The spectrum of cardiac action potentials varies from fast-response cells – conducting and contractile myocytes (Figure 18.1a) – to slow-response cells of pacemaker myocytes – sinoatrial (SA) and atrioventricular (AV) nodes (Figure 18.1b). Fast myocytes lose their characteristic action potential and behave more like slow myocytes when ischaemic. The action potential is divided into five phases, as follows.

Figure 18.1 (a) Action potential in a fast-response, non-pacemaker myocyte: phases 0–4, resting membrane potential –80 mV, absolute refractory period (ARP) and relative refractory period (RRP). (b) Action potential in a slow-response, pacemaker myocyte. The upward slope of phase 4, on reaching threshold potential, results in an action potential.

PHASE 0

In fast myocytes (Figure 18.1a) rapid depolarisation occurs due to activation of voltage-dependent Na⁺ channels. Activation is initiated in an all-or-none response once the threshold is reached. The Na⁺ channels are inactivated as membrane potential rises to +30 mV and remain inactivated until repolarisation occurs. Rapidity of depolarisation determines speed of conduction. In slow-response myocytes depolarisation does not involve Na⁺ channels and the slower rate of depolarisation is due to a slow inward Ca²⁺ current via L- and T-type voltage-dependent Ca²⁺ channels.

PHASE 1

Early rapid incomplete repolarisation to approximately 0 mV occurs due to activation of transient outward current from I_TO1 and I_TO2 K⁺ channels. Slow myocytes do not exhibit phase 1 or 2 characteristics (Figure 18.1b).

PHASE 2

The prolonged plateau repolarisation of fast myocytes is a consequence of low membrane conductance to all ions. The decreasing inward Ca²⁺ current of L- and T-type Ca²⁺ channels is initially balanced and then overcome by the outward K⁺ current of the delayed rectifiers or the I_k family of K⁺ channels. During this phase the rise in [Ca²⁺]_i is the trigger to release sarcoplasmic reticulum stores of Ca²⁺ and initiate the contractile process.

PHASE 3

Relatively rapid repolarisation occurs as outward K⁺ current of the delayed rectifiers increases. The I_kr K⁺ channel, one of the I_k delayed rectifiers, is the common mechanism whereby antiarrhythmic drugs prolong the action potential and refractoriness.

PHASE 4

This is a stable electrical state in fast non-pacemaker myocytes. In slow pacemaker myocytes the resting membrane potential (RMP) slowly depolarises until the action potential threshold is reached (Figure 18.1b). This inward or pacemaker current is due to I_f K⁺ channels.

Fast-response and slow-response myocytes also have important differences in properties of refractoriness. In fast myocytes, Na⁺ channels are progressively reactivated during phase 3 repolarisation as the membrane potential becomes more negative. When an extra stimulus occurs during phase 3, the magnitude of the resulting inward Na⁺ current and likelihood of impulse propagation depend on the number of reactivated Na⁺ channels. Refractoriness is therefore determined by the voltage-dependent recovery of Na⁺ channels. The absolute refractory period (Figure 18.1) is that minimum time needed for recovery of sufficient Na⁺ channels for a stimulus to result in impulse propagation. However, once propagation in fast myocytes occurs, conduction velocity is normal. In contrast, slow-response or Ca²⁺ channel-dependent myocytes exhibit time-dependent refractoriness. Even after full repolarisation further time is needed before all Ca²⁺ channels are reactivated. Stimuli during this period produce reduced Ca²⁺ current and the propagation velocity of any resulting impulse is reduced. The conduction velocity independence of premature action potentials with fast-response myocytes is lost in the setting of Na⁺ channel-blocking drugs or ischaemia because they behave increasingly like slow-response myocytes with resulting slowed impulse conduction.

GENETIC BASIS TO ARRHYTHMIA ¹

In the absence of structural abnormalities of the heart, primary electrical disease is associated with mutations in ion channel genes. The long-QT syndrome (LQTS), short-QT syndrome, Brugada syndrome (idiopathic ventricular fibrillation (VF)) and catecholaminergic polymorphic ventricular tachycardia (VT: causes of sudden cardiac death (SCD) in the young) are examples of primary electrical disease where genetic mutations encoding for ion channel proteins have been characterised. The ion channel basis of congenital LQTS has been verified with the discovery of disease-causing mutations in the KCNQ1, KCNH2 and SCN5A genes that encode for the cardiac delayed rectifier K⁺ (I_Ks, I_Kr) and sodium (I_Na) channels respectively. Single-gene mutations can also give rise to more than one distinct syndrome with mutation of the SCN5A gene producing both LQTS and Brugada syndromes. This phenotype complexity is presumably the result of interactions between gene expression and environmental factors or the effect of other modifier genes.

Genetic mutations can result in increased or decreased ion channel function. Mutations to the gene KCNQ1 encoding for I_Ks K⁺ channel can result in:

• reduced function, impaired repolarisation and LQTS type 1

• increased function, accelerated repolarisation and short-QT syndrome, which is also associated with SCD

Inheritable forms of structural ventricular disease are associated with atrial arrhythmias and SCD. Examples include hypertrophic and dilated cardiomyopathies and arrhythmogenic right ventricular dysplasia which are linked to mutations in sarcomeric, cytoskeletal and intercellular junction proteins, respectively.

The risk of cardiac arrhythmias and SCD in the setting of acquired structural heart disease such as ischaemic heart disease is in part genetically determined. Studies demonstrate an increased risk of SCD in patients who have a parental history of cardiac arrest.

MOLECULAR BASIS TO ARRHYTHMIA ¹

Structural and electrical remodelling in response to myocardial injury, altered haemodynamic loads and changes in neurohumoral signalling lead to alterations in:

• ion channel function

• intracellular calcium handling

• intercellular communication

• composition of intercellular matrix

All of these factors lead to heterogeneous slowing of conduction velocity and prolonged refractoriness.

Tachycardic remodelling of the atrium is associated with:

• reduced functional expression of L-type Ca²⁺ channels, intracellular Ca²⁺ overload and shortening action potential duration

Heart failure is associated with:

• downregulation of each of the major repolarising K⁺ currents, prolonging repolarisation and resulting insusceptiblity to early afterdepolarisation (EADs). As these changes are heterogeneous they also create the substrate for re-entrant arrhythmias

• reduced Na⁺ channel density decreasing conduction velocity

• increased expression of Na⁺-Ca²⁺ exchanger protein, leading to intracellular Ca²⁺ overload and predisposition to delayed afterdepolarisation (DAD)-mediated arrhythmia

Intercellular ion channels or connexins at gap junctions are decreased and redistributed from the intercalated disc to lateral cell borders, slowing conduction velocity and uncoupling myocytes.

Myocardial infarction scar produces:

• heterogeneous action potential duration with zones of healing myocytes with shortened action potential and surrounding hypertrophic myocytes with prolonged action potentials

• potential anatomical circuits due to fibrous tissue separating myocyte bundles

ARRHYTHMOGENIC MECHANISMS ²^–⁴

Many factors in isolation or combination give rise to the substrate of arrhythmogenesis (Figure 18.2). Arrhythmia may arise from abnormalities of impulse generation or conduction. Table 18.1 demonstrates the relationship between mechanism and type of arrhythmia, and desired antiarrhythmic effect.

