Abstract
Atypical atrial flutters refer to a spectrum of arrhythmias that vary in location and circuit dimensions. These arrhythmias are often recognized after catheter ablation or surgical treatment of atrial fibrillation (AF) or after surgery that involves a right or left atriotomy. Atypical atrial flutter requires fixed or functional barriers and regions of slow conduction. Activation mapping is used to demonstrate reentrant circuits, middiastolic potentials, fractionated potentials, and double potentials. Electroanatomic mapping systems and multipolar electrode catheters are useful in most cases. Difficulties may arise in defining complex reentrant circuits, spontaneous conversion to other atrial arrhythmias, and achieving durable lines of block in the atrium.
Keywords
atypical atrial flutter, catheter ablation, intracardiac mapping, macroreentrant atrial tachycardia, microreentrant atrial tachycardia
Key Points
- •
Atypical or nonisthmus-dependent atrial flutter and atrial macroreentry requires fixed or functional barriers and regions of slow conduction.
- •
Atypical flutters are often recognized after catheter ablation or surgical treatment of atrial fibrillation (AF) or after surgery that involves right or left atriotomies.
- •
Activation mapping is used to demonstrate reentrant circuits, middiastolic potentials, fractionated potentials, and double potentials.
- •
A multipolar electrode catheter is useful to map atypical flutter in either atrium.
- •
An electroanatomic mapping system is useful in most cases.
- •
An irrigated-tip or large-tip (8 mm) ablation catheter is usually employed.
- •
Sources of difficulty include defining complex reentrant circuits, particularly those with irregular oscillations in atrial cycle length; achieving durable lines of conduction block; and spontaneous conversion of atypical flutter to atrial fibrillation or other arrhythmias.
Introduction
Atypical atrial flutters refer to a spectrum of arrhythmias that vary in location and circuit dimensions. In many cases these arrhythmias may also coexist with atrial fibrillation (AF) or play transitional roles in the initiation or termination of AF, or in the transformation to typical atrial flutter. Atypical flutters are increasingly recognized after catheter ablation or surgical treatment of AF or after surgery that involves right or left atriotomies.
Historically there has been confusion over the use of the terms atrial flutter and atrial tachycardia (AT). It has been proposed that these terms be applied based on rate or the presence of isoelectric intervals on the electrocardiogram. The term atrial flutter is often used to refer to regular tachycardias with rates of 240 beats per minute or higher and lacking an isoelectric interval between deflections, whereas atrial tachycardia refers to arrhythmias with isoelectric intervals between P waves and/or slower cycle lengths. However, these characteristics are not specific for the mechanism of tachycardia. In this chapter, the term atypical atrial flutter will be used preferentially to refer to reentrant circuits that are not dependent on the cavotricuspid isthmus (CTI). Atypical flutters may be classified based on their chamber of origin, site within a given atrium, and size of the reentrant circuit ( Table 12.1 ).
