3- and 2-fold increases, respectively, in class alpha and mu GST mRNA levels, following the tert-butyl hydroperoxide treatment. The results of the present investigation show that alterations in Ca2+ homeostasis produced by either Ca2+ ionophore A23187 or tert-butyl hydroperoxide treatment of hepatocytes enhanced the expression of GST isozymes in primary cultured rat hepatocytes.     

Effects of antiarrhythmic drugs on cloned cardiac voltage-gated potassium channels expressed in Xenopus oocytes

The effects of 17 commonly used antiarrhythmic drugs on the rapidly activating cardiac voltage-gated potassium channels (Kv1.1, Kv1.2, Kv1.4, Kv1.5, Kv2.1 and Kv4.2) were studied in the expression system of the Xenopus oocyte. A systematic overview on basic properties was obtained using a simple and restricted experimental protocol (command potentials 10 mV and 50 mV positive to the threshold potential; concentration of 100 micromol/l each). The study revealed that 8 of 17 drugs yielded significant effects (changes >10% of control) on at least one type of potassium channel in the oocyte expression system. These drugs were ajmaline, diltiazem, flecainide, phenytoin, propafenone, propranolol, quinidine and verapamil, whereas the effects of adenosine, amiodarone, bretylium, disopyramide, lidocaine, mexiletine, procainamide, sotalol and tocainide were negligible. The drug effects were characterized by reductions of the potassium currents (except for quinidine and ajmaline). A voltage-dependence of drug effect was found for quinidine, verapamil and diltiazem. The different effect of the drugs was not related to the fast or slow current inactivation of the potassium channels (except for verapamil). Profiles of the individual drug effects at the different potassium channel types were identical for propafenone and flecainide and differed for all other substances. The study demonstrates marked differences in sensitivity to antiarrhythmic drugs within the group of voltage-operated cardiac potassium channel types. Taking the restrictions of the oocyte system into consideration, the findings suggest that several antiarrhythmic drugs exert significant effects at rapidly activating cardiac potassium channels.     

Effect of dantrolene sodium on [3H]nitrendipine binding to rat cardiac plasma membranes

Dantrolene sodium (Dantrium) has antiarrhythmic activity in addition to its direct-acting skeletal muscle relaxant activity. Dantrolene sodium exerts its skeletal muscle relaxant action by reducing Ca2+ release for sarcoplasmic reticulum. The mechanism by which dantrolene sodium produces its antiarrhythmic effects is not well defined. The effects of dantrolene sodium on [3H]nitrendipine binding to rat cardiac plasma membranes were, therefore, investigated to determine whether the antiarrhythmic action involves an interaction with calcium channels. Whereas 1,4-dihydropyridines maximally inhibited [3H]nitrendipine binding with IC50 values less than 1 nM, verapamil and gallopamil (D 600) inhibited the binding not more than 70% with IC50 values at microM concentrations. Dantrolene sodium caused only minimal inhibition at concentrations up to 100 microM. Thus, the antiarrhythmic action of the drug probably involves a mechanism(s) other than an interaction with the nitrendipine binding site of the slow inward calcium channel.     

Unexpected mexiletine responses of a mutant cardiac Na+ channel implicate the selectivity filter as a structural determinant of antiarrhythmic drug access

Gating properties of Na(+) channels are the critical determinants for the state-dependent block by class I antiarrhythmic drugs; however, recent site-directed mutagenesis studies have shown that the Na(+) channel selectivity filter region controls drug access to and dissociation from the binding site. To validate these observations, we have exploited a naturally occurring cardiac Na(+) channel mutation, S1710L, located next to the putative selectivity filter residue of domain 4, and evaluated the pharmacological properties to mexiletine using whole-cell, patch-clamp recordings. Consistent with the large negative shift of steady-state inactivation and the enhanced slow inactivation, the S1710L channel showed greater mexiletine tonic block than wild-type (WT) channel. In contradiction, S1710L showed attenuated use-dependent block by mexiletine and accelerated recovery from block, suggesting that the drug escape though the external access path is facilitated. Extracellularly applied QX-314, a membrane-impermeant derivative of lidocaine, elicited significantly enhanced tonic block in S1710L similar to mexiletine. However, recovery from internally applied QX-314 was accelerated by 4.4-fold in S1710L compared with WT. These results suggest that the drug access to and dissociation from the binding site through the hydrophilic path are substantially altered. Moreover, K(+) permeability was 1.9-fold increased in S1710L, verifying that the mutated residue is located in the ion-conducting pore. We propose that the Na(+) channel selectivity filter region is a structural determinant for the antiarrhythmic drug sensitivity in addition to gating properties of the indigenous Na(+) channels that govern the state-dependent drug block.     