Figure 18.2 Factors that combine to form the substrate of arrhythmogenesis.

Table 18.1 Classification of mechanisms of arrhythmia and desired antiarrhythmic drug action

ABNORMAL IMPULSE GENERATION (Table 18.2)

ENHANCED NORMAL AUTOMATICITY

Automaticity is the property of spontaneous impulse generation by cardiac fibres. This results from spontaneous depolarisation during phase 4, due to an inward current carried by K⁺ in SA node or subsidiary pacemaker myocytes.

Table 18.2 Causes of abnormal impulse generation

Enhanced normal automaticity	Adrenergic stimulation
Abnormal automaticity	Ischaemia
Early afterdepolarisations	Hypoxia
	Hypercapnia
	Catecholamines
	Class IA antiarrhythmic drugs
	Class III antiarrhythmic drugs
	Other drugs that prolong repolarisation
Delayed afterdepolarisations	Digoxin toxicity
	Increased intracellular Na⁺
	Decreased extracellular K⁺
	Increased intracellular Ca²⁺
	Intracellular Ca²⁺ overload
	Myocardial infarction
	Reperfusion after ischaemia

ABNORMAL AUTOMATICITY

Abnormal automaticity is the mechanism by which spontaneous impulses are generated in fibres that are partially depolarised by a pathological process. This less negative RMP is associated with inactivation of the normal ionic currents of phase 4 depolarisation and the pacemaker potential results from inward Na⁺ and Ca²⁺ currents and is not readily susceptible to overdrive suppression from normal pacemaker activity. Due to their less negative membrane potentials, these abnormal automatic fibres inactivate the phase 0 fast inward Na⁺ current, resulting in an impaired rate of impulse conduction (as well as contractility) which further contributes to arrhythmia. In this setting, Ca²⁺ carries the major inward current on depolarisation in these fibres.

TRIGGERED ACTIVITY

Abnormal impulse generation from triggered activity originates from oscillations in the membrane potential that are initiated or triggered by a preceding action potential. There are two types of oscillations, EAD and DAD. EAD occurs during phase 2 or 3 of the action potential, whereas DAD occurs after the termination of depolarisation. The signal-averaged electrocardiograph (ECG) can detect after-depolarisations.

1 EAD appear as subthreshold humps during the plateau or depolarisation phases. On reaching threshold, single or multiple action potentials can be induced. Plateau EAD are caused by an increased inward Ca²⁺ current (at this level of membrane potential, fast inward Na⁺ channels are inactivated) and produce slow rising and propagating action potentials. Phase 3 EAD are caused by a reduction in outward K⁺ currents and produce relatively rapidly rising and propagating action potentials. EAD amplitude and likelihood of triggered arrhythmia increase as driving rate decreases and action potential is prolonged. Tachyarrhythmias induced by EAD are more likely to occur on the background of a bradycardia.

2 DAD are produced by Na⁺-Ca²⁺ exchanger current-induced oscillations in inward calcium current. These oscillations are caused by [Ca²⁺]_i overload saturating sarcoplasmic reticulum sequestration mechanisms, thereby leading to Ca²⁺-induced Ca²⁺ release. Unlike EAD, DAD depend on previous rapid rhythm for their initiation.

ABNORMAL IMPULSE CONDUCTION ⁵

Abnormal impulse conduction may cause an arrhythmia by the phenomena of re-entry. Re-entry describes the re-excitation of an area or entire heart by a circulating impulse. Although the classic ‘bifurcating Purkinje fibre’ model of Schmitt and Erlanger has given way to a much more complex picture, the essential electrophysiological requirements for re-entrant excitation remain. Requirements for re-entry are (Figure 18.3):

• conduction block in one limb of the circuit

• slowed conduction in the other limb

• the impulse returns back along the limb initially blocked to re-enter and re-excite the pathway proximal to the block and complete the re-entry pathway

Figure 18.3 Re-entrant excitation. (a) Normal cardiac impulse conduction results in the impulse being extinguished. (b) Conduction down one limb is blocked by segment of refractory tissue (excitable gap). (c) The impulse is conducted back up the limb and arrives at the excitable gap which has recovered from refractoriness and retrograde conduction is complete. If geometry and electrical properties are favourable, the excitable gap circulates around the re-entry loop and arrhythmia is initiated.

When these properties are present, the chance of a circulating impulse producing re-entrant excitation depends on pathway geometry, the electrical properties and length of the depressed area and conduction velocity within each component. The segment of the re-entry pathway that is initially refractory and therefore blocks conduction down one limb and recovers in time to conduct the return impulse is termed the ‘excitable gap’. Therefore, the generation and subsequent maintenance of a circuit depend on this excitable gap of non-refractory tissue circulating between the advancing depolarising wave front and the repolarising tail. The resulting re-entrant impulse can be self-terminating, causing ectopic beats, or lead to atrial or ventricular tachyarrhythmias.

Risk of re-entry can be further modelled and quantified. Cardiac wavelength (λ) is the physical distance an electrical impulse travels in one refractory period. λ equals conduction velocity × refractory period (or action potential duration). Re-entry is critically dependent on the λ being shorter than the potential reentrant pathway. If λ exceeds the path length then the advancing impulse encroaches on the refractory tail and re-entry is terminated. Reducing λ (decreasing conduction velocity or refractory period) promotes re-entry circuits.

Re-entry may be terminated by:

• increasing conduction velocity: the excitable gap is abolished by the wave front arriving too early and meeting refractory tissue

• increasing refractory period: the excitable gap is lost

• slowing conduction: a unidirectional block can be converted into a complete block

Ordered re-entry occurs along anatomical pathways which are ‘macroscopic’ loops (macro-re-entry), as in Wolff–Parkinson–White (WPW) syndrome. Functional circuits can be created following myocardial infarction, resulting in VT. ‘Microscopic’ loops (micro-re-entry) occur at the level of single fibres where antegrade and retrograde impulse propagation occurs in parallel fibres. Random re-entry refers to the generation of a circulating impulse, not from a fixed circuit but from constantly changing electrophysiologically distinct fibres or pathways created by the circulating impulse, resulting in atrial fibrillation (AF) or VF.

The cellular properties that lead to impaired conduction include:

• inactivation of the phase 0 fast Na⁺ channels, which reduces both the magnitude and rate of propagation of any resultant action potential

• intercellular uncoupling, which increases resistance to action potential propagation and slows conduction. Intercellular coupling is reduced by ischaemia, [Ca²⁺]_i overload, acidosis and reduced expression of intercellular ion channel connexin proteins in diseases such as chronic heart failure

ELECTROPHYSIOLOGICAL EFFECTS OF ISCHAEMIA

Both hypoxia and acidosis are implicated in the production of a less negative RMP in ischaemia. A rise in extracellular K⁺ results from impairment of the adenosine triphosphate (ATP)-dependent K⁺ inward channels. As [K⁺]_o/[K⁺]_i is the major determinant of the RMP, intracellular K⁺ loss results in a less negative RMP.

