Author + information
- Received October 24, 2016
- Revision received January 4, 2017
- Accepted January 4, 2017
- Published online September 18, 2017.
- Bhupesh Pathik, MBBSa,b,
- Geoffrey Lee, MBChB, PhDa,b,
- Frédéric Sacher, MD, PhDc,
- Pierre Jaïs, MDc,
- Grégoire Massoullié, MDc,
- Nicolas Derval, MDc,
- Matthew G. Bates, MBChB, PhDa,
- Jonathan Lipton, MD, PhDa,
- Stephen Joseph, MBBS, PhDa,d,
- Joseph Morton, MBBS, PhDa,b,d,
- Paul Sparks, MBBS, PhDa,b,
- Peter Kistler, MBBS, PhDb,e,f and
- Jonathan M. Kalman, MBBS, PhDa,b,∗ ()
- aDepartment of Cardiology, Royal Melbourne Hospital, Melbourne, Australia
- bFaculty of Medicine, Dentistry, and Health Sciences, University of Melbourne, Melbourne, Australia
- cDepartment of Cardiac Electrophysiology, LIRYC Institute, INSERM 1045, Bordeaux University Hospital, Pessac, France
- dDepartment of Cardiology, Western Health, Footscray, Victoria, Australia
- eDepartment of Cardiology, Alfred Hospital, Melbourne, Australia
- fBaker IDI Heart and Diabetes Institute, Melbourne, Australia
- ↵∗Address for correspondence:
Dr. Jonathan Kalman, Royal Melbourne Hospital, Department of Cardiology, Grattan Street, Parkville, Victoria 3052, Australia.
Objectives Using high-resolution 3-dimensional (3D) mapping, the aim of this study was to further characterize right atrial macro–re-entrant tachycardias and answer unresolved questions in the understanding of this arrhythmia.
Background Despite advances in understanding of the mechanisms of right atrial macro–re-entrant tachycardias, many questions lack definitive answers. The advent of high-resolution 3D mapping provides an opportunity to gain further insights into the nature of these common circuits.
Methods A total of 25 patients with right atrial macro–re-entrant tachycardia were studied. High-resolution 3D mapping (Rhythmia mapping system, Boston Scientific, Natick, Massachusetts) was performed. Regional voltage and conduction velocity were determined. Maps were analyzed to characterize wave front propagation patterns in all atrial regions. The relationship between substrate and conduction was evaluated.
Results A total of 42 right atrial macro–re-entrant circuits were observed. The most common location of the posterior line of block was the posteromedial right atrium (73%). This line of block continued superiorly into the superior vena cava, taking an oblique course to finish on the anterior superior vena cava aspect in 73%. Conduction delay at the crista terminalis was less common (23%). Conduction slowing or block was seen at the limbus of the fossa ovalis (73%) and Eustachian ridge (77%). Highly variable and localized areas of slow conduction were also observed in the inferior septum (45%), superior septum (27%), anterosuperior right atrium (23%), and lateral right atrium (23%). Localized conduction slowing was seen in the cavotricuspid isthmus in 50% of patients, but there was no generalized conduction slowing in this isthmus. The voltage in regions of slow conduction was significantly lower compared with areas of normal conduction velocity (p < 0.001). Conduction channels were observed in 55% of patients.
Conclusions High-resolution 3D mapping has provided new insights into the nature of right atrial macro–re-entrant tachycardias. Variable regions of abnormal atrial substrate were associated with conduction slowing and block. Individual variation in propagation patterns was observed in association with this variable substrate. (Mapping of Atrial Arrhythmias Using High Spatial Resolution Mapping Catheters and the Rhythmia Mapping System; ACTRN12615000544572)
- atrial macro–re-entry
- cavotricuspid isthmus
- macro–re-entrant atrial tachycardia
- three-dimensional electroanatomic mapping
Macro–re-entrant atrial tachycardia is a common arrhythmia that was first described almost a century ago (1,2). Our understanding of this arrhythmia has continued to evolve with progressive advancements in technology (3–6). Initial entrainment studies using fluoroscopy were instrumental in characterizing the major components of the right atrial macro–re-entry circuits (7–9). Subsequent studies using 3-dimensional (3D) electroanatomic mapping systems have provided further insights, particularly in the setting of abnormal atrial anatomy, multiple circuits, and regions of scar (10–15).
