The cardiac action potential is the fundamental electrical event that triggers mechanical contraction. Its precise ionic architecture varies by cell type and underpins the site-specific effects of antiarrhythmic drugs. Understanding this architecture is not merely academic: it is the pharmacodynamic map from which all antiarrhythmic mechanisms derive.¹

Fast-response cells exhibit a steeply negative resting membrane potential (approximately −90 mV) maintained primarily by the inwardly rectifying potassium current (IK1). Their action potential is characterized by five distinct phases.^1,2

Phase 0 (rapid depolarization) is triggered when membrane potential reaches threshold (~−70 mV); voltage-gated fast sodium channels (Nav1.5, encoded by SCN5A) open rapidly, generating a large inward Na+ current (INa). The upstroke velocity (dV/dtmax) in fast-response cells reaches 200 to 1000 V/s, underlying fast conduction velocity that is exploited by Class I antiarrhythmic agents. Phase 1 (early rapid repolarization) occurs as fast Na+ channels inactivate within milliseconds; a transient outward K+ current (Ito) drives initial repolarization, producing the characteristic notch on the action potential. Ito density is highest in epicardial and Purkinje cells, accounting for the transmural voltage gradient exploited in Brugada syndrome. Phase 2 (plateau) is sustained by a balance between inward L-type Ca2+ current (ICaL) and outward K+ currents (rapid delayed rectifier potassium current (IKr), slow delayed rectifier potassium current (IKs)); ICaL triggers calcium-induced calcium release (CICR) from the sarcoplasmic reticulum, coupling electrical excitation to mechanical contraction. Class IV agents and amiodarone act predominantly at this phase.

Phase 3 (rapid repolarization) is driven by progressive inactivation of ICaL combined with increasing outward K+ currents, predominantly IKr encoded by hERG and IKs encoded by KCNQ1. IKr block is the mechanism shared by virtually all Class III agents and is the principal mechanism of drug-induced QT prolongation. Phase 4 (resting membrane potential) is electrically quiescent in non-pacemaker cells; the Na+/K+-ATPase restores ionic gradients, and IK1 maintains the stable negative resting potential. Any pathologic depolarization during phase 4 constitutes triggered activity or abnormal automaticity.

Nodal cells operate on distinctly different ionic machinery. Their resting membrane potential is less negative (−60 to −70 mV), rendering fast Na+ channels largely inactivated. Depolarization is instead mediated by ICaL, resulting in a slow upstroke velocity (1–10 V/s) and correspondingly slow conduction.² These cells exhibit spontaneous phase 4 depolarization, the pacemaker potential, driven by: (1) the "funny" current If, a mixed Na+/K+ inward current activated on hyperpolarization; (2) declining IK deactivation; and (3) a late-activating ICaL component. This intrinsic automaticity makes nodal cells targets for rate control agents including beta-blockers and non-dihydropyridine calcium channel blockers.^1,3

Proposed by Vaughan Williams in 1970 and subsequently refined, this classification categorizes antiarrhythmic drugs by their primary electrophysiologic action on the cardiac action potential.⁶ Despite well-recognized limitations, it remains the dominant clinical framework and the organizing principle of this module series.

Class I agents block voltage-gated fast Na+ channels (Nav1.5), reducing the maximum upstroke velocity (Vmax) of phase 0 and slowing conduction velocity. They are further divided into three subclasses based on the kinetics of channel blockade, a distinction with major clinical implications:^6,7 Class Ia agents (quinidine, procainamide, disopyramide) exhibit intermediate dissociation kinetics, prolong action potential duration (APD), and produce moderate conduction slowing with QT interval prolongation. Class Ib agents (lidocaine, mexiletine) dissociate rapidly, shorten APD, and are selectively active in ischemic or rapidly firing tissue. Class Ic agents (flecainide, propafenone) dissociate slowly, leave APD unchanged, and produce marked QRS widening through profound conduction slowing.