The consequences of this are:

• abnormal automaticity

• inactivation of the fast inward Na⁺ channels, which slows the conduction velocity

ELECTROLYTE ABNORMALITIES AND ARRHYTHMIA ⁶

POTASSIUM

Hyper- and hypokalaemia both cause arrhythmia mediated by the resultant changes in RMP (Table 18.3). In ischaemia, hyperkalaemia at the local tissue level caused by a pathological extracellular shift of K⁺ is the major factor contributing to ventricular arrhythmia in this setting. In hypokalaemia, the dispersion of pacemaker activity and the effect on repolarisation are similar to the electrophysiological effects of cardiac glycosides and β-adrenergic agonists, and it is not surprising that a combination of these factors is associated with an increased incidence of arrhythmia. The increased risk of death in hypertensive patients treated with thiazide diuretics has been attributed to hypokalaemic (and possibly hypomagnesaemic)-induced arrhythmia (Multiple Risk Factor Intervention Trial). Thiazide-induced hypokalaemic ventricular ectopy is worsened by exercise.⁷ Hypokalaemia is associated with VF and VT following acute myocardial infarction (AMI). The increased incidence of VF/VT with a serum K⁺ less than 3.5 mmol/l is clearly established and the probability of VT increases as the serum K⁺ decreases. During AMI the incidence of VF/VT was 15% at 4.5 mmol/l, 38% at 3.5 mmol/l, 55% at 3.0 mmol/l and 67% at 2.5 mmol/l.⁸

Table 18.3 Arrhythmogenic effects of potassium disturbance

MAGNESIUM

The antiarrhythmic properties of Mg²⁺ are clearly established but a causal relationship between hypomagnesaemia and arrhythmia is largely circumstantial. Decreased extracellular Mg²⁺ by itself has little effect on the electrophysiological properties of myocytes or the ECG. Hypomagnesaemia has been implicated in the genesis of VT/VF in patients with hypertension and heart failure receiving thiazide or loop diuretics, acute alcohol intoxication or withdrawal and possibly with AMI. The product of K⁺ and Mg²⁺ is the best predictor of arrhythmia in hypertensive patients taking thiazide diuretics.⁹

AUTONOMIC NERVOUS SYSTEM AND VENTRICULAR ARRHYTHMIA ¹⁰

The autonomic nervous system, particularly vagal tone, has a significant effect on the occurrence of post myocardial infarction VF, as seen by the following:

• High vagal tone is associated with better outcome and less susceptibility to exercise-induced VF with new ischaemia in animal models.

• Post myocardial infarction exercise training results in an increased vagal tone, which inhibits induced VF.

• Implantable electrical vagal stimulation and muscarinic agents, including edrophonium, are protective.

• The protection is not heart rate-related as the protection remains even when atrial pacing is used to maintain the heart rate.

• The administration of atropine increases the likelihood of developing VF.

Vagal tone can be measured by variability in the heart rate (RR interval) or blood pressure rise induced by the pressor agent phenylephrine. Heart rate variability is considered a measure of tonic vagal activity whereas the phenylephrine method is considered a measure of magnitude of the vagal reflex in response to stimulus. A reduced vagal tone has been found postinfarction in humans, which returns to normal over a 3–6-month period. There is no relationship between vagal tone and ejection fraction and the origin of reduced vagal tone postinfarction appears to be due to afferent stimulation in response to necrotic tissue and impaired cardiac contractile geometry. This reduced vagal tone has also been shown to be predictive of mortality and inductility of arrhythmia at electrophysiological study (EPS).

PROARRHYTHMIC EFFECTS OF ANTIARRHYTHMIC DRUGS ^11,¹²

Concomitant proarrhythmia with the use of antiarrhythmic drugs is increasingly recognised. The ‘quinidine syncope’ due to VF and polymorphic VT at therapeutic concentrations was also seen with disopyramide. The Cardiac Arrhythmia Suppression Trial (CAST) clearly defined the magnitude of this deleterious side-effect in drugs that were previously perceived to be of benefit.¹³ This study, which involved flecainide, encainide and morizicine (a class IA drug), was terminated early because of adverse outcome in the flecainide and encainide groups (relative risk of arrhythmic death or non-fatal cardiac arrest of 3.6, 95% confidence interval (CI) 1.7–8.5). Proarrhythmia is reported between 5.9% and 15.8% depending on agent, clinical setting and definition of proarrhythmia, and now considered ubiquitous with all antiarrhythmic drugs.

Proarrhythmia has been defined as an increase in frequency of ventricular ectopic beat (VEB) or aggravation of the target arrhythmia on Holter monitor or exercise test. Manifestations of proarrhythmia not only include VEB, monomorphic and polymorphic VT and VF, but also bradyarrhythmias and Afl with 1:1 AV conduction. Most proarrhythmic events occur soon after starting the drug, but late arrhythmias are also a significant problem.

Proarrhythmia appears to be correlated with the degree of drug-induced QT prolongation or characteristics of sodium channel blockade. Sodium channel blocking agents with a long time constant for recovery of the sodium channel blockade cause more pronounced blockade, even at slow heart rates, slow conduction to a greater extent and are mostly proarrhythmic. Agents with a short time constant of sodium channel blockade, where sodium channel blockade is more pronounced at fast heart rates (e.g. class IB: lidocaine and mexiletine) are less proarrhythmic than drugs with long time constants (e.g. class IC: flecainide and propafenone). Class III drugs and quinidine proarrhythmia correlate with degree of QT prolongation.

The mechanism of drug proarrhythmia is probably via both slowing of conduction and abnormal automaticity. Paradoxically slowing conduction, which may block a re-entry circuit, may also create the very substrate needed for re-entry, unidirectional block and an excitable gap. The existence of a re-entrant circuit requires the circulating wave front of the impulse not to catch up with the refractory tissue behind the tail. Re-entry is more likely to occur with a shorter refractory period and reduced conduction velocity (Figure 18.4).¹¹

Figure 18.4 (a) Graph of refractory period of an excitable gap versus its conduction velocity around a theoretical re-entrant circuit. When conduction velocity is high enough so that refractory period of excitable gap exceeds circuit time, re-entry is impossible. Arrow demonstrates the action of an ‘ideal’ antiarrhythmic drug, which prolongs refractory period and increases conduction velocity. (b) With antiarrhythmic drugs that increase refractory period and slow conduction, the net effect of an antiarrhythmic drug may have no effect on proarrhythmia (arrows 1 and 4), decrease proarrhythmia (arrow 2) or increase proarrhythmia (arrow 3) depending on properties of a potential re-entrant circuit.

(Adapted from Schwartz PJ, La Rovere MT, Vanoli E. Autonomic nervous system and sudden cardiac death. Circulation 1992; 85 (suppl. 1): 77–91 with permission.)

Increasing conduction velocity is an ideal antiarrhythmic property but there are no antiarrhythmic drugs that accelerate conduction. However antiarrhythmic drugs readily slow conduction and the degree of conduction slowing and therefore proarrhythmic tendency correlates with the potency of antiarrhythmic properties.