Biochemical and biophysical studies of the interaction of class I antiarrhythmic drugs with the cardiac sodium channel

Class I antiarrhythmic drugs block the cardiac sodium channel and are now used widely for a variety of cardiac arrhythmias. The nature of the interaction of the drugs with these channels, and their mechanism of action, have been areas of considerable recent interest. Here we review several aspects of drug action. Evidence that sodium channel blockade itself is antiarrhythmic arises from experiments in which the sodium channel-specific toxin tetrodotoxin was shown to prevent ventricular fibrillation in rabbit hearts. A radioligand binding assay for the cardiac sodium channel was then developed using [³H]Batrachotoxinin (BTX) Benzoate and freshly isolated rat cardiac myocytes. Class I antiarrhythmic drug binding identified in this model fits conventional criteria for binding to a receptor. This receptor behaves like an allosteric protein; class I antiarrhythmic drugs appear to bind to and stabilize closed channels. The mechanism of drug binding and electrophysiologic consequences have been examined in a number of model systems by microelectrode studies, whole cell voltage clamping, and more recently voltage clamp studies of single isolated channels in lipid bilayers. Using the latter method, we have shown that lidocaine causes two distinct kinds of sodium channel blockade. The first is a very slow block which can be demonstrated with simple aryl compounds. The second kind of block is a very rapid block which can also be caused by simple alkyl amines. Thus, the two major structural moieties of lidocaine each cause a specific form of block. The structure-activity relationships of class I drugs were explored with homologs of lidocaine in the radioligand binding assay. With this method the existence of a number of receptor subdomains has been demonstrated, each of which recognizes specific structural moieties of class I drugs. Finally, we have explored both drug proarrhythmia and drug resistance. Drug proarrhythmia may be due to homogeneous slowing of conduction with little effect on refractoriness. One form of drug resistance may be due to induction of synthesis of new cardiac sodium channels by exposure to animals with chronic class I drug administration. Mexiletine induces a three-fold increase in rat cardiac sodium channels within 3 days due to increased sodium channel mRNA synthesis. Thus drug resistance might be due to drug-induced increases in the numbers of cardiac sodium channels. © 1994 Wiley-Liss, Inc.     

Antiarrhythmic drugs and epilepsy

For a long time it has been suspected that epilepsy and cardiac arrhythmia may have common molecular background. Furthermore, seizures can affect function of the central autonomic control centers leading to short- and long-term alterations of cardiac rhythm. Sudden unexpected death in epilepsy (SUDEP) has most likely a cardiac mechanism. Common elements of pathogenesis create a basis for the assumption that antiarrhythmic drugs (AADs) may affect seizure phenomena and interact with antiepileptic drugs (AEDs).Numerous studies have demonstrated anticonvulsant effects of AADs. Among class I AADs (sodium channel blockers), phenytoin is an established antiepileptic drug. Propafenone exerted low anti-electroshock activity in rats. Lidocaine and mexiletine showed the anticonvulsant activity not only in animal models, but also in patients with partial seizures. Among beta-blockers (class II AADs), propranolol was anticonvulsant in models for generalized tonic-clonic and complex partial seizures, but not for myoclonic convulsions. Metoprolol and pindolol antagonized tonic-clonic seizures in DBA/2 mice. Timolol reversed the epileptiform activity of pentylenetetrazol (PTZ) in the brain. Furthermore, amiodarone, the representative of class III AADs, inhibited PTZ- and caffeine-induced convulsions in mice. In the group of class IV AADs, verapamil protected mice against PTZ-induced seizures and inhibited epileptogenesis in amygdala-kindled rats. Verapamil and diltiazem showed moderate anticonvulsant activity in genetically epilepsy prone rats. Additionally, numerous AADs potentiated the anticonvulsant action of AEDs in both experimental and clinical conditions. It should be mentioned, however, that many AADs showed proconvulsant effects in overdose. Moreover, intravenous esmolol and intra-arterial verapamil induced seizures even at therapeutic dose ranges.     