However, despite the progress in our understanding, there are still questions that lack definitive answers:
• Is conduction through the cavotricuspid isthmus (CTI) slow (16–18)?
• Is the posterior line of block at the crista terminalis or more medially in the posterior right atrium?
• What are the patterns of activation in the septum and around the coronary sinus (CS) os?
• How often is dual-loop re-entry present, and what are the anatomic locations of propagating wave fronts?
• Are there other regions of abnormal conduction, and what is the heterogeneity in this apparently stereotypic arrhythmia?
• Is there a relationship between activation patterns and underlying atrial substrate?
The recent advent of high-resolution 3D electroanatomic mapping that automatically annotates electrograms and allows the rapid acquisition of thousands of points provides a level of detail and resolution not previously achievable (19–21). Using high-resolution 3D mapping, we sought to further characterize right atrial macro–re-entrant circuits and answer the aforementioned unresolved questions in our understanding of this arrhythmia.
A total of 25 patients with right atrial macro–re-entrant tachycardia undergoing catheter ablation at 2 institutions were studied. All patients gave written informed consent, and the study protocol was approved by the local research and human ethics committee.
Antiarrhythmic medications were discontinued for 5 half-lives prior to the procedure. All patients underwent transesophageal echocardiography if not anticoagulated prior to the procedure to exclude left atrial thrombus. Procedures were performed under either conscious sedation or general anesthesia. A decapolar catheter was placed in the CS in all patients and a hexapolar catheter at the His location according to operator discretion. For patients in sinus rhythm at the time of procedure, burst atrial pacing was used to induce the macro–re-entrant atrial tachycardia. According to operator discretion, in selected cases, a duodecapolar catheter was placed in the right atrium encircling the tricuspid annulus. Intracardiac electrograms and 12-lead surface electrocardiograms were recorded simultaneously on a computerized digital amplifier system (EPMed Systems, West Berlin, New Jersey; or Labsystem Pro, Bard, Tewksbury, Massachusetts). Bipolar electrograms were filtered at 30 and 300 Hz. Unipolar electrograms were filtered at 1 and 300 Hz. Unipolar signals were referenced to the indifferent electrode in the inferior vena cava (IVC).
3D electroanatomic mapping
3D electroanatomic mapping was performed with the Rhythmia mapping system (Boston Scientific, Natick, Massachusetts). Bipolar activation maps were created during macro–re-entrant atrial tachycardia in all 25 patients and during proximal CS pacing in sinus rhythm following ablation in 7 patients. The Rhythmia mapping system uses both magnetic and impedance-based localization. Two location references are used: a stationary electrode reference (commonly the CS catheter) and a magnetic location reference back patch. The IntellaMap-Orion (Boston Scientific) mapping catheter was used to create the 3D electroanatomic map. The IntellaMap-Orion has 64 flat microelectrodes (0.8 mm in diameter) with a 2.5-mm interelectrode distance (center to center) in a basket configuration with 8 splines. The catheter is 8.5-F, with bidirectional deflection and variable degrees of basket deployment (basket diameter ranges from 3 to 22 mm) that can be adjusted on the basis of individual patient anatomy.
The Rhythmia mapping system allows real-time automated signal analysis. Automated mapping is mediated through continuous mapping via user-defined beat acceptance criteria. Cardiac beats were selected for inclusion in the map on the basis of multiple criteria, including: 1) cycle length stability (±5 ms); 2) relative timing of a reference electrogram positioned in the CS; 3) respiratory gating; and 4) mapping catheter motion. Chamber surface geometry was created using the basket catheter gated to cardiac and respiratory cycles. We only included electrograms within 2 mm of the surface geometry.
For activation time, the electrograms were automatically annotated at the maximum negative dV/dt for unipolar signals or the peak sharp of the bipolar electrogram. If the electrogram had multiple components, using an algorithm known as Intelligent Annotation, the system would take into account not only the amplitude but also the timings of the electrograms in the surrounding area to select the potential to annotate. On the basis of the timings of the activation map, a propagation map was generated of the atrial macro–re-entrant circuit.