Prolonging the refractory period is also an ideal antiarrhythmic property, which increases the likelihood of abolishing any excitable gap by ensuring the wave front of a re-entrant circuit meets refractory tissue. The potency of class IA and III antiarrhythmic agents is dependent on the prolongation of the refractory period. This property is also protective against proarrhythmia due to re-entry mechanism. The effect of class IB agents on shortening the refractory period will contribute to proarrhythmia by this mechanism in this class.

Surface mapping of the heart has been used to quantify proarrhythmic effect. The scale of potency of proarrhythmia has been found to be:

Amiodarone was not included in this study but presumably its proarrhythmic potential is similar to other class III agents and less than the class I agents.

Antiarrhythmic drugs are effective at suppressing abnormal automaticity, with the exception of triggered automaticity due to EAD. Class IA, class III and many non-antiarrhythmic drugs can produce proarrhythmia via EAD. These drugs increase not only the frequency of EAD, but also the likelihood of them leading to triggered tachyarrhythmias. Slowing repolarisation, which leads to QT prolongation and slower heart rate, is central to this increased frequency and sensitivity to EAD. EAD manifests as prominent and bizarre T-U waves on the ECG and, if triggered activity results, VEB and ventricular tachyarrhythmias may occur. Torsade de pointes is the classical resulting arrhythmia, although less classical polymorphic VT and VF result. Risk of proarrhythmia via this mechanism correlates with the degree of QT prolongation.

All antiarrhythmic drugs are capable of producing bradyarrhythmias via decreasing normal automaticity and slowing conduction. Digoxin can be proarrhythmic via the production of triggered activity due to DAD.

Antiarrhythmic drug proarrhythmia is facilitated by several factors, which are frequently found in patients on antiarrhythmic drugs or with heart disease (Table 18.4).

Table 18.4 Factors facilitating antiarrhythmic drug proarrhythmia

Toxic blood levels due to excessive dose or reduced clearance from old age, heart failure, renal disease or hepatic disease

Severe left ventricular dysfunction. Ejection fraction less than 35%

Pre-existing arrhythmia or arrhythmia substrate

Digoxin therapy

Hypokalaemia or hypomagnesaemia

Bradycardia

Combinations of antiarrhythmic drugs and concomitant drugs with similar toxicity

(Adapted from Campbell TJ. Proarrhythmic actions of antiarrhythmic drugs: a review. Aust NZ J Med 1990; 20: 275–82, with permission.)

MANAGEMENT OF THE PATIENT WITH A CARDIAC ARRHYTHMIA

HISTORY AND PHYSICAL EXAMINATION

A careful history is important. Specific questions should confirm or exclude palpitations, syncope, chest pain, shortness of breath, ischaemic heart disease (especially previous myocardial infarction), congestive cardiac failure, valvular heart disease, thyrotoxicosis and diuretic therapy without adequate potassium supplements. A family history is helpful for arrhythmias associated with inherited disorders (e.g. LQTS and hypertrophic obstructive cardiomyopathy). The physical examination looks for underlying structural heart disease and signs to assist diagnosis, and assesses haemodynamic consequences of the arrhythmia.

VAGAL MANOEUVRES

Vagal manoeuvres may be undertaken during examination. These reflexly increase vagal tone, thereby prolonging AV node conduction and refractoriness. The effect may be:

• transient slowing of sinus tachycardia as SA nodal discharge rate is slowed

• termination of AV nodal re-entry tachycardia (AVNRT) and AV re-entry tachycardias (AVRT)

• unmasking (but not reversion) of atrial tachycardia, flutter (Figure 18.5) and fibrillation.

Figure 18.5 Atrial flutter with 2:1 atrioventricular (AV) block. Carotid sinus massage (CSM) increases AV block to 4:1 then 6:1.

VT is not affected. Carotid sinus massage is most commonly used. Valsalva manoeuvre or iced water to the face may be useful. Eyeball pressure should be avoided as eye damage may result. Carotid sinus massage is performed with the patient supine, with head extended and turned away from the side to be massaged. After auscultation to exclude carotid bruits, the carotid bifurcation is gently palpated by placing two fingers anterior to the sternocleidomastoid muscle, just below the angle of the jaw. Massage is applied one side at a time, and never both sides simultaneously. It is contraindicated in those with known or suspected cerebrovascular disease.

INVESTIGATIONS

A 12-lead ECG should be recorded with a longer rhythm strip (usually lead II or V₁). If P-waves are not visible, atrial activity may be recorded using an oesophageal electrode or pacing lead, or via a central venous catheter or the right atrial injectate port of a pulmonary artery catheter, using 20% saline and a bedside monitor.¹⁴

Holter monitoring requires prolonged (usually 24–72 h), non-invasive, ambulatory ECG monitoring, sometimes combined with exercise testing.

EPS, which involves invasive electrophysiological testing with programmed electrical stimulation, attempts to reproduce the spontaneously occurring arrhythmia.^15,¹⁶ EPS is not clearly superior to Holter monitoring in evaluating drug treatment for ventricular arrhythmias.

Other investigative techniques being studied include signal-averaged ECG, heart rate variability and electrical alternans measurement.^10,¹⁷

MANAGEMENT OF SPECIFIC ARRHYTHMIAS

Treatment has two aspects: acute termination of the arrhythmia and long-term prophylaxis. The decision whether to treat depends on the rhythm diagnosis, haemodynamic consequences, aetiology of the arrhythmia and the prognosis (e.g. risks of sudden death or long-term complications).

ECTOPIC BEATS

These are premature impulses originating from the atria, AV junction or ventricles. The coupling interval (time between the ectopic and the preceding beat) is shorter than the cycle duration of the dominant rhythm.

PREMATURE VENTRICULAR ECTOPIC BEATS

These are also known as ventricular premature beats and ventricular premature complexes. The ventricle is not normally activated via the rapidly conducting bundle branches, and a wide QRS complex results from slow ventricular conduction.

ECG

There is no preceding P-wave.

• Premature complexes occur before the next expected QRS

• QRS is wide (> 120 ms)

• T-wave of opposite polarity to the QRS (Figure 18.6)

• VEB is not conducted retrogradely to the SA node.

• SA node is therefore not reset, and there is temporary AV dissociation with a full compensatory pause; the interval between the normal QRS complexes on either side of the VEB will usually be twice that of the dominant sinus rhythm.

Figure 18.6 Sinus rhythm with an interpolated ventricular ectopic beat (VEB) without a compensatory pause and a VEB with a following non-conducted P-wave, resulting in a compensatory pause.

Occasionally VEB may not produce any pause, and are said to be interpolated (see Figure 18.6). Interpolated VEB occur when the background sinus rhythm is slow. The retrograde conduction into the AV node renders it partially refractory to the next impulse and its conduction through the AV node is slowed and the PR interval is prolonged. A VEB following each sinus beat is ventricular bigeminy (Figure 18.7). Ventricular trigeminy refers to recurring sequences of a VEB followed by two sinus beats. Two VEB in succession are a couplet (Figure 18.8), and three, a triplet.

Figure 18.7 Sinus rhythm with ventricular bigeminy.