| Right Atrium | ||
|---|---|---|
| Circuit | Boundaries | Clinical Scenarios |
| Upper-loop reentry | SVC and crista terminalis | De novo or with CTI-dependent flutter |
| Right atrial free-wall reentry | Right atrial scar, lateral tricuspid annulus | Prior right atriotomy, atrial myopathy |
| Dual-loop reentry (combined lower loop, upper loop, free wall, and/or tricuspid valve) | Combinations of right atrial scar, SVC, tricuspid annulus, with or without CTI | De novo or with CTI-dependent flutter, prior right atriotomy, atrial myopathy |
| Left Atrium | ||
| Circuit | Boundaries | Clinical Scenarios |
| Perimitral annular reentry | Mitral annulus, left atrial isthmus | Following left atrial ablation for AF, after maze surgery, de novo |
| Roof-dependent reentry | Linear ablation lines around left and/or right pulmonary veins, possibly electrically active pulmonary vein tissue | Following left atrial ablation for atrial fibrillation with encircling lesions, after maze surgery, de novo |
| Periseptal tachycardia | Right pulmonary veins and mitral annulus or fossa ovalis with right pulmonary veins or mitral annulus | Following atriotomy, left atrial ablation for atrial fibrillation, de novo |
| Miscellaneous | ||
| Circuit | Boundaries | Clinical Scenarios |
| Lesional tachycardias | Surgical scars | Following left atriotomy, post maze surgery |
| Microreentry or localized reentry | Variable | Following left atrial ablation for AF, after maze surgery, atrial myopathy |
| Coronary sinus mediated reentry | Coronary sinus with left or right atrial myocardium | De novo |
Anatomy
The specific configuration of a reentrant circuit is dependent on the local or global anatomy of the atrium, as well as conduction and refractory properties of the atrial myocardium. As a general rule, reentry requires the presence of two limbs, which are anatomically and/or functionally contiguous but dissociated. These limbs are dependent on the presence of an inexcitable central barrier (i.e., atrioventricular annulus, venous ostium, or scar) or a functional line of block. In the right atrium (RA), natural barriers to conduction are the tricuspid annulus, inferior vena cava (IVC), superior vena cava (SVC), crista terminalis, Eustachian ridge, and fossa ovalis. In the left atrium (LA), the mitral annulus and pulmonary venous ostia serve as critical conduction barriers, along with electrically silent areas that may be found in myopathic atria or result from prior ablation or surgery.
Pathophysiology
Although atypical atrial flutters can arise in structurally normal hearts, these arrhythmias predominantly occur in patients with organic heart disease, after cardiac surgery, or AF ablation. In patients with organic heart disease, the pathogenesis is thought to involve elevated atrial pressure, which causes interstitial fibrosis resulting in conduction slowing and block. LA reentry that arises de novo in the atria frequently involves regions of patchy scar, which presumably occurs as the result of atrial myopathy. These areas of patchy scar can be identified as electrically silent areas during electroanatomic mapping. Other important determinants are slowing of atrial conduction velocity, heterogeneity of atrial refractoriness, and initiating or triggering foci.
In response to premature stimuli, arcs of functional block develop and—if of sufficient length—initiate and support reentry. Functional lines of block develop in structures such as the crista terminalis and the Eustachian ridge, both of which play a critical role in the formation of CTI-dependent atrial flutter. Gaps in these functional lines of block, as in the crista terminalis, permit formation of atypical circuits. Functional lines of block may develop in many other locations, including variable sites in the LA. A combination of fixed and functional block has been demonstrated in animal models of lesional tachycardia, in which a line of functional block develops as an extension to a fixed anatomic lesion. In this case, the anatomic lesion might not be large enough to support reentry, but the combined fixed and functional barrier provides the critical path length needed to maintain reentry. Formation and breakdown of these arcs of block are responsible for interconversion of flutter circuits with each other and the transition to AF.
With the growth of ablation procedures for AF, atypical flutters have become increasingly frequent after both catheter and surgical ablation. Discontinuities in ablation lines can result in unidirectional rate-dependent block or conduction slowing. Furthermore, areas of excluded myocardium, such as those occurring with wide encircling pulmonary vein isolation, give rise to large central barriers that permit circus movement reentry. Many reentrant ATs are complex circuits that involve dual loops (figure-of-eight reentry) or less commonly triple loops. In multiple loop tachycardias, conduction times are nearly equal around the individual loops, but sometimes a dominant circuit exists as a driver, which entrains other loops.
Small reentrant circuits, known as localized reentry or microreentry , can occur in a variety of clinical circumstances, such as catheter ablation of AF, surgical maze procedures, and surgical repair of congenital heart disease. These small circuits can also occur de novo in diseased atria. A unifying condition in cases of localized reentry is the presence of an atrial scar, and these circuits usually localize near prior ablation lines or surgical incisions. Disruption of the myocardial architecture may give rise to markedly slow conduction, anisotropic differences in conduction velocity, and unidirectional block, which allow small circuits to develop and perpetuate.
