Abstract
The term typical atrial flutter (AFL) is reserved for a macroreentrant circuit with the activation wavefront rotating clockwise or counterclockwise around the tricuspid annulus and using the cavotricuspid isthmus (CTI) as an essential part of the reentry circuit. Typical AFL is the most common type of macroreentrant atrial tachycardia.
There are four main issues that must be addressed in the treatment of AFL: (1) ventricular rate control; (2) restoration of sinus rhythm; (3) maintenance of sinus rhythm; and (4) prevention of systemic embolization. Catheter ablation is recommended as first-line therapy for most patients with symptomatic or recurrent typical AFL, whether paroxysmal or persistent. The ablation procedure is associated with high long-term success rates (92% after a single procedure and 97% after multiple procedures), and low risk of serious complications (0.4%). In addition to improvement of symptoms and quality of life, successful ablation offers a potential cure of the arrhythmia, reducing the risk of thromboembolism, and potentially eliminating the need for long-term anticoagulation and antiarrhythmic drug therapy. The CTI is the ideal target of AFL ablation, and complete bidirectional CTI block is the primary endpoint of ablation of typical AFL.
Keywords
atrial flutter, macroreentrant atrial tachycardia, cavotricuspid isthmus, double-wave reentry, cardioversion
Outline
Pathophysiology, 339
Right Atrial Anatomy, 339
Typical Atrial Flutter Circuit, 340
Interrelationship of Atrial Flutter and Atrial Fibrillation, 341
Double-Wave Reentry, 342
Epidemiology and Natural History, 343
Clinical Presentation, 343
Initial Evaluation, 343
Principles of Management, 344
Rate Control, 344
Restoration of Sinus Rhythm, 345
Maintenance of Sinus Rhythm, 346
Prevention of Thromboembolism, 347
Management of Coexistent Atrial Flutter and Atrial Fibrillation, 347
Electrocardiographic Features, 347
Electrophysiological Testing, 348
Induction of Tachycardia, 351
Tachycardia Features, 351
Diagnostic Maneuvers During Tachycardia, 353
Entrainment, 353
Termination, 355
Overdrive Suppression, 355
Acceleration, 355
Transformation, 356
Mapping, 356
Activation Mapping, 356
Entrainment Mapping, 356
Electroanatomic Mapping, 358
Noncontact Mapping, 359
Ablation, 359
Target of Ablation, 359
Ablation Technique, 359
Endpoints of Ablation, 364
Confirmation of Bidirectional Cavotricuspid Isthmus Block, 365
Outcome, 371
Organized atrial tachycardias (ATs) are broadly categorized as either focal or macroreentrant. Focal ATs exhibit a centrifugal activation pattern originating from a discrete site, and can have automaticity, triggered activity, and microreentrant mechanisms. A macroreentrant AT incorporates a relatively large reentrant circuit around a central obstacle. Depending on whether or not the cavotricuspid isthmus (CTI) is critical to the reentry circuit, macroreentrant ATs are divided into two groups: “CTI-dependent” or “non-CTI-dependent” macroreentrant AT ( see Table 11.1 ). CTI-dependent macroreentrant ATs include typical atrial flutter (AFL), lower loop reentry, and intra-isthmus reentry.
The term “ atrial flutter ” has traditionally been used to refer to a continuously waving pattern on the electrocardiogram (ECG), without an isoelectric baseline in at least one lead, whatever the tachycardia cycle length (TCL). “Typical AFL ” is reserved for a macroreentrant circuit with the activation wavefront rotating clockwise or counterclockwise around the tricuspid annulus and using the CTI as an essential part of the reentry circuit. “ Atypical AFL ” is only a descriptive term for an AT with an ECG pattern of continuous undulation of the atrial complex, different from that in typical AFL. However, the term “ atypical AFL” introduces unnecessary confusion, and a mechanistic description of the AT circuit in relation to atrial anatomy is preferred.
Pathophysiology
Right Atrial Anatomy
The right atrial (RA) endocardial surface is composed of many orifices and embryonic remnants, accounting for an irregular, complex surface. The RA endocardium is architecturally divided into three anatomically distinct regions; each is a remnant of embryological development. The posterior smooth-walled RA, derived from the embryonic sinus venosus, receives the superior vena cava (SVC), the inferior vena cava (IVC), and the coronary sinus (CS). It also contains the fossa ovalis, the sinus node, and the atrioventricular node (AVN). The anterolateral trabeculated RA, derived from the “true” embryonic RA, is lined by horizontal, parallel ridges of muscle bundles that resemble the teeth of a comb (the pectinate muscle). It contains the RA appendage and free wall. The atrial septum is primarily derived from the embryonic septum primum and septum secundum.
The SVC enters the roof of the RA between the base of the RA appendage and the superior margin of the interatrial septum. The IVC enters the posterolateral portion of the floor of the RA along the inferior margin of the interatrial septum. The orifice of IVC is guarded by a fibrous or fibrous-muscular semilunar flap, the valve of the IVC (the eustachian valve). The CS enters the inferior aspect of the RA adjacent to the inferior margin of the interatrial septum, slightly more anterior and medial relative to the orifice of the IVC, and closer to the tricuspid annulus. Commonly, the CS os is guarded by a semilunar valve, the valve of the CS (valve of Thebesius). On the lower third of the interatrial septum lies the fossa ovalis ( eFig. 12.1 ).