Figure 18.8 Atrial fibrillation with a ventricular couplet.

CLINICAL

Even when frequent, complex, or in short runs of non-sustained VT, VEB are not associated with risk of sudden death in asymptomatic healthy adults.¹⁸ However, there is increased risk of cardiovascular death with:

• exercise-induced VEB: risk of death 2.53 (95% CI 1.65–3.88)¹⁹

• AMI and VEB. Frequent and complex VEB often precede VF or sustained VT and are a marker of risk of subsequent SCD

Apart from ischaemic heart disease, VEB may be associated with cardiomyopathy, valvular disease, myocarditis and non-cardiac precipitating factors (e.g. electrolyte and acid–base disturbances, hypoxia and drugs such as digoxin).

TREATMENT

Drug treatment of VEB is rarely indicated and may be dangerous.

• Correct potassium and magnesium.

• Severely symptomatic patients with frequent complex VEB may benefit from judicious β-blockade.

• The underlying cause of VEB is often more clinically relevant than the arrhythmia. Following myocardial infarction, β-adrenergic blockers, which are indicated for long-term benefit, will also likely suppress VEB.

• Prophylactic lidocaine following AMI will increase total mortality and has been abandoned.^20,²¹

• Attempts at long-term VEB suppression with class IC agents (flecainide and encainide), even if successful, increase mortality.¹³

SUPRAVENTRICULAR TACHYCARDIAS ^22,²³ (Table 18.5)

Supraventricular tachycardias (SVT) are any tachycardias that require atrial or AV nodal tissue for their initiation and maintenance.

• SVT are usually conducted rapidly through the bundle branches so that QRS complexes are narrow.

• All narrow-complex tachycardias are SVT and wide-complex tachycardias are usually ventricular.

• However, SVT may be wide-complex in the setting of bundle branch block (BBB) and pre-excitation.

Table 18.5 Classification of supraventricular tachycardias

Atrioventricular (AV) node-dependent

AV nodal re-entry tachycardia: re-entry within the AV node

AV re-entry tachycardia: re-entry includes accessory pathway between atria and ventricles

Accelerated idionodal rhythm: increased automaticity of AV nodea

AV node-independent

Atrial flutter: re-entry confined to atria

Atrial fibrillation: multiple re-entry circuits confined to atria

Unifocal atrial tachycardia: usually due to increased automaticity

Multifocal atrial tachycardia: increased automaticity or triggered activity

Others: sinus node re-entry tachycardia

A clinically useful classification divides SVT into AV node-dependent and AV node-independent.

Distinguishing between AV node-dependent and independent SVTs can be difficult. Vagal manoeuvres or drugs that prolong AV nodal refractoriness (e.g. adenosine) may assist in diagnosis:

• Temporary AV block with unchanged atrial rate indicates AV node independence.

• Slowing or reversion of the tachycardia diagnoses AV dependence.

AV NODE-DEPENDENT SVT

In these SVT, sometimes referred to as junctional tachycardias, the re-entry circuit or ectopic focus involves the AV node or junction. Blocking the AV node with drugs such as adenosine or vagal manoeuvres will terminate these SVT.

AV NODE-INDEPENDENT SVT

Also referred to as atrial tachycardias, the atrial tissue only is required for the initiation and maintenance of the tachycardia. Blocking the AV node will not terminate these SVT; it will merely slow ventricular rate.

AV NODAL RE-ENTRY TACHYCARDIA (AVNRT)(Figure 18.9)

Re-entry tachycardia is confined to the AV node. Antegrade conduction to the ventricles usually occurs over the slow pathway and retrograde conduction over the fast pathway.

Figure 18.9 Arteriovenous nodal re-entry tachycardia (AVNRT) has both pathways in the AV node. The conduction occurs over the slow pathway and retrogradely over the fast pathway. AV re-entry tachycardia (AVRT) involves antegrade conduction through the AV node and retrograde conduction through an accessory pathway.

AV RE-ENTRY TACHYCARDIA (SEE Figure 18.9)

The re-entry pathway consists of the AV node and an accessory pathway, which bypasses the AV node. The accessory pathway may be evident during sinus rhythm, with the ECG showing pre-excitation: short PR interval, delta wave and widening of the QRS (see WPW, below, under pre-excitation syndrome). However, in 25% of cases, the accessory pathway conducts only retrogradely from ventricle to atria and the ECG pre-excitation will be concealed in sinus rhythm. Orthodromic AVRT, with antegrade nodal and retrograde accessory pathway circuit, is the most common regular SVT in patients with accessory pathway.

ECG

Figure 18.11 Arteriovenous re-entry tachycardia (AVRT). Narrow QRS tachycardia at 135 beats/min. Inverted P-wave in leads I, II, III and aVF just following the QRS complex.

Figure 18.12 Arteriovenous re-entry tachycardia (AVRT). Rate is 214 beats/min. P-wave deflection is just seen on the upslope of the T-wave in lead V₁.

CLINICAL

TREATMENT ^24,²⁵

Acute treatment is identical to AVNRT, but verapamil should be avoided in WPW syndrome, as it may block the AV node, facilitating very rapid conduction to the ventricles via the accessory pathway.²⁷

PREVENTION

Drugs such as sotalol and flecainide may prevent recurrence of the tachycardia. Radiofrequency ablation of the accessory pathway is usually curative.²⁶

ACCELERATED IDIONODAL RHYTHM

Increased automaticity of the AV junction (above the inherent discharge rate of 40–60 beats/min) is the usual cause of this arrhythmia. The often-used term ‘non-paroxysmal AV junctional tachycardia’ is cumbersome and misleading: junctional rate is commonly 60–100 beats/min, not strictly a tachycardia. AV dissociation is often present, but there may be synchronisation of the two pacemakers – so-called isorhythmic dissociation.

ECG

There are narrow complexes on the ECG at a regular rate (60–130 beats/min) (Figure 18.13), often with independent atrial activity. With isorhythmic dissociation, P-wave is either fixed relative to the QRS complex (usually just after) or oscillates to and fro across the QRS in a rhythmical manner.

Figure 18.13 Accelerated idionodal rhythm: 105 beats/min. Inverted P-wave is immediately following the QRS complex in leads II, III and aVF.

CLINICAL

It may be observed in normal persons, but is often associated with structural heart disease, especially following inferior myocardial infarction. Digoxin intoxication is another important cause.

TREATMENT

In most cases, the rhythm is transient and well tolerated, and no treatment is required. Treatment is otherwise directed towards the underlying cause.

UNIFOCAL ATRIAL TACHYCARDIA

This is sometimes called ectopic atrial tachycardia to distinguish it from the atrial tachycardias (referring collectively to unifocal atrial tachycardia, Afl and AF). However, it is inappropriate to call atrial tachycardia paroxysmal atrial tachycardia. Paroxysmal, by definition, indicates an abrupt onset and termination, which applies less commonly to unifocal atrial tachycardia. Vagal manoeuvres will not terminate this arrhythmia, but AV block may be induced, or increased if already present.