Cardiac surgery that involves atrial incisions is an important cause of atypical flutter. The location and extent of an incision, in addition to intrinsic disease of the atria, influence the likelihood of developing atypical flutter. Patients who have right atrial surgery are also at risk of developing atypical flutter involving the surgical scar in the lateral RA, whereas those who have a left atriotomy may develop atypical flutter in the LA. In addition, patients who have right or left atriotomies are also prone to develop CTI-dependent flutter.
Diagnosis
Macroreentrant Versus Focal Atrial Tachycardia
A fundamental consideration in evaluating atypical flutter is to establish if the arrhythmia is macroreentrant or focal in origin ( Box 12.1 ). In general, P wave durations are shorter and surface isoelectric intervals are longer with focal tachycardias than with macroreentry. A study of 75 patients with atypical flutters after AF ablation found that a P wave duration of less than 185 ms had a sensitivity of 85% and specificity of 97% in identifying a small reentrant circuit. However, it is possible for a macroreentrant tachycardia to have an isoelectric interval if part of circuit consists of a slowly conducting isthmus generating little electric force. When the tachycardia cycle length varies by 15% or more, a focal mechanism is suggested, but a regular cycle length can occur with both focal and macroreentrant atrial tachycardias (ATs).
Macroreentrant Tachycardias
- •
Entrainment with fusion (with last beat entrained but not fused)
- •
Electroanatomic mapping of >90% of tachycardia cycle length with adjacent early and late areas of activation
- •
Insensitivity to adenosine (in dose sufficient to cause AV block)
Microreentrant Tachycardias or Localized Reentry
- •
Fragmented electrograms that encompass a large proportion of the tachycardia cycle length on single electrode or multiple closely spaced electrodes
- •
Entrainment with progressively longer postpacing intervals at increasing distances from the apparent source
- •
Insensitivity to adenosine (in dose sufficient to cause AV block)
Left Atrial Tachycardias
- •
Passive conduction in the right atrium, with early septal activation and fusion of wave fronts in the RA lateral wall
- •
Absence of RA activation during long segments of the cycle length (mapping <50% of tachycardia cycle length)
- •
Large variations in the RA cycle length with a relatively fixed cycle length in the LA
- •
Entrainment pacing at multiple sites in the RA yielding postpacing intervals >30 ms
AV, Atrioventricular; LA, left atrium; RA, right atrium.
The electrocardiographic (ECG) morphology of atypical flutters is highly variable, and substantial overlaps exists between flutter morphologies arising from different anatomic circuits, so that localization of atrial flutters from the surface ECG has limited utility. However, some broad generalizations can be made that distinguish CTI-dependent from non-CTI-dependent flutters, as well as right from left atrial tachycardias. CTI-dependent flutters often show characteristic morphologies, and recognizing deviation from these patterns allows one to conclude that a tachycardia might be atypical flutter. Typical counterclockwise CTI-dependent flutter often has a superiorly directed flutter wave axis with a small terminal positive deflection in the inferior leads (so called sawtooth pattern). Lead V 1 has an overall positive deflection with an initial isoelectric or inverted component, and the flutter wave transitions from positive to negative across the precordium. Clockwise CTI-dependent flutter is more variable, but it characteristically shows positive, notched flutter waves in the inferior leads, and lead V 1 usually shows negative flutter waves. It should be emphasized that even CTI-dependent flutters can show variability in ECG morphology, and the typical patterns described above might also arise from non-CTI-dependent mechanisms. Typical CTI-dependent flutter that occurs after ablation for AF may also present with atypical ECG morphologies.
Generally flutters that are predominately negative in V 1 arise from the RA, whereas those that show broad positive waves in V 1 (without isoelectric or negative initial components) usually arise from the LA ( Fig. 12.1 ). Left atrial flutters usually demonstrate either positive flutter waves in the inferior leads or low-amplitude/isoelectric signals in the limb and other precordial leads. Intermediate patterns that do not fit these various descriptions are common and can be difficult to localize. Regardless of the morphology, confirmation of the tachycardia origin with endocardial mapping is required.