CTI-dependent macro–re-entrant atrial tachycardia was defined as a macro–re-entrant circuit around the tricuspid annulus and dependent on propagation through the CTI.
Upper loop re-entry was defined as a macro–re-entrant atrial tachycardia involving the upper portion of the right atrium that is independent of the CTI.
Lower loop re-entry was defined as macro–re-entrant atrial tachycardia localized to the lower right atrium that is CTI dependent with breakthrough across the lower portion of the posterior line of block.
Conduction gap was defined as the remnant of conducting tissue bounded on both sides by regions of conduction block as determined from the propagation map.
Double potentials were defined as bipolar atrial electrograms with 2 discrete deflections per beat separated by either an isoelectric baseline or a low-amplitude interval (22).
Global and regional atrial voltage analysis
The median global right atrial bipolar voltage was compared between patients. Regional voltage analysis of the right atrium was performed by dividing the chamber into the following 10 segments: anterosuperior, anterolateral, anteroseptal, CTI, posterior superior, posterior inferior, lateral superior, lateral inferior, septal superior, and septal inferior. In the left anterior oblique view, medial to the superior vena cava (SVC) was defined as septal and lateral to the SVC was defined as lateral. In the right anterior oblique view, anterior and posterior segments were defined in relation to the sulcus terminalis. The median bipolar voltage of each segment was calculated.
Regional conduction velocity
The mean conduction velocity in each of the 4 segments around the tricuspid annulus (anterosuperior, anterolateral, anteroseptal, and CTI) was calculated in patients with CTI-dependent macro–re-entrant atrial tachycardia. Conduction velocity between 2 points was determined by expressing the distance between the points as a function of the difference in the local activation times. Isochronal maps of the atrium were created at 10-ms intervals in local activation times to determine regional conduction velocity. Conduction velocity for each segment was determined as the mean of the conduction velocity among 10 pairs of points along the direction of wave front activation. The mean conduction velocity across all patients was compared among the 4 segments around the tricuspid annulus. Atrial conduction slowing was defined as a local conduction velocity of 10 to 20 cm/s and conduction block as <10 cm/s (23).
Localizing posterior line of block
In patients with CTI-dependent macro–re-entrant atrial tachycardia, propagation maps were reviewed for the location of the posterior line of conduction block. In a subset of patients, maps were also created during stable CS pacing in sinus rhythm, and the site of the posterior line of block was compared with the activation map during the tachycardia. The posterior line of block was defined as the region with local conduction velocity <10 cm/s (23) with the presence of double potentials and change in activation sequence from craniocaudal to caudocranial. The anatomic region of the crista terminalis was represented by the sulcus terminalis, which was clearly defined on the high-resolution 3D electroanatomic maps as a groove from the anterior aspect of the SVC to the anterior aspect of the IVC. The sinus venosus was defined as the region of the posterior right atrium between the SVC and IVC located posteromedial to the crista terminalis (sulcus terminalis). Propagation maps were also analyzed for the presence of breakthrough across the posterior line of block.
Anatomic location of circuits and wave front propagation in relation to the SVC
The number of active circuits in each patient was identified using the propagation maps and confirmed with entrainment mapping. In addition, propagation maps were analyzed for the location of the upper turn around point of the circuit with respect to the SVC.
In all patients with CTI-dependent macro–re-entrant atrial tachycardia, activation maps were analyzed for the site and extent of passive activation into the SVC. Regions and patterns of slow conduction and block involving the SVC were also determined.
Analysis of voltages at zones of slow conduction
In patients with CTI-dependent macro–re-entrant atrial tachycardia, propagation maps were reviewed for areas of slow conduction with isochronal crowding and conduction block. The anatomic location of these sites of slow conduction and block, the pattern of their distribution, and their effect on the direction of wave front propagation were determined. The mean bipolar voltage in these zones of slow conduction and block was determined within each patient and compared with the mean bipolar voltage of adjacent areas with normal conduction velocity determined within each patient.