ECG

P-wave morphology is abnormal but monomorphic. Atrial rate is often 130–160 beats/min, and may occasionally exceed 200 beats/min. Atrial rate distinguishes unifocal atrial tachycardia from atrial flutter (Afl), with Afl greater than 250 beats/min. The QRS complexes will usually be narrow (Figure 18.14). AV block is common (Figure 18.15).

Figure 18.14 Unifocal atrial tachycardia with 1:1 arteriovenous conduction; rate is 140 beats/min. Large, inverted P-waves are seen in lead II.

Figure 18.15 Unifocal atrial tachycardia with 2:1 arteriovenous conduction. Atrial rate is 170 beats/min.

CLINICAL

Digitalis intoxication is the most common cause, especially when AV block is present. Other causes include myocardial infarction, chronic lung disease and metabolic disturbances.

TREATMENT ²⁵

If applicable, digitalis is stopped and the toxicity treated. Otherwise digoxin may be used to control the ventricular rate. β-Adrenergic blockers or amiodarone are alternatives. Rapid atrial pacing may be ineffective if the arrhythmia is due to increased automaticity, although it may increase AV block, thereby slowing ventricular rate. Synchronised DC shock may be necessary, but is avoided if digitalis intoxication is suspected.

MULTIFOCAL ATRIAL TACHYCARDIA ²⁸

Multifocal atrial tachycardia (MAT) is defined as an atrial rhythm, with a rate greater than 100 beats/min, with organised, discrete non-sinus P-waves having at least three different forms in the same ECG trace. The baseline between P-waves is isoelectric, and the PP, PR and RR intervals are irregular. This is an uncommon arrhythmia, also known as chaotic or mixed atrial tachycardia.

ECG

There are irregular atrial rates, usually 100–130 beats/min, with varying P-wave morphology (at least three different P-wave morphologies and varying PR interval) and some degree of AV block (Figure 18.16). Most P-waves are conducted to the ventricles, usually with narrow QRS complexes.

Figure 18.16 Multifocal atrial tachycardia with the rate about 130 beats/min. There is varying P-wave morphology and PR intervals. Note the wide complex preceded by a P-wave. The aberrant intraventricular conduction is related to the long–short cycle length sequence.

CLINICAL

MAT is often misdiagnosed and inappropriately treated as AF. This rhythm occurs most commonly in critically ill elderly patients with chronic lung disease and often cor pulmonale, and is associated with a very high mortality from underlying disease. Theophylline has been implicated as a precipitating cause, and rarely digoxin.

TREATMENT

Treatment should correct the underlying cause (e.g. treatment of cardiorespiratory failure, electrolyte and acid–base abnormalities and theophylline toxicity). Spontaneous reversion is common, and few patients require antiarrhythmic therapy. Magnesium is the drug of choice for acute control.²⁹ β-Blockers are probably more effective than diltiazem, but because of the common association of MAT with obstructive lung disease have limited utility.³⁰ Digoxin and cardioversion are ineffective, which highlights the need to differentiate MAT from AF. Longer-term control is best achieved with diltiazem in patients with good left ventricular (LV) function and amiodarone in those without.

ATRIAL FLUTTER ³¹

Atrial rate during classical Afl is 250–350 beats/min, and in most cases, close to 300 beats/min. Afl is due to a single re-entry circuit lying within the right atrium and the wave of depolarisation in most patients is anticlockwise. If the right atrium is significantly enlarged the rate may be considerably slower. Studies in patients who had recently undergone cardiac surgery subdivided Afl into type I and II on the basis of rate and typical responses to atrial pacing.

Type I flutter was slower – rate 240–320 beats/min – and was readily entrained with overdrive pacing. Type II flutter was faster than type I, with rates of 340–430 beats/min. Type II flutter could not be entrained or terminated by pacing. Type II is thought to arise from a circus pathway with a very short excitable gap.

ECG

Afl waves (characteristic sawtooth appearance with no isoelectric baseline) are best seen in V₁ (Figure 18.17) or aVF, but leads II and III may also be useful. The flutter waves are usually negative in aVF. Rapid QRS waves may obscure typical flutter waves, and vagal manoeuvres may unmask them (see Figure 18.5). AV conduction block (usually 2:1) is usually present, so that alternate flutter waves are conducted to the ventricles, with a ventricular rate close to 150 beats/min. Frequently flutter waves are not obvious and a ventricular rate of 150 beats/min leads to the presumption of Afl (Figure 18.18). Type II Afl results in greater atrial and ventricular rates (Figure 18.19). Treatment with drugs that affect AV node conduction may lead to higher degrees of AV block (Figure 18.20) and/or variable AV block with irregular QRS duration. Rarely, Afl with 1:1 conduction occurs. This is usually associated with sympathetic overactivity or class I antiarrhythmic drugs (which slow atrial discharge rate to 200 beats/min, thereby allowing each atrial impulse to be conducted) (Figure 18.21). QRS complexes are usually narrow, as conduction through the bundle branches is normal.

Figure 18.17 Atrial flutter with 2:1 arteriovenous conduction. Atrial rate is 270 beats/min (arrows V₁) and ventricular 135 beats/min. Characteristic ‘sawtooth’ inverted flutter waves are evident in leads II, III and aVF.

Figure 18.18 Atrial flutter with 2:1 arteriovenous (AV) conduction. Inverted flutter waves are difficult to differentiate from T-waves. Rate of 144 beats/min confirms atrial flutter with 2:1 AV conduction.

Figure 18.19 Atrial flutter with 2:1 arteriovenous conduction. Type II atrial flutter is confirmed by the rapid atrial rate of 380 beats/min.

Figure 18.20 Atrial flutter varying between 3:1 and 4:1 arteriovenous (AV) conduction due to drug effect slowing AV node conduction.

Figure 18.21 Atrial flutter with 1:1 conduction with a rate of 240 beats/min. Up-sloping of ST segment is easily mistaken as part of the QRS complex, giving the appearance of a broad QRS tachycardia in some leads. Lead III shows true narrow width of QRS complex.

CLINICAL

AFL is less common than AF. It may occur in ischaemic heart disease, cardiomyopathy, rheumatic heart disease and thyrotoxicosis, and after cardiac surgery.

TREATMENT ²⁵

No drug will reliably terminate Afl, although ibutillide and dofetilide have been shown to be most likely to result in pharmacological reversion. Attempts at slowing ventricular rate by drugs that will increase the degree of AV block are worthwhile in the first instance. Drugs such as digoxin, diltiazem, β-adrenergic blockers, sotalol and amiodarone may be tried; the choice depends on LV function. Flecainide and procainamide may occasionally be effective at terminating Afl. However, class IA and IC drugs may lead to 1:1 AV conduction. Class I drugs should probably be avoided unless ventricular response has been slowed with calcium channel or β-adrenergic blocking drugs.

Synchronised DC cardioversion, often with low energies (25–50 J), is a reliable treatment option. Rapid atrial pacing faster than the flutter rate will terminate classical or type I Afl in most patients.

Anticoagulation guidelines are the same as that for AF, although there are less supporting data.