Evidence for a macroreentrant mechanism can be obtained through manifest entrainment, entrainment with concealed fusion (concealed entrainment), or electroanatomic mapping (see Mapping section). Manifest entrainment is recognized as fixed fusion at any given paced cycle length and by progressive fusion with progressively rapid overdrive pacing. Because the surface flutter wave may not be visible or may be obscured by ventricular depolarization or repolarization, intracardiac electrograms provide a surrogate marker for orthodromic and antidromic capture and the degree of fusion. Manifest entrainment with progressive fusion implies a dimensionality to the tachycardia circuit that includes a separate entrance and exit site, a condition that does not exist for focal tachycardia and is unlikely to be demonstrated for microreentry. The pacing site is considered to be within the tachycardia circuit when the postpacing interval after entrainment is 30 ms (or less) greater than the tachycardia cycle length ( Fig. 12.2 ). Entrainment from two opposite quadrants (such as septal and lateral LA, or anterior and posterior LA), each with return cycle lengths within 30 ms of the tachycardia cycle length, indicates the presence of a macroreentrant circuit. Thus in-circuit entrainment from the septal and lateral LA often indicates perimitral reentry; in-circuit entrainment from the anterior and posterior LA is consistent with a roof-dependent LA flutter. Entrainment may also give indirect information about the distance of a pacing site from the reentrant circuit, with shorter postpacing intervals indicating closer proximity to the circuit.
The response to adenosine is useful in distinguishing reentrant from focal ATs, in that it exerts mechanistic-specific effects on atrial arrhythmias. With rare exceptions, adenosine does not terminate the vast majority of reentrant atrial arrhythmias, whereas it either terminates focal ATs attributed to triggered activity or transiently suppresses focal AT as a result of enhanced automaticity ( Fig. 12.3 ). Adenosine-insensitive ATs with apparently focal activation patterns show characteristics of localized reentrant circuits, including low-amplitude, long-duration fractionated electrograms, and can be entrained. This simple tool provides a reliable means for establishing a tachycardia mechanism before mapping ( Fig. 12.4 ).
Localization of Reentry to the Right Atrium or Left Atrium
The activation pattern of the coronary sinus (CS) is useful in localizing ATs. RA tachycardias typically exhibit proximal-to-distal activation in the CS, unless activation proceeds predominantly over Bachmann bundle (or the catheter is positioned very distally in the CS), in which case the distal CS may be activated earlier. LA tachycardias show a variety of activation patterns depending on the particular reentrant circuit ( Fig. 12.5 ). Some LA tachycardias, such as clockwise perimitral annular reentry and tachycardias originating near the lateral LA (by the left pulmonary veins or appendage), exhibit distal-to-proximal CS activation. However, a proximal-to-distal sequence does not necessarily localize the arrhythmia to the RA and may be seen with counterclockwise perimitral reentry or tachycardias involving the septum or right pulmonary veins. Chevron and reverse chevron activation patterns occur with roof-dependent LA macroreentry. The chevron pattern occurs when a wave front descends the posterior wall and activates the midposterior mitral annulus, and then propagates in both septal and lateral directions along the inferior LA. The reverse chevron pattern occurs when a wave front descends the anterior LA wall and then fuses in the posterior mitral annulus. Focal arrhythmias in the posterior or anterior LA may also produce chevron or reverse chevron patterns. It should be recognized that activation of the CS may be dissociated from the LA endocardium because of a muscular sleeve, which envelops the CS and is attached to the LA through discrete connections. Electrograms recorded in the CS may show disparate activation patterns, and careful analysis may show that the endocardial LA is activated in a pattern different from the CS. Macroreentry involving the CS musculature or the ligament of Marshall as a critical part of the circuit has also been described.