All statistical analysis was performed using SPSS version 23.0 (IBM, Armonk, New York). Normality of all quantitative variables was checked using the Shapiro-Wilk test. Continuous variables are reported as mean ± SD or as median and interquartile range (IQR) as appropriate. Categorical variables are reported as numbers and percentages. Comparisons of conduction velocity between different segments, voltage between slow conduction and normal conduction velocity, or voltage between within and periphery of conduction channels were performed using analysis of variance with patient as a random effect. A p value <0.05 was considered to indicate statistical significance.
A total of 25 patients with right atrial macro–re-entrant tachycardia underwent 3D electroanatomic mapping (Table 1). The mean age was 60 ± 13 years. The median CHA2DS2-VASc score was 1 (IQR: 0 to 2). Twelve patients (48%) had histories of cardiac surgery: 4 patients (16%) had prior coronary artery bypass grafting, 7 patients (28%) had previous mitral valve repair, and 1 patient had a history of patch closure for congenital ventricular septal defect. Among the 7 patients with prior mitral valve repair, the left atrium was accessed via Waterston’s (interatrial) groove with a direct left atriotomy performed in 6 patients. In 1 patient, the surgical report was no longer available for review. In the patient with a history of patch closure of congenital ventricular septal defect, no atriotomy was performed (only a right ventricular incision was performed). In no documented case was a right atriotomy performed. The mean number of points collected for the 3D electroanatomic map was 18,877 ± 8,959, acquired in a mean of 22 ± 11 min. The baseline characteristics of patients with and without prior cardiac surgery are shown in Table 2. In both groups, median CHA2DS2-VASc (congestive heart failure, hypertension, age ≥75 years, diabetes, previous stroke, vascular disease, age 65 to 74 years, and female sex) score was 1 (cardiac surgery [IQR: 1 to 2] vs. no cardiac surgery [IQR: 0 to 2]).
Distribution of atrial macro–re-entrant circuits
A total of 42 right atrial macro–re-entrant circuits were observed in 25 patients. Mean tachycardia cycle length (TCL) was 260 ± 37 ms. Twenty (48%) re-entry circuits were consistent with counterclockwise CTI-dependent macro–re-entrant atrial tachycardia, 2 (5%) clockwise CTI-dependent macro–re-entrant atrial tachycardia, and 1 example of intraisthmus re-entry (2%). There were 3 cases (7%) of upper loop re-entry, 15 of lower loop re-entry (36%), and 1 lateral wall circuit (2%).
Conduction velocity around the tricuspid annulus and in the CTI
The overall mean conduction velocity in the CTI was 111 ± 13 cm/s, and there was no significant difference compared with the other segments around the tricuspid annulus (p = 0.64). However, 9 of 18 patients with CTI-dependent macro–re-entrant atrial tachycardia (without prior ablation) demonstrated highly localized regions of conduction slowing in the CTI, with a mean width of 0.30 ± 0.01 cm (lateral isthmus, n = 2; central isthmus, n = 2; septal isthmus, n = 5). In the other 9 patients, there was no regional conduction slowing in the CTI.
Anatomic location of the posterior line of block in CTI-dependent macro–re-entrant atrial tachycardia (counterclockwise and clockwise)
In patients with CTI-dependent macro–re-entrant atrial tachycardia, there was variability in the location of the posterior line of block. The most common site (16 of 22 patients) was a relatively straight SVC-to-IVC line in the posterior (sinus venosus) right atrium located posteromedial to the crista terminalis (sulcus terminalis) (Figures 1A and 1B, Online Video 1). In 6 patients, the line was more lateral and occurred in the region of the crista terminalis. In this latter group of patients, the line of block was less discrete, and double potentials could be recorded over a broader zone of up to 0.9 ± 0.2 cm. Discrete double potentials were observed along the posterior line of block in all patients (Figure 1C).
In 7 patients with CTI-dependent macro–re-entrant atrial tachycardia, electroanatomic maps of the right atrium were created during stable CS pacing at 600- and 300-ms cycle lengths after reversion to sinus rhythm. The posterior line of block was present in the same anatomic location as during tachycardia and at both cycle lengths (Figure 2, Online Video 2). In 3 of the 7 patients, the line of block was intact along its entire length, and in 4 patients there was functional break across the superior aspect of the line in the same location that conduction break occurred during tachycardia.