PREVENTION

Prevention is difficult. Drugs used include sotalol and amiodarone at low doses. Class IC agents (e.g. flecainide) may be used in patients without significant structural heart disease. Increasingly recurrent or refractory Afl may be cured by radiofrequency ablation to create a linear lesion between the inferior tricuspid annulus and the eustachian ridge at the anterior margin of the inferior vena cava to interrupt the re-entry circuit.²⁶

ATRIAL FIBRILLATION ³²

AF is the most common arrhythmia requiring treatment and/or hospital admission. The incidence increases with age: 5% of individuals over 70 years have this arrhythmia. There is also an age-independent increase in frequency due to increase in obesity and obstructive sleep apnoea. LV dyfunction increases risk of AF (4.5-fold in men and 5.9 in women) with atrial stretch and fibrosis causing electrical and atrial ionic channel remodelling.

AF is common in:

• congestive cardiac failure (40%)

• coronary artery bypass grafting (25–50%)

• critically ill patients (15%)

Idiopathic or lone AF (i.e. with no structural heart disease or precipitating factor) in someone aged under 60 years has an excellent prognosis; however, AF developing after cardiac surgery, for instance, is associated with increased stroke, life-threatening arrhythmias and longer hospital stays.

ECG

Atrial activity is chaotic with rapid (350–600 beats/min) and irregular depolarisations varying in amplitude and morphology (fibrillation waves). Ventricular response is irregularly irregular (Figure 18.22). Most atrial impulses are not conducted to the ventricles, resulting in an untreated ventricular rate of 100–180 beats/min. QRS complexes will usually be narrow. When the ventricular rate is very rapid or very slow, ventricular irregularity may be missed (Figure 18.23).

Figure 18.22 Atrial fibrillation. Irregular fibrillation waves with varying amplitude and morphology.

Figure 18.23 Atrial fibrillation with rapid ventricular rate. Ventricular irregularity and fibrillation waves are less evident when the rate is rapid.

CLINICAL

AF is more common in patients with underlying heart disease (particularly those with a dilated left atrium) and abnormal atrial electrophysiology. Causes include ischaemic and valvular heart disease, pericarditis, hypertension, cardiac failure, thyrotoxicosis and alcohol abuse. AF may also occur after cardiac surgery and thoracotomy. AF can be chronic, or intermittent with paroxysmal attacks. Chronic AF has a poorer prognosis.

AF is associated with:

• adverse haemodynamic effects. Rapid ventricular rate and loss of atrial systole may increase pulmonary capillary wedge pressure, while stroke volume and cardiac output decline

• systemic embolism and stroke

• tachycardiomyopathy. Reversible global cardiomyopathy secondary to rapid heart rate. Assessing LV function with echocardiogram before and after AV node ablation for AF refractory to medical therapy suggests that 10% of patients with AF have AF-induced tachycardiomyopathy ³³

TREATMENT ^25,^34,³⁵

The goals of treatment include ventricular rate control, anticoagulation where appropriate and conversion to sinus rhythm. There is increasing evidence available on the ‘rate versus rhythm’ control debate. Results from several recent major studies have challenged the previous belief that achievement of sinus rhythm is important in the long term (Table 18.6). When comparing control of ventricular rate versus reversion to sinus rhythm no clear survival benefit is apparent. However composite end-points of death, stroke and recurrent hospilisation favour rate control only.³⁶^–³⁹

Table 18.6 Atrial fibrillation rate versus rhythm control debate

The possible reasons why rhythm control has not been shown to be superior include:

• Trials have included predominantly elderly high-risk patients.

• Sinus rhythm is difficult to achieve (39–63%).

• Rate control strategies can result in sinus rhythm in up to 35% of patients.

• Underlying heart disease that initiates the AF persists.

• There may be antiarrhythmic drug side-effects.

• Anticoagulation is still required even if rhythm control is successful.

However rhythm control (if possible) appears superior in patients with LV dysfunction, with both amiodarone and dofetilide reducing mortality when sinus rhythm is achieved.^40,⁴¹ The paucity of data in younger patients (less than 60 years) favours initial attempts at rhythm control, particularly in those with structurally normal hearts, in the hope that progressive atrial electrical and anatomical remodelling is prevented.

RECENT ONSET OR PAROXYSMAL AF

VENTRICULAR RATE CONTROL

The urgency of ventricular rate control depends on the clinical situation and spontaneous reversion of AF is common. Treatment may not be necessary, and a reasonable strategy is based on clinical status:

• Haemodynamically unstable with rapid ventricular rate requires immediate synchronised DC shock (in addition to drug therapy) to control rate urgently.

• Haemodynamically stable, symptomatic with depressed LV function: semiurgent synchronised DC shock or drug therapy, digoxin or amiodarone to control ventricular rate.

• Haemodynamically stable, symptomatic, normal LV function: control of ventricular response with β-adrenergic blockers, diltiazem, digoxin (poor control with exertion and settings with increased sympathetic tone), magnesium (short-term), amiodarone or sotalol.

• Haemodynamically stable, with no structural heart disease and minimal or no symptoms: no immediate treatment is an option. Most cases will revert spontaneously within 24 h. Single-dose flecainide for paroxymal AF has been recommended (contraindicated in patients with structural heart disease).⁴²

Ideal rate control can be defined as a resting heart rate of 80 beats/min, peak rate of 110 beats/min with 6-minute walk and an average of 100 beats/min.

CONVERSION TO SINUS RHYTHM

Antiarrhythmic drugs or DC shock cardioversion can be used. The likelihood of short- and long-term success depends on the clinical situation. Conversion to sinus rhythm is more important in young patients and those with heart failure. Maintenance of sinus rhythm is problematic: sinus rhythm at 1 year is 60% with amiodarone and 40% with sotalol, and associated with significant drug cardiac and extracardiac toxicities. The risk of stroke and need for antithrombotic therapy due to frequent AF recurrences, which may be asymptomatic, remain. Achieving sinus rhythm (especially greater than 60 years) is less important than previously thought.³⁶

DC SHOCK CARDIOVERSION

DC shock cardioversion is indicated either before 24–48 h or after appropriate anticoagulation protocol. Combining DC shock with antiarrhythmic drugs to promote maintenance of sinus rhythm is favoured, especially if risk factors for relapse exist. Cardioversion is less likely to be successful if:

• AF has been present for over 1 year

• left atrial size is greater than 45 mm

• untreated conditions are present, e.g. thyrotoxicosis, valvular heart disease and heart failure

Critically ill patients who are septic, postoperative or on drugs such as catecholamines are likely to relapse.

ANTICOAGULATION AND CARDIOVERSION ^43,⁴⁴

Loss of atrial contraction with AF is associated with stasis of blood flow and formation of blood clots in the left atrium, particularly the atrial appendage. Reversion to sinus rhythm and return of more effective atrial contraction may cause expulsion of any atrial clots and systemic emboli. Once AF has been present for more than 48 h – some authors stipulate 24 h – the risk of systemic emboli is significant and anticoagulation is required prior to DC shock cardioversion. The current recommended period of anticoagulation prior to DC shock cardioversion is 3 weeks. This 3-week period can be shortened to 1 day for heparin and 5 days for warfarin if the left atrium can be demonstrated free of clot using transoesophageal echocardiography. With this accelerated approach heparin dose should be titrated to an activated partial thromboplastin time 2–3 times control or warfarin to produce an international normalised ratio (INR) of 2.0–3.0. In many clinical situations such as recent surgery or other bleeding risks, anticoagulation is contraindicated and elective cardioversion should be delayed until recommended anticoagulation cover is safe. Following successful cardioversion to sinus rhythm the risk of systemic embolism continues as the propensity to form atrial clot remains due to atrial contractile stunning and anticoagulation should be continued for 4–6 weeks.