The posterior smooth-walled RA and the anterolateral trabeculated RA are separated by the crista terminalis on the lateral wall and the eustachian ridge in the inferior aspect. The sulcus terminalis, where the sinus node is located, is a subtle groove on the epicardial surface of the heart corresponding to the crista terminalis. The crista terminalis is a roughly C -shaped, convex, thick muscular band that runs from the high septum, anterior to the orifice of the SVC superiorly, and courses caudally along the posterolateral aspect of the RA. In its inferior extent, it courses anteriorly to the orifice of the IVC. The crista terminalis can vary in size and thickness, most often appearing as distinct ridge, but occasionally can be a broad, a flat, or a thin structure. As the crista reaches the region of the IVC, it is extended by the eustachian valve and eustachian ridge.
The eustachian valve is the remnant of the embryonic sinus venosus valve, which manifests as a flap of variable thickness and mobility along the orifice of the IVC; this valve can continue as a ridge (eustachian ridge) superiorly along the floor of the RA to the CS os, to join the valve of Thebesius, forming the tendon of Todaro, and then continuing onto the interatrial septum as the inferior limbus of the fossa ovalis. The tendon of Todaro is a fine, tendinous cord that runs as an extension of the free-edge of the eustachian ridge toward the central fibrous body. Its insertion into the central fibrous body marks the position of the compact AVN at the apex of the triangle of Koch.
The tricuspid annulus lies anterior to the body of the RA, and its inferior portion lies a short distance (approximately 1 to 4 cm) anterior to the eustachian ridge, although its course varies among individuals. The CTI is the part of the RA between the ostium of the IVC and the eustachian ridge (posteriorly) and the tricuspid annulus (anteriorly). The CTI runs in an anterolateral-to-posteromedial direction, from the low anterior RA to the low septal RA. The CTI belongs to the trabeculated part of the RA, and its surface is very rough. Its width and muscle thickness are variable, from a few millimeters to more than 3 cm in width and more than 1 cm in depth. The CTI becomes wider in a medial-to-lateral direction, and it is thinnest in its central portion. A thick eustachian ridge (greater than 4 mm) is seen in 24% of patients. The eustachian ridge (often composed of partly or largely fibrous tissue) occurs as an elevation on the CTI. The area between the tricuspid annulus and the eustachian ridge is referred to as the sub-eustachian isthmus, whereas the downslope of the eustachian ridge leads to the junction of the RA and IVC. The pectinate muscles fan out from the crista terminalis and encroach upon the CTI for variable distances, but typically spare the myocardium just atrial to the tricuspid valve. This smooth portion of the CTI is referred to as the vestibular portion. The pectinate muscles are more prominent on the lateral side of the CTI, but become progressively thinner as they ramify branches toward the CS os. In the normal heart, CTI anatomy can be flat, “hilly” (from prominent eustachian ridge and pectinate muscles), concave, or have a pouch-like recess. Occasionally there is a depression (sub-eustachian pouch or sinus of Keith) on the CTI just lateral to the CS os, which can be very deep.
Typical Atrial Flutter Circuit
Typical AFL is the most common type of macroreentrant AT. The macroreentrant circuit is defined by anatomical barriers including the tricuspid annulus, crista terminalis, IVC orifice, eustachian ridge, CS os, and probably the fossa ovalis. However, the lines of conduction block necessary to provide adequate path length for the flutter reentry circuit can be functional or anatomical. The anterior boundary of the tachycardia circuit has been well established as being the tricuspid ring. The posterior boundaries, however, are more complex and not as well-defined. The posterior borders occur at a variable distance from the anterior border, narrowest in the region of the eustachian ridge (CTI) and widest in the anterior part of the RA.
The CTI provides the protected zone of relatively slow conduction necessary for the flutter reentry circuit. The area of slowest conduction is probably localized in the lateral aspect of the CTI in younger patients and in the medial aspect in older patients. Conduction velocity in the CTI during pacing in sinus rhythm is slower in patients with typical AFL compared with those without any history of AFL. The mechanism of the slower conduction velocity in the CTI, relative to the interatrial septum and RA free wall, is uncertain but can be related to the anisotropic fiber orientation. With aging or atrial dilation, intercellular fibrosis can change the density of gap junctions and produce nonuniform anisotropic conduction through the trabeculations of the CTI.
Key to the development of typical AFL is the formation of a line of block in the region between the SVC and IVC in the RA free wall. This line of block acts as a critical lateral boundary that prevents short-circuiting of the flutter wavefront, whereby the reentrant wave catches the “tail of refractoriness” and hence extinguishes. This line of block is usually functional, but may be fixed. The creation of a line of block between the vena cavae or anterior in the RA free wall permits induction of AFL in otherwise normal atria. This explains the observation that AFL most often does not start immediately after a premature atrial complex (PAC) or burst rapid atrial pacing; rather its onset is usually preceded by a transitional rhythm (atrial fibrillation [AF]) of variable duration, which helps induce the functional line of block between the venae cavae. When a fixed (i.e., anatomic) intercaval line of block exists (e.g., atriotomy scar following surgical repair of congenital heart disease), antecedent AF may not be necessary to produce AFL.
The crista terminalis plays an important role as a functional barrier during typical AFL. Conduction delay and rate-related transverse block across the crista terminalis has been consistently observed in sinus rhythm and during pacing. During typical AFL, a line of transverse conduction block along the crista terminalis serving as a lateral boundary can be determined by the presence of double and split potentials recorded. Structural characteristics of the crista terminalis influence transverse conduction; steep slope and arborization of the crista terminalis have been implicated as geometric factors in its transverse conduction block. Typical AFL is more likely to occur in the setting of a thicker and continuous crista terminalis, and these patients are more likely to exhibit transverse crista terminalis conduction block at longer pacing cycle lengths (PCLs), as opposed to controls. Similarly, the region posterior to the crista terminalis (the posterior smooth-walled RA) also has been shown to demonstrate functional transverse conduction block during AFL or rapid pacing.