Extension of the posterior line of block into the SVC
A highly stereotypical line of block into the SVC was observed as the superior continuation of the posterior line of block in 16 of 22 patients (73%). This passed into the SVC obliquely to complete as a horizontal line passing to the septal aspect of the anterior SVC (Figures 1E and 1F, Online Video 1). This superior extension of the line of posterior block into the SVC prevented an active wave front from bypassing the superior extent of the posterior line of block. In 6 patients, insufficient detail was present in the SVC map to characterize this line.
Superior turnaround site of the CTI-dependent macro–re-entrant atrial tachycardia circuit: Relationship to SVC
In all patients with CTI-dependent macro–re-entrant atrial tachycardia, there was an active superior turnaround point of the circuit located anterior to the SVC. The width and location of this turnaround was highly variable, with 5 of 22 wave fronts (23%) closer to the tricuspid annulus and 8 of 22 wave fronts (36%) closer to the SVC in this region. In 9 of 22 patients (41%), the wave front was broad and equidistant to the tricuspid annulus and SVC.
In addition, in 13 of 22 cases (59%) of CTI-dependent macro–re-entrant atrial tachycardia, there was evidence of active breakthrough across the superior portion of the posterior line of block resulting in a double-loop pattern of activation (1 anterior and 1 posterior to the SVC) (Figure 3, Online Video 3). In a further 2 of 22 patients, this breakthrough across the posterior line was located inferiorly.
Conduction patterns at the eustachian ridge and in relation to the CS ostium
In 12 of 22 patients with CTI-dependent macro–re-entrant atrial tachycardia, a line of conduction block was carried from the Eustachian ridge to be continuous with a line of block posterior to the CS os (Figure 4A, Online Video 4). Along this line of conduction block, double potentials were observed (Figure 4C). In these patients, the line of block directed the activation wave front anterior to the CS os along the tricuspid annulus as it exited the CTI.
In 5 of 22 patients, slow conduction occurred across the Eustachian ridge, exemplified by the presence of fractionated bipolar electrograms. In the remaining 5 patients, normal conduction was observed across this zone between the Eustachian ridge and the posterior CS os without any evidence of conduction slowing. In these patients, the active wave front traveled both anterior and posterior to the CS os.
Limbus of fossa ovalis as a site of conduction slowing and block
In 16 of 22 patients (73%) with CTI-dependent macro–re-entrant atrial tachycardia, conduction slowing or block was observed at the anatomic region of the right atrium corresponding to the limbus of the fossa ovalis (Figures 4A and 4D, Online Video 5). This anatomic location was demonstrable on the maps and catheter location at the site of block confirmed as the limbus using transesophageal echocardiography. During CTI-dependent macro–re-entrant atrial tachycardia, this region appeared as a zone of slow conduction in 8 of 16 patients (50%) and as a line of complete block in 8 of 16 patients (50%). Fractionated bipolar electrograms were observed at the limbus in the 8 patients with conduction slowing and double potentials in the 8 patients with complete block (Figure 4F). In 6 patients, there was no evidence of conduction slowing or block at the limbus of the fossa ovalis.
Septal wall activation patterns in counterclockwise CTI-dependent macro–re-entrant atrial tachycardia
The nature of septal activation was determined by the nature of the lines of block at the Eustachian ridge, CS os, and fossa ovalis and resulted in 3 distinct patterns. In the first pattern, linear block was continuous from the Eustachian ridge and the posterior CS os to the limbus of the fossa ovalis driving the wave front anterior to these structures before it broadened at the superior aspect of the limbus in 5 of 20 patients with counterclockwise CTI-dependent macro–re-entrant atrial tachycardia (Figure 4A, Online Video 4).
In the second pattern, conduction block was not continuous from the Eustachian ridge and CS os to the limbus of the fossa ovalis, allowing the ascending wave front to divide and continue both anterior and posterior to the limbus (Figure 4D, Online Video 5). This pattern was observed in 10 of 20 cases of counterclockwise CTI-dependent macro–re-entrant atrial tachycardia. In the third pattern, because of an absence of conduction block between the Eustachian ridge and CS os or at the limbus, a single broad wave front ascended the septum (5 of 20 cases) (Figure 4G).