Use-dependence is a defining pharmacodynamic property of Class I agents: channel blockade intensifies at higher heart rates because more channels cycle through the open/inactivated states per unit time, increasing drug binding. This property makes Class Ic agents particularly hazardous during rapid tachycardias in the presence of structural heart disease, which is the mechanistic basis for the mortality signal observed in the CAST trial.⁷

Class II agents competitively block β1- (and in the case of non-selective agents, β2-) adrenergic receptors, attenuating sympathetically mediated increases in heart rate, atrioventricular nodal conduction, and myocardial contractility. Their antiarrhythmic effects are mediated through:³ Reduction in the slope of spontaneous phase 4 depolarization in pacemaker and Purkinje cells (decreased If activity). Prolongation of AV nodal effective refractory period and slowing of conduction through the AV node (increased PR interval). Suppression of DAD-mediated triggered activity in catecholamine-sensitive arrhythmias (CPVT, exercise-induced VT). Reduction of ischemia-related arrhythmia substrate via decreased myocardial oxygen demand.

Class III agents prolong the cardiac action potential duration (APD) and effective refractory period (ERP) by blocking repolarizing K+ currents, primarily IKr (rapid delayed rectifier, encoded by hERG/KCNH2) and/or IKs (slow delayed rectifier, encoded by KCNQ1). The net effect is QT interval prolongation on surface ECG. Prolonging the ERP without slowing conduction eliminates the excitable gap in re-entrant circuits, the pharmacologic rationale for rhythm control in atrial and ventricular arrhythmias.^6,8 Amiodarone is the class archetype but acts through multiple mechanisms (Classes I, II, III, and IV), making it the broadest-spectrum antiarrhythmic in clinical use. Its unusual pharmacokinetics (an apparent volume of distribution exceeding 60 L/kg and an elimination half-life of 40 to 55 days) necessitate loading protocols and produce toxicities that evolve over months to years.⁸

Class IV agents block L-type voltage-dependent Ca2+ channels (Cav1.2). Their principal antiarrhythmic effect is on slow-response tissue (SA and AV nodes), where ICaL mediates the upstroke. Verapamil and diltiazem reduce AV nodal conduction velocity and prolong AV nodal refractoriness, making them effective for ventricular rate control in atrial fibrillation and for terminating AV nodal-dependent re-entrant tachycardias (AVNRT, AVRT).⁹ Class IV agents have minimal effect on fast-response tissue at therapeutic concentrations. This selectivity underlies both their efficacy and their safety profile in the atria but also their contraindication in ventricular tachycardia with wide QRS morphology, where Ca2+-channel blockade can precipitate hemodynamic collapse in the absence of AV nodal slowing.⁹

The Vaughan Williams classification has endured in clinical practice for over five decades, but its limitations are well-established and clinically significant:^6,10 Multi-class activity: Most agents, particularly amiodarone and sotalol, possess effects spanning multiple classes. Assigning them a single class misrepresents their pharmacology. Active metabolites: Procainamide's active metabolite N-acetylprocainamide (NAPA) has predominantly Class III activity, distinct from the parent compound's Class Ia profile. Reverse use-dependence: Class III agents (especially pure IKr blockers like dofetilide) paradoxically produce greater APD prolongation at slow heart rates, the condition most conducive to EAD formation and TdP. No mechanistic guidance: The classification is based on drug action, not arrhythmia mechanism. Knowing that a drug is "Class Ic" does not indicate which arrhythmias it should or should not be used for. Tissue heterogeneity ignored: The same drug may have substantially different effects in normal versus diseased myocardium, in the atria versus the ventricles, and at differing heart rates.

Mechanistic knowledge translates directly into several clinical imperatives that recur throughout antiarrhythmic prescribing:^5,7 Structural heart disease constrains drug selection: Any agent that slows conduction (Class I) can convert a micro-re-entrant circuit from a non-sustained to a sustained or fatal arrhythmia in scarred myocardium. The CAST trial demonstrated that encainide and flecainide, despite suppressing ambient ventricular ectopy, increased mortality 2.5-fold in post-MI patients. Class Ic agents are therefore contraindicated in structural heart disease. Rate matters: Use-dependent block (Class I) intensifies at rapid rates. Reverse use-dependence (Class III) attenuates protection at rapid rates. Designing a rhythm control strategy requires anticipating the heart rate environment in which the drug will act. Proarrhythmia is class-specific: QT prolongation and risk of torsades de pointes (Classes Ia, III), negative inotropy (Class Ic, verapamil), AV block (Classes II, IV, adenosine, digoxin), and sinus node suppression all derive predictably from mechanism. Anticipating them is part of rational prescribing. Combination therapy requires caution: Combining agents from different classes does not necessarily produce additive benefit but reliably produces additive toxicity. Class III + Class Ia combinations, for example, carry compounded risk of torsades de pointes.