It is possible to establish a diagnosis of LA tachycardia during mapping in the RA and thus identify the need for transseptal catheterization and LA mapping. Criteria for identifying an LA origin through intracardiac mapping include the following:
- 1.
Passive conduction into the RA, which may be demonstrated as fused wave fronts in the lateral wall of the RA.
- 2.
Earliest RA activation in the septum, typically in the region of Bachmann bundle or the CS ostium, especially if areas of early activation encompass a large area.
- 3.
Large variations in the RA cycle length with a relatively fixed cycle length in the LA, implying LA/RA dissociation or conduction block.
- 4.
Entrainment pacing at multiple sites in the RA (including the CTI and the RA free wall) yielding postpacing intervals greater than 30 ms.
Although fusion of wave fronts in the lateral wall of the RA is common during LA tachycardias, it is possible to record a single wave front mimicking counterclockwise or clockwise atrial flutter. This situation depends on (1) the location of the multipolar catheter in the lateral wall, (2) the location of conduction breakthrough from the LA (i.e., preferential conduction over Bachmann bundle or the CS), and (3) the presence of conduction block in the CTI. Typically, activation time in the RA is substantially less than the tachycardia cycle length. Exceptions occur when the tachycardia cycle length is short or if conduction is substantially slowed in the RA, incorrectly implying the presence of a RA tachycardia.
Mapping
Activation Mapping
Conventional activation mapping with multielectrode catheters is usually employed to define the mechanisms of atypical flutters. In the RA, a multielectrode catheter positioned around the tricuspid annulus can provide information suggestive about the tachycardia mechanism. For example, counterclockwise lower-loop reentry reveals lateral-to-septal activation in the CTI and areas of breakthrough in the lateral RA, which can cause fusion of wave fronts or an ascending wave front in the lateral wall (see Fig. 12.3 ).
Double potentials, which usually signify lines of block, may be identified through conventional activation mapping. If reentry proceeds around a line of block, double potentials are widely split in the middle of the line, and they progressively narrow toward the end of the line where the wave front pivots. An example of this may be found in RA free-wall macroreentry ( Fig. 12.6 ). Middiastolic and fragmented potentials can be recorded within critical zones of slow conduction, but verification of participation in the tachycardia circuit is desirable through other means, such as entrainment maneuvers.
Activation mapping in the LA relative to a fixed reference can identify or exclude mitral reentry or roof-dependent flutter. Opposite activation sequences in the superior and inferior mitral annulus (e.g., lateral-to-septal activation along the superior annulus, and septal-to-lateral activation along the inferior annulus) identify perimitral reentry. By contrast, similar directions of activation (e.g., lateral-to-septal) in both the superior and inferior annulus exclude perimitral reentry. Opposite activation along the anterior and posterior walls is seen with roof-dependent reentry.
Entrainment Mapping
Concealed entrainment is an essential tool for identifying sites that participate in a reentrant arrhythmia. Entrainment criteria for identifying appropriate sites for ablation are intended to localize protected zones within a circuit. These include (1) concealed entrainment (with P wave and intracardiac activation sequences resembling those in tachycardia), (2) a postpacing interval within 30 ms of the tachycardia cycle recorded at the pacing site, and (3) a stimulus to P wave interval during pacing equal to the electrogram to P wave interval during tachycardia. Because the P wave may not be visible in many cases of macroreentrant AT, identifying the initial inscription of the P wave may be arbitrary, and an intracardiac reference is often used as a surrogate.
Limitations or pitfalls in using concealed entrainment must be recognized: (1) rate-related conduction slowing may occur, and therefore the postpacing interval might exceed the tachycardia cycle length; (2) failure to capture might occur in some critical regions of a reentrant circuit; (3) acceleration or termination of AT with pacing may occur. Inadvertent termination or slowing can be minimized by pacing 20 ms or less shorter than the tachycardia cycle length.