Global and regional right atrial voltage analysis
In each patient, there was a marked variation in median bipolar voltage from segment to segment. For example, in 1 patient, the median bipolar voltage in the lateral inferior segment was 2.5 mV (IQR: 0.5 to 6.7 mV) compared with 0.3 mV (IQR: 0.1 to 1.4 mV) in the posterior inferior segment. Between patients, this did not demonstrate any stereotypic regional pattern but also showed marked variability (as to which right atrial region was high or low voltage). For example, in 1 patient, the median bipolar voltage in the lateral inferior segment was 8.9 mV (IQR: 4.2 to 13.3 mV), whereas in another patient, the median bipolar voltage in the same segment was 0.80 mV (IQR: 0.3 to 4.2 mV). Between patients, there was also marked variation in absolute bipolar voltage such that there was no apparent “normal range” in this population. The greatest median global right atrial bipolar voltage was 2.1 mV (IQR: 0.7 to 4.5 mV). The lowest median global right atrial bipolar voltage was 0.2 mV (IQR: 0.1 to 0.8 mV). A moderate negative correlation was observed between median global right atrial bipolar voltage and TCL (r = −0.56, p = 0.03). The lower the median global right atrial bipolar voltage, the longer the TCL.
CTI-dependent macro–re-entrant atrial tachycardia: Variable localized areas of slow conduction: Correlation with localized low voltage
In all patients with CTI-dependent macro–re-entrant atrial tachycardia, unexpected and very localized regions of slow conduction were observed in variable areas of the right atrium. We described earlier that 9 patients with CTI-dependent macro–re-entrant atrial tachycardia had localized conduction slowing in the CTI and that 16 patients had conduction slowing or block at the limbus of the fossa ovalis. In addition, localized slow conduction was observed in the inferior septum (10 of 22 patients), superior septum (6 of 22 patients), anterior superior right atrium in a linear pattern from SVC to tricuspid annulus (10 of 22 patients), and lateral right atrium (5 of 22 patients) (Figure 5). The bipolar voltage was found to be significantly lower in regions of slow conduction compared with surrounding areas of normal conduction velocity (mean 0.4 ± 0.1 mV vs. 3.4 ± 1.5 mV; p < 0.001) (Figure 6, Online Videos 6A and 6B). In patients with counterclockwise CTI-dependent macro–re-entrant atrial tachycardia, the electrocardiogram was stereotypical, and we did not observe any obvious correlation with the variable regions of slow conduction present in the right atrium.
Conduction channels in the right atrium: Correlation with voltage channels
In 55% of patients (12 of 22) with CTI-dependent macro–re-entrant atrial tachycardia, there were examples of conduction occurring preferentially through channels of relative normal voltage when bordered by areas of relative low voltage or scar (Figure 4B, Online Video 4). In the majority of cases (7 of 12), the conduction channel was located in the superior septum. Other locations of the conduction channels included the inferior septum (3 of 12), the anterior superior right atrium (4 of 12), and the lateral wall (3 of 12 patients). The bipolar voltage within the conduction channel was significantly greater than the bipolar voltage along the borders of the conduction channel (mean 2.7 ± 2.1 mV vs. 0.3 ± 0.2 mV; p < 0.001).
Upper loop re-entry
In 3 patients, a re-entrant circuit that could broadly be described as rotating around the base of the SVC was observed. Two of these patients had histories of mitral valve repair. The remaining patient had no structural heart disease, prior cardiac surgery, or catheter ablation. In these 3 patients, there was also a line of conduction block in the sinus venosus region of the posterior right atrium with an oblique conduction channel located in the superior aspect of this line of block (Figures 7A to 7C, Online Video 7). Within the channel in all cases a broad local fractionated bipolar electrogram was recorded (Figure 7D). Ablation at this site terminated the tachycardia within 5 s in all 3 patients.
Comparison of patients with and without prior cardiac surgery
Table 3 summarizes the distribution of atrial macro–re-entrant circuits, location of the posterior line of block, and regions of conduction slowing or block in patients with and without prior cardiac surgery. The distribution of right atrial macro–re-entrant circuits was the same in both groups, with counterclockwise CTI-dependent macro–re-entrant atrial tachycardia the most common arrhythmia. In addition, in both groups, the posteromedial right atrium (sinus venosus) was the location of the posterior line of block in a majority of patients. Conduction slowing or block at the limbus of the fossa ovalis and Eustachian ridge as well as localized zones of slow conduction and low voltage were observed in both groups in a similar frequency.