The width of the activation wavefront in typical AFL varies considerably, and is determined by the distance between the anterior and posterior boundaries at any given part of the circuit. It is very narrow inferiorly at the CTI and substantially wider moving upward. Substantial variability in the upper part of the circuit is a result of the large distance between anterior and posterior borders and anatomical barriers superiorly, combined with variability in the completeness of the posterior border. Recent studies suggest that the posterior block line is located along the posteromedial RA wall, posterior to the crista terminalis. Despite a relatively similar activation sequence, the active circuit (as determined by entrainment mapping) is variable. Most commonly, the reentrant wavefront courses not around the tricuspid annulus but obliquely between anterior and posterior borders away from the tricuspid annulus along any available, more rapidly conducting segments. Consequently, significant portions of the RA, including areas around the tricuspid annulus, can often be passively activated. In many subjects, the upper portions of the circuit pass behind the RA appendage and lie near or at the posterior circuit border, or they bifurcate around the SVC or RA appendage. The posterior border can extend completely or partially between the IVC and the SVC.
Typical AFL is of two types: counterclockwise and clockwise. In counterclockwise AFL (“counterclockwise” as viewed in the left anterior oblique [LAO] view from the ventricular side of the tricuspid annulus), the activation wavefront propagates caudocephalically up the septal side of the tricuspid annulus toward the crista terminalis and advances cephalocaudally down the lateral wall of the RA to reach the lateral tricuspid annulus, after which it propagates through the CTI. In clockwise AFL (also known as “reverse typical” AFL), activation propagates in the direction opposite to that in counterclockwise typical AFL ( Fig. 12.1 ). In both types of typical AFL, the flutter circuit is entirely confined within the RA. Left atrium (LA) activation occurs as a bystander and follows transseptal conduction across the inferior CS-LA connection, Bachmann’s bundle, and/or fossa ovalis.
Counterclockwise AFL is the most common form of typical AFL. Clockwise AFL is observed in only 10% of clinical cases, despite the fact that it is easily inducible with programmed electrical stimulation. Clockwise AFL can be induced in the electrophysiology (EP) laboratory in approximately 50% of patients who clinically present with only counterclockwise AFL. The 9 : 1 clinical predominance of counterclockwise AFL can be related to the localization of an area with a low safety factor for conduction in the CTI, close to the atrial septum. In addition, counterclockwise AFL is more likely to be induced with rapid atrial pacing from the CS os. Conversely, clockwise AFL is more likely to be induced with pacing from the low lateral RA pacing. These observations may be related to the anisotropic properties of the CTI and the development of rate-dependent conduction delays and unidirectional block necessary for tachycardia induction, which may be affected by the site of stimulation.
Interrelationship of Atrial Flutter and Atrial Fibrillation
Although AF and typical AFL frequently coexist, the pathophysiological interrelationship between the two arrhythmias is still uncertain. Clinical AFL occurs in more than one-third of patients with AF. AF often precedes the onset of AFL and can also develop after the successful catheter ablation of AFL. Similar predisposing factors are found in both arrhythmias, including age, hypertension, heart failure, sleep apnea, and chronic pulmonary disease. It is likely that the atrial electrical and structural remodeling that underlies or is induced by AF can also promote the occurrence of AFL, or vice versa. Evidence also suggests that AF plays an important role in the genesis of typical AFL. It is also possible that at least some episodes of AFL can degenerate into AF ( Fig. 12.2 ). AFL with sufficiently short cycle lengths (CLs) can result in fibrillatory conduction, which manifests as clinical AF.
Multiple studies have shown that, in the vast majority of instances of induced or spontaneous AFL, antecedent AF is necessary for the development of AFL ( Fig. 12.3 ). This is because it is during the AF that a critical lateral boundary (i.e., a functional line of block) necessary for the development of AFL forms between the SVC and IVC. Hence, without preceding AF, it is difficult to initiate AFL. Thus it is not surprising that recent studies have implicated pulmonary vein (PV) triggers in the initiation of typical AFL. The PV triggers induce a transitional period of AF, which then organizes to AFL once the functional line of block critical for the flutter circuit is formed. If this functional boundary does not develop, AF will either persist or spontaneously convert back to sinus rhythm. Not infrequently, the use of antiarrhythmic drug therapy (especially class IC agents) facilitates the conversion of AF to AFL, presumably by changing the atrial substrate to favor development of the intercaval line of block that could not form previously without drug effects.
Among patients who initially present with isolated typical AFL, the incidence of AF is very high, around 25 times higher than in the general population, even after the elimination of AFL with catheter ablation. Following AFL ablation, AF can develop in up to 82% of these patients. Hence it appears that AFL is often an early marker of atrial electrical disease that frequently progresses to AF. Eradication of the flutter circuit alone by CTI ablation is not expected to eliminate AF that likely was the underlying trigger for AFL in most patients. After CTI ablation, the atria still are susceptible to PV triggers, which continue to induce AF episodes. After successful ablation of AFL, AF wavefronts can no longer “reorganize” into typical AFL because the CTI (the target for AFL ablation) is no longer available as a critical component of an AFL reentrant circuit, at which time AF either persists or terminates. Frequently, however, AF simply becomes clinically manifest. In fact, it is likely that AF already exists in many patients with supposed “isolated” AFL, but has not been clinically documented because of the preferential organization of AF wavefronts into AFL. Intensified continuous cardiac monitoring frequently confirms the additional existence of AF in many patients with suspected isolated AFL.