ANTIARRHYTHMIC DRUGS

The drugs used for ventricular rate control – digoxin, diltiazem and β-adrenergic blockers – are unlikely to result in pharmacological cardioversion.

Antiarrhythmic drugs that may cardiovert are unfortunately relatively ineffective and may possibly be dangerous. They are more effective at retaining sinus rhythm. About 50% will remain in sinus rhythm 1 year after cardioversion with drugs and 25% without drugs.

Quinidine is more effective than placebo, but increases mortality through proarrhythmia (generally class IA and IC antiarrhythmic drugs are contraindicated). Ibutilide and dofetilide are newer antiarrhythmic drugs with particular success at pharmacological cardioversion. Pretreatment with ibutilide increased DC shock cardioversion from 72% for placebo to 100%. In placebo failures, cross-over to ibutilide resulted in a 100% success rate with subsequent cardioversion. Ibutilide also resulted in reduction in DC shock energy required from 228 ± 93 J to 166 ± 80 J. However, ibutilide was associated with a 3% incidence of sustained polymorphous VT.⁴⁵ Dofetilide appears to cardiovert AF and Afl pharmacologically in about a third of patients (intravenous (IV) better than oral, recent onset better than prolonged and Afl may be more responsive than AF). Dofetilide is far superior to placebo and sotalol, with similar recurrence rates to amiodarone.⁴⁶

Other drugs currently used to promote onset of sinus rhythm and prevent AF relapse include amiodarone, sotalol, procainamide, flecainide and propafenone. Amiodarone was found to be superior in preventing AF recurrence with a recurrence rate of 35% compared to a recurrence rate of 63% for sotalol and propafenone.⁴⁷

The factors dictating choice are:

• degree of LV depression: amiodarone has the least and flecainide the greatest

• risk of proarrhythmia: worse with flecainide and propafenone

• long term side-effect profile: amiodarone is the worst

When using amiodarone for prevention of AF recurrence there was an 18% incidence of adverse effects versus 11% for sotalol and propafenone.⁴⁷

ATRIAL FIBRILLATION ABLATION THERAPY

Ablation techniques for AF have been continuously refined since the original Maze III surgical procedure which involved numerous atrial incisions to form a maze-like pattern of scarring, blocking propagation of arrhythmia. The ultility of this procedure was limited because it was surgical, with longer bypass times, postoperative bleeding and impaired atrial contractility. The magnitude of this original procedure was based on the belief that the entire atrium was involved in the initiation and maintenance of the fibrillatory conduction. This may be true for long-standing AF but paroxysmal AF appears to originate primarily at the junction of the left atrium and pulmonary veins. AF in 94% of patients is initiated by rapid discharges from one or more foci at or near the pulmonary vein orifices.⁴⁸ Atrial tissue in this area has heterogeneous electrophysiological properties and there is also clustering of vagal inputs, which creates substrate for rapid discharges that initiate microre-entrant circuits or ‘rotors’. These high-frquency periodic rotors send spiral wave fronts of activation into surrounding atria. Localised ablation of a single dominant foci and rotor is inadequate as there are usually multiple foci.

There is renewed interest in surgical AF ablation therapy in conjunction with cardiac surgery. Complications have been reduced with energy (cryotherapy, radiofrequency) rather than incisions and the extent of lesions reduced. The minimum lesion set is now considered to be encirclement of pulmonary veins, linear lesion from the inferior pulmonary vein to mitral annulus and from the coronary sinus to the inferior vena cava.

Left atrial catheter (transatrial septum) AF ablation isolating all four pulmonary veins using radiofrquency is being heralded as the possible AF cure. Results are improving as all pulmonary veins are now isolated and the encircling lesion is clear of the pulmonary vein antrum (reducing pulmonary vein stenosis). Success rates of 81% (75–88%) free of AF and off drugs are reported. Success appears long-term as recurrence occurs early. A further 10–20% may become responsive to antiarrhythmic drugs which were previously ineffective. Repeating the procedure can increase success to > 90% with failure only in patients found to have extensive atrial scarring (predicting and excluding patients with this extensive atrial scarring is a major future challenge). Although not yet the universal cure the results are two- to threefold better than antiarrhythmic drugs alone.

Complication rates are also falling associated with:

• intracardiac echocardiography ensuring safer transeptal puncture and positioning of isolating lesions clear of the pulmonary vein antra

• higher levels of procedural anticoagulation

• strict limitations on radiofrequency energy output

Transient ischaemic attacks, strokes, tamponade/perforation and symptomatic pulmonary vein stenosis are all well below 1% respectively. Proarrhythmia resulting from re-entrant tachycardias from incomplete ablative lesions is more common. Some are advocating ablation as first-line treatment whereas most are selecting younger patients (less than 70 years) with paroxysmal AF for whom antiarrhythmic therapy has failed, left atrial diameter is less than 5 cm and ejection fraction is greater than 40%.²⁶ Head-to-head studies comparing ablation and antiarrhythmic drugs are appearing with suggested survival benefit, improved quality of life, reduced adverse effects and cost-effectiveness after approximately 3 years with catheter AF ablation therapy.^49,⁵⁰

CHRONIC ATRIAL FIBRILLATION

Although most patients undergo at least one attempt at cardioversion to sinus rhythm, many are left in chronic AF, particularly those with dilated atria, poor LV function and valvular heart disease. In this setting, treatment aims at ventricular rate control and prevention of embolic stroke.

VENTRICULAR RATE CONTROL

Digoxin is often the drug of choice, particularly in patients with poor LV function. It is often ineffective at controlling ventricular rate during exercise and physiological stress. Judicious addition of a small dose of a β-adrenergic blocker may improve digoxin control. Amiodarone is particularly useful in patients with poor LV function. Higher-dose β-adrenergic blocker, diltiazem, sotalol or flecainide can be used in patients with good LV function. Rarely, His-bundle ablation with permanent cardiac pacing may be required for severe cases refractory to drug therapy.

ANTICOAGULATION FOR CHRONIC ATRIAL FIBRILLATION

Consider for all patients, especially those with risk factors (Table 18.7).

Table 18.7 Prognostic factors for ischaemic stroke and systemic embolism in patients with atrial fibrillation

High	Previous stroke, transient ischaemic attack, systemic embolism
	Mitral stenosis
	Prosthetic heart valve
Moderate	Age > 75 years
	Left atrial size > 45 mm
	Hypertension
	Congestive cardiac failure
	Diabetes mellitus
	Left ventricular ejection fraction < 35%
Low	Female
	Age 65–74 years
	Coronary artery disease
	Thyrotoxicosis