The present study, using ultra-high-density mapping, builds on the extensive existing research describing the nature of right atrial macro–re-entrant circuits to provide important refinements in our understanding of the arrhythmia mechanism. We made the following observations.
1. Conduction velocity in the CTI was not uniformly slower than the other segments around the tricuspid annulus.
2. The posterior line of conduction slowing and block in CTI-dependent macro–re-entrant atrial tachycardia was most commonly located in the sinus venosus region of the posterior right atrium running linearly from the SVC to the IVC. The crista terminalis was less frequently the anatomic site of posterior conduction delay or block.
3. In approximately 60% of patients with CTI-dependent macro–re-entrant atrial tachycardia, there was a conduction gap across the superior portion of the posterior line of block resulting in 2 active wave fronts, 1 traveling anterior and 1 posterior to the SVC (lower loop).
4. The right atrium was characterized by localized zones of low voltage and scar as well as conduction channels that determined the direction of wave front propagation. These occurred in relation to anatomic structures such as the limbus of the fossa ovalis and the Eustachian ridge but also in other regions such as the anterior right atrium, lateral wall, septum, and CTI.
Conduction velocity in the CTI
In the present study, using high-resolution mapping, we did not observe uniform conduction slowing in the CTI. We did see highly localized conduction slowing in varying regions of the CTI in 50% of patients, but such regional conduction slowing was not unique to the isthmus. Our findings are in contrast to those of early studies using fluoroscopy, which identified the CTI as a zone of slow conduction (16–18). These studies used much lower resolution mapping than the present study and fluoroscopy to identify anatomy, and these issues may explain the different results. The CTI, as the narrowest portion of the CTI-dependent macro–re-entrant atrial tachycardia circuit, may nevertheless be more likely to develop unidirectional conduction block during rapid atrial pacing than other atrial regions (24).
Anatomic location of posterior line of block
A posterior line of conduction slowing or block in the right atrium has been identified as pivotal for the maintenance of CTI-dependent macro–re-entrant atrial tachycardia by preventing a short-circuiting of the active wave front by rapid propagation across the posterior right atrium (9). Prior studies using intracardiac echocardiography, fluoroscopy, and low-resolution mapping have made conflicting observations, with some suggesting that the posterior line of block is at the crista terminalis (7,9,25) and others observing block in the posteromedial (sinus venosus) right atrium (26,27). However, in the present study, we identified that this line of block occurred in the posteromedial right atrium in the majority of patients and was less commonly observed at the crista terminalis. Importantly, the line of block continued superiorly into the SVC muscle sleeve coursing obliquely to finish as an anterior horizontal line. This SVC continuation prevents bypass of the posterior line of block using the SVC muscle sleeve and has not previously been described.
Of note, these earlier studies (7,9,25–27) did not uniformly demonstrate conduction slowing of block along either location of the posterior line during atrial pacing. In contrast, although we observed variability, with some patients showing fixed and others functional block, in all patients the pattern correlated closely with that observed during the tachycardia.
Work in swine hearts observed rate-dependent block at the posteromedial sinus venosus right atrium at the site of abrupt change in muscle fiber orientation and changes in wall thickness and collagen content (28) to explain why conduction block occurs at this region. In a study of autopsied human hearts, Gami et al. (29) observed a second ridge (in addition to the crista terminalis) in approximately 20% of patients, which corresponded with the posteromedial line of block. The frequency described in this work is lower than observed in our study population, likely because the autopsy study was not restricted to patients with macro–re-entrant atrial tachycardia.
Upper loop re-entry
Although the circuit of “upper loop” re-entry may be quite variable, we observed that there was a critical region of conduction slowing in all patients that corresponded to a channel through the posterior line of block located in sinus venosus region of the posterior right atrium. This finding is in contrast to earlier work regarding the mechanism of upper loop re-entry. Using 3D noncontact mapping with the balloon catheter, Tai et al. (30) delineated the re-entrant circuit around the SVC with conduction through the gap in the crista terminalis. Either may be possible in different patients, but it should be emphasized that the current technology provides a combination of anatomic and electrophysiological detail not previously attainable.