On the other hand, in patients presenting with both AF and AFL, the successful elimination of AF was found to prevent the recurrence of both AF and AFL. PV isolation alone was equally as effective as the combined ablation strategy (PV isolation plus CTI ablation), and more effective than CTI ablation alone in providing long-term freedom from both arrhythmias. Even in patients presenting with isolated AFL and no prior history of AF, PV isolation (without CTI ablation) could prevent the recurrence of AFL.
In summary, data suggest that bursts of AF may serve as the primary electrical disorder in patients with typical AFL. These bursts induce a more stable macroreentrant arrhythmia (typical AFL) in susceptible individuals. After eradication of the flutter substrate by CTI ablation, the same PV triggers manifest as AF. PV isolation can potentially control both arrhythmias in patients with AFL, and those with coexistent AFL and AF.
Double-Wave Reentry
A typical AFL circuit with a large excitable gap may allow a second excitation wave to be introduced into the flutter circuit by a critically timed atrial extrastimulus (AES), so that two wavefronts occupy the same circuit simultaneously. This type of AFL is designated “double-wave reentry.”
Double-wave reentry is manifest by acceleration of the tachycardia rate, but with identical surface ECG morphology and intracardiac activation sequence. It can be recognized by the simultaneous activation of the superior and inferior regions of the tricuspid annulus, with all activation being sequential. This rhythm rarely lasts for more than a few beats and can serve as a trigger for AF. Because the CTI is still a necessary part of the circuit, double-wave reentry is amenable to CTI ablation.
Epidemiology and Natural History
It is estimated that the overall incidence of AFL in the United States is 88 per 100,000 person-years. The prevalence of AFL increases with age to almost 600 per 100,000 among those older than 80 years. AFL accounts for approximately 15% of supraventricular arrhythmias. Although in clinical practice AFL appears to be less common than paroxysmal supraventricular tachycardia, population-based data show that in the general population, AFL is diagnosed for the first time more than twice as often. Adjusted for age, the incidence of AFL in men is 2 to 3 times that in women.
Paroxysmal AFL can occur in patients with no apparent structural heart disease, whereas chronic AFL is usually associated with underlying heart disease, such as valvular or ischemic heart disease or cardiomyopathy. At highest risk of developing AFL are men, older adults, and individuals with preexisting heart failure or chronic obstructive lung disease. In approximately 60% of patients, AFL occurs as part of an acute disease process, such as acute pericarditis, acute exacerbation of chronic pulmonary disease, acute pneumonia, thyrotoxicosis, alcoholism, following cardiac or pulmonary surgery, or during acute myocardial infraction (MI).
The natural history of typical AFL is often interlinked with AF. Typical AFL and AF frequently coexist. The majority (75%) of patients with AFL also have documented AF at the time of presentation. Successful CTI ablation to cure AFL does not seem to improve the natural history of progression to AF, even in patients with AFL as the only clinical arrhythmia. Up to 82% of these patients develop AF; the majority of those develop AF in the first year after ablation of AFL.
Clinical Presentation
Patients with AFL can be completely asymptomatic, or they may present with a spectrum of symptoms ranging from palpitations, lightheadedness, fatigue, reduced activity tolerance, or dyspnea, to acute pulmonary edema or acute coronary syndrome in susceptible patients.
The clinical manifestations of AFL strongly depend on the ventricular rate during the AFL, the presence of structural heart disease, and the underlying functional status. Fast ventricular rates and the loss of effective atrial contraction have significant hemodynamic consequences, especially in patients with systolic or diastolic heart failure. Furthermore, AFL with a chronically rapid heart rate can lead to tachycardia-mediated cardiomyopathy and heart failure. In fact, some patients remain asymptomatic until they present with a thromboembolic event or with decompensated heart failure secondary to tachycardia-induced cardiomyopathy. AFL occurs in approximately 25% to 35% of patients with AF, in which case AFL may be associated with worsening of the intensity of symptoms because of more rapid ventricular rates.
Initial Evaluation
Clinical symptoms are not usually helpful in distinguishing typical AFL from other atrial tachyarrhythmias. Documentation of the arrhythmia during spontaneous symptoms on ECG, ambulatory cardiac monitoring, or cardiac implantable electronic devices (loop recorders, pacemakers, defibrillators) is important to establish the diagnosis. The 12-lead ECG diagnosis of typical AFL is frequently accurate, but it can occasionally be misleading (see later).
Initial assessment includes the determination of cardiopulmonary stability, symptom onset and severity, thromboembolic versus hemorrhagic risk, and potential substrates or triggers of AFL. An echocardiogram is necessary to evaluate for structural heart disease. Evaluation for ischemic heart disease is considered in patients with angina, heart failure, or high risk for coronary artery disease. Additional laboratory evaluation typically includes the assessment of serum electrolytes, blood counts, renal and hepatic function, as well as thyroid function.
Principles of Management
Management of AFL should be aimed at identifying and treating underlying causes of the arrhythmia, as well as reducing symptoms, improving quality of life, and preventing cardiovascular morbidity and mortality associated with AFL. In addition, unlike AF, curing AFL is an attainable treatment goal.
There are four main issues that must be addressed in the treatment of AFL: (1) ventricular rate control, (2) restoration of normal sinus rhythm (NSR), (3) maintenance of NSR, and (4) prevention of systemic embolization ( Figs. 12.4 and 12.5 ).
Rate Control
Ventricular rate control during AFL is important to prevent hemodynamic instability, improve symptoms and functional capacity, and prevent tachycardia-mediated cardiomyopathy. Oral or intravenous (IV) AVN blockers are utilized for rate control, depending on the severity of symptoms and the degree of hemodynamic compromise caused by the tachycardia. Notably, the ventricular rate can be very difficult to control in typical AF, more so than during AF, because of the slower and more regular atrial rate (see Fig. 12.2 ). As a consequence, controlling clinical symptoms frequently requires cardioversion.