Mexiletine. A review of its pharmacodynamic and pharmacokinetic properties, and therapeutic use in the treatment of arrhythmias

As a member of the class Ib antiarrhythmic drugs mexiletine's primary mechanism of action is blocking fast sodium channels, reducing the phase 0 maximal upstroke velocity of the action potential. It increases the ratio of effective refractory period to action potential duration, but has little effect on conductivity. Unlike quinidine it does not prolong QRS and QT (QTc) intervals. In the dosage range 600 to 900 mg daily mexiletine effectively suppresses premature ventricular contractions (PVCs) in 25% to 79% of patients, with or without underlying cardiac disease. In comparative studies the response rate was comparable to that with quinidine or disopyramide. However, the use of antiarrhythmic therapy in patients with asymptomatic arrhythmias is controversial. More importantly, mexiletine abolishes spontaneous or inducible ventricular tachycardia or fibrillation in the short term in 20% to 50% of patients with refractory arrhythmias. Arrhythmia suppression is maintained in 57% to over 80% of these early therapeutic successes in the long term, with mexiletine alone or in combination with another antiarrhythmic drug. As with other antiarrhythmic drugs, there is no substantial evidence that administration of mexiletine after acute myocardial infarction improves long term prognosis. Although the incidence of adverse effects associated with mexiletine is high, the majority are minor gastrointestinal or neurological effects which can be adequately controlled through dosage adjustment. Furthermore, mexiletine has minimal effects on haemodynamic variables, or on cardiac function in patients with or without pre-existing deterioration of left ventricular function, and it appears to have a low proarrhythmic potential. Thus, while the therapeutic efficacy of mexiletine for the prevention or suppression of symptomatic ventricular arrhythmias may be no greater than that of other antiarrhythmic drugs, and less than that of some (e.g. amiodarone), it is effective in a significant proportion of patients refractory to other treatments and can be administered without causing adverse haemodynamic effects to patients with complicating factors such as acute myocardial infarction or congestive heart failure.     

Synthesis and toxicopharmacological evaluation of m-hydroxymexiletine, the first metabolite of mexiletine more potent than the parent compound on voltage-gated sodium channels

The first synthesis of m-hydroxymexiletine (MHM) has been accomplished. MHM displayed hNav1.5 sodium channel blocking activity, and tests indicate it to be ～2-fold more potent than the parent mexiletine and to have more favorable toxicological properties than mexiletine. Thus, MHM and possible related prodrugs might be studied as agents for the treatment of arrhythmias, neuropathic pain, and myotonias in substitution of mexiletine (metabolite switch), which has turned out to be tainted with common toxicity.     

Inhibitory effects of class I and IV antiarrhythmic drugs on the Na+-activated K+ channel current in guinea pig ventricular cells

Recently we have reported that class III antiarrhythmic drugs including amiodarone inhibit the Na⁺-activated K⁺ (K_Na) channels in isolated cardiac cells. In this study effects of antiarrhythmic drugs having class I and/or IV properties on the single K_Na channel current were examined in inside-out membrane patches of guinea pig ventricular cells by using patch clamp techniques. The K_Na channel current, which was activated by increasing [Na⁺]_i from 0 mM to 100 mM in the presence of 150 mM [K⁺]_o, showed a large slope conductance (212 pS) and inward-going rectification. Quinidine (100 μM), mexiletine (100 μM) and flecainide (10 μM) were selected as representative of class Ia, Ib and Ic drugs, respectively. These drugs at relatively high concentrations incompletely inhibited the K_Na channel by decreasing the open time (flickering block). The class IV drug verapamil inhibited the K_Na channel current mainly by decreasing the open probability although the IC₅₀ value of verapamil (3.36 μM) was higher than the therapeutic concentrations. Bepridil and SD-3212, antiarrhythmic drugs having both class I and IV properties, potently inhibited the K_Na channel current by decreasing the open probability. The IC₅₀ values of bepridil and SD-3212 for inhibiting the K_Na channel current was 0.51 μM and 0.53 μM, respectively, both of which are within the therapeutic range. Most antiarrhythmic drugs inhibit cardiac K_Na channels by different modes and at different concentrations. The K_Na channel blocking action of bepridil and SD-3212 may partly contribute to the prolongation of the action potential duration by these drugs at rapid stimulation rates. Received: 20 April 1998 / Accepted: 16 September 1998     