Electroanatomic Mapping
Electroanatomic mapping provides direct visualization of a reentrant circuit, which is defined as the shortest distance of continuous activation comprising the tachycardia cycle length. A hallmark of macroreentrant arrhythmias is the presence of areas of early activation adjacent to late regions, with intermediate values connecting these two regions (see Figs. 12.5 and 12.6 ). In practice, it is necessary to account for 90% or more of the tachycardia cycle length to visualize a reentrant circuit. Displaying activation information as an isochronal map may clearly demonstrate the direction of wave-front propagation, which is perpendicular to the isochronal steps ( Fig. 12.7 ). Ultrahigh-resolution mapping uses proprietary multipolar mapping catheters to collect a large number of activation points and can identify critical isthmuses of reentrant circuits that are suitable for ablation.
A substantial percent of the tachycardia cycle length may be occupied by conduction through a diseased segment with low-amplitude and fractionated electrograms. If a large part of the cycle length is not accounted for in the map, a careful search for such electrograms should be undertaken to identify a gap in the reentrant circuit; such sites are often characterized by fragmented, low-amplitude electrograms and are appropriate targets for ablation ( Fig.12.8 ). The specific choice of assigning activation times to long duration or split electrograms can greatly affect the appearance of an electroanatomic map and may potentially obscure the delineation of a reentrant circuit. New automated mapping technology uses algorithms that integrate bipolar and unipolar electrograms with neighboring activation times to annotate complex electrograms. By displaying an atrial voltage map, areas of scar can be identified that facilitate the localization of channels that form potential reentrant circuits. Areas with no detectable voltage (defined as the noise limit of recording systems) represent a dense scar and thus a region of fixed conduction block.
Entrainment mapping may be combined with electroanatomic mapping to define critical components of a reentrant circuit. This combined technique is especially useful in situations where the electroanatomic map is ambiguous and it is difficult to distinguish critical components from bystander regions.
Misinterpretation of electroanatomic maps may occur unless care is taken to avoid the following pitfalls ( Table 12.2 ):
- 1.
Incomplete mapping and low resolution: It is important to acquire activation points during tachycardia with sufficient density and to sample different regions in the atrium of interest. Interpolation of activation times in the map may lead to misinterpretation of the rhythm and failure to identify critical components of the circuit. Advanced mapping technology that constructs high-density maps with shorter acquisition times is useful in addressing this pitfall.
- 2.
Fractionated electrograms: If highly fractionated and wide potentials are present, it might be difficult to assign an activation time. In this case, the critical isthmus might not be identified, and the reentrant arrhythmia might be confused with a focal rhythm.
- 3.
Central obstacles or conduction block: Failure to identify areas of scar or central obstacles to conduction may confuse interpretation of an electroanatomic map because interpolation of activation times through areas of conduction block may give the appearance of wave front propagation and obscure the reentrant circuit. If this occurs, it is impossible to identify a critical isthmus to target for ablation. Adjusting the voltage threshold for identifying areas of scar may ameliorate this. It should be recognized that very low-voltage areas (<0.05 mV) can serve as critical components of tachycardia circuit. Mapping with a relatively large-tip ablation catheter with widely spaced bipolar electrograms may not identify low-amplitude signals, which are more easily recorded with smaller, closely spaced electrodes and mapping systems that generate low noise levels. A line of block can be inferred if there are adjacent regions with wave front propagation in opposite directions, separated either by a line of double potentials or dense isochrones.
- 4.
Conduction delay in either atrium: Slow conduction or local block may prolong activation in either the RA or LA, and conduction time in the passively activated chamber may approach the tachycardia cycle length. For example, if conduction block is present in the CTI, an LA tachycardia can give rise to a craniocaudal activation pattern of the lateral RA ( Fig. 12.9 ). Even if conduction is present in the CTI, RA activation in the lateral wall may be craniocaudal if conduction occurs over Bachmann bundle. In these situations, entrainment is an important adjunctive tool, which would define the RA as a bystander and clarify the electroanatomic map.