Beta-blockers or nondihydropyridine calcium-channel blockers (verapamil and diltiazem) are the drugs of choice for rate control, and appear to have equivalent efficacy. Care should be used in administering these medications in patients with acutely decompensated heart failure. Beta-blockers are preferred in patients with cardiomyopathy, ischemic heart disease, and following surgical procedures. Verapamil and diltiazem are preferred in patients with reactive airway disease.
Digoxin is less effective and requires a longer time to achieve rate control, but may be considered if beta-blockers and calcium channel blockers have failed or have intolerable side effects. Digoxin reduces the resting heart rate, but it is seldom effective in ambulatory patients because its effects are mediated by the enhancement of vagal tone, which is offset during exertion. Thus digoxin has traditionally been used as a second-line agent, usually in sedentary patients or those with heart failure or hypotension. Recently, however, several systematic reviews and meta-analyses found that digoxin use was associated with a greater risk for mortality in patients with AF, regardless of concomitant heart failure. Some studies have suggested that AF nullifies the effect of digoxin in reducing hospitalizations for heart failure patients. Hence the long-term use of digoxin has been discouraged.
Amiodarone may be considered for rate control when other AVN blockers are unsuccessful or not tolerated. IV amiodarone is useful for acute control of the ventricular rate, and can be of particular value in acutely ill patients or those with acutely decompensated heart failure or severe hemodynamic compromise. Because of the possibility of termination of AFL by amiodarone, though very small, pericardioversion anticoagulation strategies (as discussed later) should be considered, depending on the individual patient’s risk/benefit profile. Given its potential toxicity, amiodarone is not recommended for long-term rate control.
In patients with AFL and ventricular preexcitation causing rapid ventricular response, prompt direct-current cardioversion is recommended, especially when hemodynamic compromise is present. IV procainamide or ibutilide to restore NSR or to slow the ventricular rate may be considered in hemodynamically stable patients. Importantly, drugs that preferentially slow AVN conduction without prolonging bypass tract refractoriness (such as verapamil, diltiazem, adenosine, oral or IV digoxin, and IV amiodarone) can accelerate the ventricular rate and potentially precipitate hemodynamic collapse and ventricular fibrillation (VF) in high-risk patients with Wolff-Parkinson-White syndrome. Unlike the IV route of administration, chronic oral amiodarone therapy can slow or block bypass tract (BT) conduction. Limited data exist regarding the use of beta-blockers; nonetheless, these drugs theoretically pose a similar potential risk in this situation, and they should be used with caution.
Restoration of Sinus Rhythm
Restoration and maintenance of NSR in patients with AFL is preferred to rate control strategy. The elimination of AFL is associated with relief of symptoms, improved functional status and quality of life, reduced risk of systemic thromboembolism, and prevention of tachycardia-induced cardiomyopathy. In addition, reduction in atrial remodeling can potentially help reduce the future risk of AF. Therefore rate control strategy is reserved to patients with contraindication to anticoagulation, those with intraatrial thrombi, or patients with very poor functional status and multiple comorbidities when the risks associated with rhythm control strategy outweigh the benefits.
Several options are available for termination of AFL, including external direct-current therapy, antiarrhythmic drugs, overdrive atrial pacing, and catheter ablation. The timing of attempted cardioversion is influenced by the duration of AFL, the severity of the patient’s symptoms, the adequacy of rate control, and the risk of thromboembolism. Prompt cardioversion is recommended for patients with rapid ventricular rates and hemodynamic compromise attributed to AFL (hypotension, acute heart failure, myocardial ischemia) or ventricular preexcitation. Cardioversion is also considered to restore NSR in stable but symptomatic patients with persistent AFL, especially when ventricular rate control remains suboptimal. For stable patients with adequate heart rate control and minimal symptoms, conversion to sinus rhythm may be deferred until catheter ablation, if performed in a timely manner.
Peri-Cardioversion Anticoagulation
In spite of the organized atrial rhythm and apparent preserved atrial contraction during AFL, the thromboembolic risk is practically no different for this rhythm than for AF. In stable patients with AFL of a duration longer than 48 hours or of unknown duration, any mode of cardioversion (electrical, chemical, pacing, or ablation) should be delayed until the patient has been anticoagulated at appropriate levels for 3 to 4 weeks or transesophageal echocardiography (TEE) has excluded atrial thrombi. TEE may also be considered in patients with high thrombotic risk (e.g., severe valvular or congenital heart disease, prior thromboembolic events, severe cardiomyopathy), even when the duration of AFL is less than 48 hours.
If urgency of cardioversion (because of severe symptoms or hemodynamic instability) precludes TEE, therapeutic doses of low-molecular-weight heparin or unfractionated heparin should be administered as soon as possible, concurrent with or, preferably, prior to cardioversion.
Electrical Cardioversion
External direct-current cardioversion of AFL is successful in more than 95% and is typically achieved with relatively lower levels of energy (i.e., 5 to 50 joules) as compared with AF. In general, electrical cardioversion is preferred to chemical cardioversion, given the higher efficacy and the lower low risk of proarrhythmia; however, it requires sedation or anesthesia, and is contraindicated in patients with digitalis toxicity or those with hypokalemia.