Pharmacokinetics and the antiarrhythmic effect of mexiletine in patients with chronic ventricular arrhythmias

A new antiarrhythmic agent, 1-(2,6- dimethylphenoxy )-2-aminopropane (mexiletine), was investigated in 10 patients with chronic premature ventricular contractions (PVCs) to evaluate the antiarrhythmic efficacy and the pharmacokinetics after single intravenous, single oral and repeated oral dosings of mexiletine 150 mg. Mexiletine was well absorbed from the intestinal tract. The relative bioavailability was 83.2 +/- 8.9% (mean +/- S.E.). The time-concentration curve of mexiletine fitted in well with two-compartment open model. Elimination half-life, volume of distribution and plasma clearance were 10.54 +/- 0.26 h, 2.10 +/- 0.49 l/kg, 6.01 +/- 0.63 ml/min/kg, respectively. The computer-simulated time-concentration curves of multiple oral dosings , which were based on the kinetic parameters from single oral dosing, conformed well with measured concentrations. This might be applied to predict the plasma level of mexiletine. The steady state of plasma mexiletine level was reached 4-5 days after 450 mg/day dosings and ranged 0.75-2.18 micrograms/ml. In 6 of 10 patients, the frequency of PVCs was suppressed more than 75% as compared with the pre-medication value. Mexiletine was well tolerated at a dose of 450 mg/day. However, of 4 patients with the dose increased to 600 mg/day, the administration was ceased in three patients due to gastrointestinal symptoms and tremor. All of these adverse reactions disappeared when the administration was stopped. These results suggest that mexiletine is effective against ventricular arrhythmias and the dosage should be carefully adjusted. The prediction of plasma level would be applied to the dosage regimen of mexiletine.     

l-carnosine and verapamil inhibit hypoxia-induced expression of hypoxia inducible factor (HIF-1 alpha) in H9c2 cardiomyoblasts.

Contractile failure of myocardial cells is a common cause of mortality in ischemic heart disease. In response to hypoxic conditions, cells upregulate the activity of hypoxia-inducible factor 1 (HIF-1) and express a number of genes encoding proteins that either enhance O (2)delivery or increase cellular ATP levels. HIF-1 is a heterodimer of bHLH-PAS proteins, HIF-1 alpha and HIF-1 beta. Both subunits are constitutively expressed under normoxic conditions, but HIF-1 alpha levels are kept low by proteolytic degradation, then stabilized under conditions of low O (2)by a mechanism that is poorly understood. Here we tested the hypothesis that expression of HIF-1 alpha in cardiac cells may be affected by two known cardioprotective agents. We tested l-carnosine, a naturally occurring dipeptide which has been shown to improve myocardial contractility during hypoxia, and verapamil, a calcium channel blocker frequently prescribed for the treatment of heart disease. The levels of HIF-1 alphamRNA remained relatively stable during time course hypoxia (1% O (2)) in H9c2 cardiomyoblasts, then increased slightly after 24 h. In cells pretreated with 1 microM carnosine, the levels of mRNA were transiently reduced, but then increased after 24 h similar to the controls. The levels of HIF-1 alpha protein increased rapidly in H9c2 cells within 30 min of hypoxia, but this induction was significantly reduced in cells treated with either carnosine or verapamil. In addition, treatment of cells with these agents further reduced the low levels of HIF-1 under normoxic conditions. These results suggest that l-carnosine and verapamil may affect the regulated proteolytic degradation of HIF-1 alpha in heart cells during hypoxia.     

Expression and function of glycine-gated Cl- channels]

Glycine is a major inhibitory neurotransmitter in the spinal cord and brain stem. Glycine acts by increasing the Cl- permeability through activation of a specific receptor/ion channel complex consisting of a pentameric subunit assembly. Molecular cloning has disclosed the nature of receptor subunits alpha and beta. While the role of the beta subunit is still unclear, the alpha subunit functions in both ligand (agonist/antagonist) binding and ion channel formation. It has been demonstrated that there are two isoforms of the alpha subunit, alpha 1 and alpha 2. The mRNAs encoding these subunit isomers are transcribed from different genes, in spite of their structural similarity. The alpha 1 mRNA is abundant in adult spinal cord, whereas the alpha 2 mRNA is mainly expressed in developing spinal cord as well as various regions of brain tissue. The single channel properties were examined in outside-out patches excised from Xenopus oocyte membrane expressing alpha 1 or alpha 2 homomeric receptors. The mean open time of alpha 2 channels was 70-times longer than that of alpha 1 channels. The subunit switching from alpha 2 to alpha 1, and resulting shortening of channel kinetics, may ensure a rapid motor control in adult animals.     