Overdrive Atrial Pacing
Overdrive atrial pacing can terminate AFL in about 82% (range 55% to 100%), and is especially effective in patients receiving antiarrhythmic medications. Overdrive pacing is particularly useful in patients with preexisting atrial pacing wires (as part of a permanent pacemaker or defibrillator, or temporary epicardial pacing wires following cardiac surgery). In these patients, overdrive atrial pacing may be preferred to electrical cardioversion since it obviates the need for sedation. Insertion of a temporary pacing wire for overdrive pace termination of AFL may be considered when electrical cardioversion is contraindicated (e.g., in the setting of digitalis toxicity) or when sedation is not feasible. Transesophageal atrial pacing, on the other hand, is rarely utilized, since it requires sedation and esophageal intubation, and is much less effective than electrical cardioversion.
Overdrive atrial pacing is started at a rate 5% to 10% faster than the flutter atrial rate and is maintained for 15 or more seconds. Overdrive pacing is repeated at incrementally faster rates until NSR is restored as AF develops. When overdrive atrial pacing alone fails, high-frequency (50-Hz) burst atrial pacing or overdrive pacing with atrial extrastimuli can be effective. One potential drawback to atrial pacing is the potential conversion of AFL to AF; even then, induction of AF can be associated with better ventricular rate control and less symptom, and may subsequently revert spontaneously to NSR.
Pharmacological Cardioversion
Pharmacological cardioversion of AFL is generally less effective than electrical cardioversion and carries the potential risk of proarrhythmia, but can be an option when sedation is not available or not well tolerated or when indicated by patient preference.
Ibutilide and dofetilide are the most effective agents for pharmacological conversion of AFL. Other antiarrhythmic agents, including sotalol, amiodarone, class IA (e.g., procainamide), or class IC agents (e.g., flecainide, propafenone), have limited efficacy. AVN blockers (beta-blockers, digoxin, and calcium-channel blockers) are generally not effective for restoration of NSR.
IV ibutilide is the drug of choice for chemical cardioversion; it can terminate AFL (usually within 30 minutes) in 38% to 76% of cases, regardless of the duration of the arrhythmia. The efficacy of ibutilide is significantly higher than that of IV procainamide (76% vs. 14%), IV sotalol (70% vs. 19%), and IV amiodarone (87% vs. 29%). However, ibutilide is associated with sustained polymorphic ventricular tachycardia (VT) in 1.2% to 1.7% of cases, and nonsustained VT in 1.8% to 6.7%, which is more likely to occur in patients with reduced left ventricular ejection fraction (LVEF). The risk of VT lingers for 6 or even 8 hours after ibutilide administration, regardless of whether it has terminated AFL; thus patients should not be given this medication in an emergency room setting without observing them on continuous heart rhythm monitoring for this duration. Pretreatment with magnesium can increase the efficacy and reduce the risk of torsades de pointes.
Oral dofetilide is also effective for conversion of AFL (70% to 80% conversion rate), more so than in AF. In the majority of patients, conversion to NSR is achieved within 36 hours. However, dofetilide can be associated with proarrhythmia, and its initiation requires continuous cardiac monitoring for a minimum of 72 hours. IV dofetilide, which is not available in the United States, also appears to be effective for conversion of AFL, and was found to have significantly higher efficacy than IV amiodarone (75% vs. 10%).
Because of the limited efficacy and the associated risk of proarrhythmia, chemical cardioversion is generally reserved for selected patients, especially when electrical cardioversion is not feasible because of contraindication to sedation. In addition, when the use of long-term antiarrhythmic medications is planned for the maintenance of NSR, starting drug therapy before electrical cardioversion can be beneficial, as it can help restore NSR in some patients and obviate the need for electrical cardioversion and, in other cases, can potentially enhance the efficacy of electrical cardioversion. In addition, it can help maintain NSR after successful cardioversion and, if a side effect develops that would preclude long term use of the drug, it can be stopped and a different medication tried before the electrical cardioversion is performed.
Importantly, adequate rate control with AVN blockers (beta-blockers, diltiazem, verapamil) should be achieved before instituting class IA (procainamide and disopyramide) or IC drugs (propafenone and flecainide), which can potentially slow the flutter rate and hence facilitate 1 : 1 AV conduction and paradoxically faster ventricular rates.
Catheter Ablation
Catheter ablation is the definitive treatment for AFL, and it is a reasonable option for restoration of NSR for stable patients who do not require immediate cardioversion and can wait until this procedure is performed. In fact, the presence of the arrhythmia at the time of the procedure helps reliably establish the diagnosis and mechanism of the clinical arrhythmia and differentiate it from other arrhythmias that might be inducible by programmed electrical stimulation but may not be of clinical significance.
Maintenance of Sinus Rhythm
When AFL occurs as part of an acute disease process, such as hyperthyroidism, acute MI, pulmonary embolism, or following cardiac surgery, chronic therapy for the arrhythmia is usually not required after sinus rhythm is restored. In patients with no underlying reversible disorder, the risk of arrhythmia recurrence after initial restoration of NSR is high and hence strategies to maintain NSR should be considered. When long-term rhythm control is required, catheter ablation is superior to antiarrhythmic drugs and is the preferred strategy in most patients.
Catheter Ablation
Catheter ablation is recommended as first-line therapy for most patients with symptomatic or recurrent typical AFL, whether paroxysmal or persistent. The ablation procedure is associated with high long-term success rates (92% after a single procedure and 97% after multiple procedures), and low risk of serious complications (0.4%). In addition to improvement of symptoms and quality of life, successful ablation offers a potential cure of the arrhythmia, reducing the risk of thromboembolism, and potentially eliminating the need for long-term anticoagulation and antiarrhythmic drug therapy.
Antiarrhythmic Drug Therapy
Complete maintenance of NSR often is unachievable with current drug therapy. The average 1-year recurrence rate associated with dofetilide is more than 35%, and is even higher for flecainide (approximately 50%). Data are limited regarding the efficacy of other drugs, as most studies combined AFL with AF, with the vast majority of the patients having AF, and without specifying the results for each arrhythmia. Given the significant superiority of catheter ablation and its low complication rate, long-term antiarrhythmic drug therapy is no longer recommended for most patients with AFL.
For patients in whom catheter ablation is no feasible, or when long-term antiarrhythmic drug therapy is preferred, the choice of the antiarrhythmic agent is similar to that used for rhythm control in AF ( see Chapter 15 ). The selection of pharmacological agents (sotalol, dofetilide, amiodarone, flecainide, and propafenone) is largely driven by the safety profile, and should consider coexisting sinus nodal or AVN disease, heart failure, associated therapies, and comorbidities. The presence and extent of concomitant cardiovascular disease have to be carefully considered. A safer, although possibly less efficacious, drug is usually recommended before resorting to more effective but less safe therapies.
Prevention of Thromboembolism
AFL is associated with increased risk of systemic thromboembolism, although likely to a lesser degree than in AF. The risk factors for development of embolic events in AFL are similar to those described for AF. Short-term stroke risk following cardioversion of AFL ranges from 0% to 7%, and the annual thromboembolism rate in patients with sustained AFL is approximately 3%. Therefore the indications for long-term and for peri-cardioversion anticoagulation in patients with AFL are the same as those in patients with AF ( see Chapter 15 ).
Management of Coexistent Atrial Flutter and Atrial Fibrillation
AF and AFL frequently coexist in the same patient. Clinical AFL occurs in more than one-third of patients with AF. In these patients, when AF is the predominant arrhythmia, catheter ablation of AFL unlikely would improve clinical outcome. On the other hand, when AFL is the dominant clinical arrhythmia, or when new-onset AFL develops in patients with AF after administering antiarrhythmic drug therapy, hybrid therapy (CTI ablation plus antiarrhythmic medications) can be effective. Ablation of the CTI eliminates the flutter and provides good symptomatic relief, while antiarrhythmic drug therapy is continued to suppress AF. Alternatively, a combined approach of PV antrum isolation plus CTI ablation may be considered. Notably, PV isolation alone (without CTI ablation) was found in small studies to be equally effective in the suppression of both AF and AFL in these patients.
Importantly, the incidence of AF is very high among patients who initially present with isolated AFL, even after ablation of the CTI. Given the fact that up to 82% of patients with “isolated” AFL eventually develop AF, some investigators have proposed additional PV isolation at the time of CTI ablation to improve long-term freedom of AF recurrences. Although small studies suggested some value of this approach, this strategy cannot be recommended for general adoption before confirmation by larger studies because of the increased procedural risk.
Anticoagulation therapy is commonly discontinued 1 month after a successful AFL ablation if no other atrial arrhythmias are apparent. However, given the high risk of development of new-onset AF, which is often asymptomatic, patients with typical AFL undergoing successful ablation seem to have an elevated future risk for strokes. Hence intensified cardiac monitoring and anticoagulation therapy may be necessary in this patient population, especially those with significant risk factors (such as obstructive sleep apnea and LA dilation) for developing AF. This is especially important because intermittent and symptom-based surveillance with ambulatory cardiac monitoring is inaccurate and unreliable for identifying patients with AF. However, the current practice guidelines do not provide adequate recommendations regarding the optimal monitoring and anticoagulation strategies for this patient population.
Electrocardiographic Features
Flutter Waves
Flutter waves appear as atrial complexes of constant morphology, polarity, and CL. Typically, flutter waves are most prominent in the inferior leads (II, III, aVF) and lead V1. In counterclockwise AFL ( Fig. 12.6 ), the flutter waves in the inferior leads resemble a picket fence (sawtooth) because these leads are primarily negative. This pattern consists of a downsloping segment, followed by a sharper negative deflection, and then a sharp positive deflection, with a positive overshoot leading to the next downsloping plateau. The relative size of each component can vary markedly. The flutter waves can exhibit pure negative deflections in the inferior leads, negative and then positive deflections that are equal in size, or a small negative and then a larger positive deflection. Those three varieties coexist with tall positive, small positive, or biphasic P waves in lead V1, respectively. With progression across the precordium, the initial component rapidly becomes inverted and the second component isoelectric usually by lead V2 to V3. This produces the overall impression of an upright flutter wave in lead V1, which becomes inverted by lead V6. A negative deflection always precedes the positive deflection in the inferior leads in counterclockwise AFL, and the degree of positivity in the inferior leads appears to be related to the coexistence of heart disease and LA enlargement. Lead I is low-amplitude isoelectric, and lead aVL is usually upright.
The initial part of negative deflections showing gradual voltage decline coincides with activation of the CTI. This short period of relative electrical silence on the surface ECG is related to the small amount of tissue being activated within the isthmus. Then, the sharp negative deflection of flutter wave is caused by caudocranial activation of the interatrial septum and the LA. On the other hand, the upstroke of flutter wave and terminal deflection (from the nadir to terminal deflection point) represent craniocaudal activation in the RA free wall ( eFig. 12.2 ). The absence of positive terminal deflection has been correlated with a greater extent of low voltage area in the RA free wall. In addition, the overall amplitude of the flutter wave, which can be a determinant of the size of positive terminal deflection, depends on the longitudinally directed vector of the RA free wall activation.