Determinants of stereospecific binding of type I antiarrhythmic drugs to cardiac sodium channels

The determinants of stereospecific binding of type I antiarrhythmic drugs to specific sites associated with the sodium channel were assessed using rat cardiac myocytes. The asymmetric carbon atoms of stereoisomers may be located at two sites within type I drugs. The structure of these drugs can be schematically illustrated as Aromatic-C1-link-C2-Amine, where C1 and C2 represent potentially asymmetric carbon atoms. We used enantiomeric pairs with either C1 or C2 asymmetric carbon atoms to assess the importance of conformation at these sites to drug binding. The affinities of enantiomers of seven sodium channel blockers were measured with a radioligand binding assay using [3H]batrachotoxinin benzoate [( 3H]BTXB) and freshly isolated cardiac myocytes. The enantiomers inhibited [3H]BTXB binding in a dose-dependent manner, with a mean Hill number of 1.0 +/- 0.1. The ratios of affinities [IC50 of (+)-isomer/IC50 of (-)-isomer] were, for the C1 pairs: quinidine, 0.29; cinchonidine, 0.55; disopyramide, 1.11; RAC 109, 5.33; and for C2 pairs: flecainide, 1.03; mexiletine, 2.15; tocainide, 3.01. The stereospecific differences in drug binding suggest that the orientations of both the aromatic and the amine groups to the rest of the drug molecule are important determinants of drug binding to the cardiac sodium channel. This also suggests the presence of at least two stereospecific domains within the binding sites for type I antiarrhythmic drugs.     

Effect of calcium channel modulators on temperature regulation in ovariectomized rats

Clinical studies evaluating a calcium channel modulator, gabapentin, for the treatment of vasomotor symptoms have been reported. The present studies evaluated three calcium channel modulators in ovariectomized (OVX) rodent models of temperature regulation. Gabapentin, reported to interact with the alpha(2)delta subunit of voltage-sensitive calcium channels and the L-type voltage-gated calcium channel blockers, verapamil and nifedipine, were examined. These series of experiments demonstrated that orally administered gabapentin, verapamil and nifedipine all acutely and dose-dependently lower tail skin temperature in both models of OVX-induced thermoregulatory dysfunction. These compounds all had a rapid onset of action, however, the efficacy of all three calcium channel modulators is less than that observed following chronic estrogen treatment. Additionally, these compounds were also tested in a telemetric rat model measuring core body temperature to evaluate any temperature effects on internal core temperature. The present data suggests that gabapentin, verapamil and nifedipine all act to globally alter temperature regulation in steroid-dependent models of thermoregulatory function.     

Suppression of ventricular arrhythmias after coronary artery ligation by pinacidil, a vasodilator drug

Pinacidil, a new vasodilator compound, has been shown to lower blood pressure in animals and humans by a direct vasodilator effect. We have studied the effects of pinacidil on experimental cardiac arrhythmias in dogs. Pinacidil did not exhibit antiarrhythmic activity on ouabain-induced arrhythmias. In contrast, verapamil had minor antiarrhythmic activity on the ouabain arrhythmia, restoring sinus rhythm in one-third of the dogs studied. Pinacidil suppressed the arrhythmia present 22-24 h after coronary artery ligation at doses which produced a significant reduction in mean arterial pressure. The antiarrhythmic action of pinacidil was not modified by pretreatment with propranolol but appeared to be blunted by the infusion of the alpha-agonist, phenylephrine. Other hypotensive agents, hydralazine and sodium nitroprusside, although producing similar reductions in mean arterial pressure to pinacidil, did not exhibit a consistent antiarrhythmic action in dogs 22-24 h after coronary artery ligation. The calcium antagonist verapamil did not display antiarrhythmic activity on this model. The mechanisms by which pinacidil exerted an antiarrhythmic action have not yet been elucidated. The results of the present study suggest that further studies with pinacidil on myocardial infarct size, myocardial perfusion, and experimental cardiac arrhythmias would be advantageous.