The surface ECG appearance of clockwise typical AFL is more variable than that of counterclockwise typical AFL, but in many respects, clockwise AFL presents an inversion of the appearance in counterclockwise AFL (see Fig. 12.6 ). Clockwise AFL generally has broad positive deflections in the inferior leads, with characteristic notching; however, there is an inverted component preceding the upright notched component. Depending on the amplitude of this component, the appearance can be of continuous undulation without an obviously predominant upright or inverted component. On other occasions, it may appear that the inverted component is dominant, thus superficially mimicking counterclockwise AFL. Lead V1 is characterized by a wide negative and usually notched deflection. There is transition across the precordium to an upright deflection in lead V6. Lead I is usually upright, and lead aVL is low-amplitude negative and notched.
Typical AFL usually has an atrial rate of 240 to 340 beats/min. However, the flutter rate can be slower in patients with conduction delays in the atrial circuit secondary to prior incomplete CTI ablation ( Fig. 12.7 ), scars from prior cardiac surgery, or antiarrhythmic drugs, whereby flutter CLs as long as 450 milliseconds (i.e., atrial rate less than 150 beats/min) can be observed. If the ventricular response is half the atrial rate, it can be difficult to identify flutter waves “buried” within the QRS or T waves ( Fig. 12.8 ). Close inspection of the QRS and T waves, and comparisons with ECGs obtained in normal sinus rhythm, can help identify buried flutter waves. Furthermore, vagal maneuvers and AVN blockers can slow AV conduction and unmask the flutter waves ( eFig. 12.3 ).
In patients who have undergone extensive LA ablation for treatment of AF, the P wave morphology during typical CTI-dependent AFL can be very different from the prior description because of alteration of intraatrial and interatrial wavefront propagation. Similarly, non–CTI-dependent macroreentrant ATs can mimic typical CTI-dependent AFL on the surface ECG. Thus arrhythmias that appear to be typical AFL may not be, whereas others that are actually typical AFL may not appear to be.
Atrioventricular Conduction
In general, AV conduction during AFL is characterized by even-integral conduction ratios; for example, with an atrial rate 300/min, the ventricular response is often 150/min or 75/min, not 90 to 100/min, as is often seen in AF. Most commonly, 2 : 1 AV conduction is present during AFL. Variable AV conduction and larger multiples (e.g., 4 : 1 or 6 : 1) are not uncommon. When present, variable AV conduction is the result of multilevel block; for example, proximal 2 : 1 AV block and more distal 3 : 2 Wenckebach block result in 5 : 2 AV Wenckebach block.
Slowing the atrial rate during AFL caused by antiarrhythmic drugs or following a prior incomplete CTI ablation can result in a paradoxical increase in the ventricular rate caused by better AVN conduction of the slower flutter beats ( Fig. 12.9 ). For this reason, it is essential that adequate AVN blockade with beta-blockers or calcium channel blockers is in place before administering antiarrhythmic drugs that can slow the atrial rate during flutter. Rapid 1 : 1 AV conduction is most commonly seen in patients with anterogradely conducting AV BTs ( Fig. 12.10 ), but it may also be present in cases of enhanced AVN conduction secondary to high sympathetic tone (e.g., exercise, sympathomimetic drugs).
QRS Morphology
The QRS complex during AFL is often identical to that during sinus rhythm. However, the atrial impulses can be aberrantly conducted because of functional bundle branch block, most frequently right bundle branch block (RBBB) (see Fig. 12.9 ). Even with normal ventricular conduction, the QRS complex can be slightly distorted by temporal superimposition of flutter waves on the QRS complex. Thus the QRS complex can appear to acquire a new or larger R, S, or Q wave.
Electrophysiological Testing
Typically a decapolar catheter (positioned into the CS with the proximal electrodes bracketing the CS os) and a multipolar (20 or 24 pole) Halo catheter (positioned at the tricuspid annulus) are used to map typical AFL. The distal tip of the Halo catheter is positioned at 6 to 7 o’clock in the LAO view, so that the distal electrodes will record the middle and lateral aspects of the CTI, the middle electrodes will record the anterolateral RA, and the proximal electrodes may record the RA septum (depending on the catheter used and RA size). Instead of the Halo and CS catheters, some operators use a single duodecapolar catheter around the tricuspid annulus, thus extending the catheter tip inside the CS. Such a catheter can straddle the CTI and provide recording and pacing from the medial and lateral aspects of the isthmus, assuming good catheter-tissue contact at these locations. In the latter arrangement, however, the body of the duodecapolar catheter crossing over the CTI can potentially hinder manipulation and positioning of the ablation catheter tip to achieve adequate tissue contact for effective ablation.
Induction of Tachycardia
Programmed electrical stimulation protocol typically involves atrial burst pacing from the high RA and CS (down to the PCL, at which 2 : 1 atrial capture occurs) and single and double AESs (down to the atrial effective refractory period [ERP]) at multiple CLs (600 to 200 milliseconds) from the high RA and CS. Administration of an isoproterenol infusion (1 to 4 µg/min) may be required to facilitate tachycardia induction.
AFL can be induced readily with programmed electrical stimulation in most patients with a clinical history of AFL. Reproducible initiation of counterclockwise AFL is possible in more than 95% of patients. Rapid atrial pacing is more likely to induce AFL than a single AES, but as likely as introducing two AESs. On the other hand, the frequency of single or double AESs initiating AFL is low in patients without a history of AFL (less than 10%). Counterclockwise AFL is more likely to be induced by stimulation from the CS os; conversely, clockwise AFL is more likely to be induced with low lateral RA pacing. Induction of AFL usually occurs once unidirectional CTI block develops during pacing ( Fig. 12.11 ). Not infrequently, a run of AF of variable duration is induced first, which then converts into AFL. The faster the pacing rate and the shorter the AES coupling intervals, the more likely it will be that AF is induced, which is usually self-terminating but can be sustained in less than 10% of patients with no clinical history of AF. The significance of induction of AF in these patients is uncertain.