Superior turnaround location of the CTI-dependent macro–re-entrant circuit
Consistent with prior studies in patients with CTI-dependent macro–re-entrant atrial tachycardia (9,31,32), we observed that the superior activation wave front propagated anterior to the SVC. However, even here there was considerable variability, with the active wave front at times a narrow band immediately anterior to the SVC or immediately adjacent to the tricuspid annulus and in other patients a true broad wave front between these structures. In approximately 60% of these patients, the wave front also traveled posterior to the SVC, with evidence of breakthrough across the posterior line of block creating a dual-loop pattern, as previously documented (14,33–35).
Limbus of fossa ovalis as a zone of conduction delay
In more than 70% of cases of CTI-dependent macro–re-entrant atrial tachycardia, a zone of conduction slowing or block was observed at the limbus of the fossa ovalis. This region of block was continuous with the Eustachian ridge to CS os line in 25% of patients in whom the wave front was therefore directed along the anterior region of the septum adjacent to the tricuspid annulus. The limbus is a muscular ridge that represents the embryological remnant of the lower margin of the septum secundum and has been shown to be a zone of abrupt change of myocardial fiber orientation (36), which may explain the observed conduction changes. To our knowledge, conduction slowing at the limbus has not previously been described, but its relevance to the arrhythmia mechanism remains to be demonstrated.
Regions of slow conduction and block: Determinants of wave front direction during right atrial macro–re-entrant tachycardia
In this study, we demonstrated the presence of localized regions of conduction slowing corresponding with areas of relative voltage reduction. These areas occurred in relation to anatomic structures (such as the limbus of the fossa ovalis) but frequently at other varied sites throughout the atrium and were present in all patients in this series. These regions were responsible in part for the heterogeneity observed in the course of the dominant wave front. Whether these local regions of conduction slowing are critical for initiation or maintenance of the macro–re-entrant circuit is unknown but certainly would be expected to stabilize a circuit by increasing the excitable gap. The combination of anatomic regions of conduction slowing and block together with localized zones of low voltage and slow conduction appear to represent the substrate necessary for arrhythmia maintenance.
In other patients, conduction channels through regions of relatively preserved voltage surrounded by areas of conduction block and low voltage were critical for the arrhythmia mechanism, such as in upper loop re-entry. Recent studies have suggested that substrate mapping may be critical in identifying the substrate necessary for the maintenance of atrial fibrillation (37). In identifying considerable structural and conduction heterogeneity in the atrium we have provided evidence in support of this concept.
The mean age of our study population was 60 ± 13 years, with a low prevalence of comorbidities and a median CHA2DS2-VASc score of 1 (IQR: 0 to 2). A total of 48% patients had histories of cardiac surgery. We cannot exclude that an older population with higher CHA2DS2-VASc scores may have different circuits and characteristics. The TCL was longer in patients with prior surgery, and overall, there was an inverse relationship between median bipolar voltage and TCL, suggesting that our study population may have had more advanced atrial remodeling. Measurements of bipolar voltage are affected by degree of contact of the mapping catheter with the endocardium and also the directionality of the bipolar signal. Although the IntellaMap-Orion mapping catheter does not have force-sensing capabilities, as is the case with other high-density mapping catheters, the presence of 64 flat microelectrodes with closely spaced bipoles arranged in a basket configuration together with the acquisition of thousands of points within 2 mm of the geometry shell leading to the creation of high-density electroanatomic maps is likely to overcome this limitation of noncontact force-sensing catheters. In addition, in areas of low voltage, electrograms were manually reviewed to assess signal quality and confirm the presence of sharp, discrete electrograms rather than far-field signals indicating poor contact.
This study provides new insights into the mechanisms underlying right atrial macro–re-entrant tachycardia. Building on extensive prior work, we have more clearly defined the regional variations in conduction and wave front propagation and the critical substrate correlates. Using a new high-resolution mapping technology, we have been able to clearly define the anatomic location and characteristics of the posterior line of block and the relevance of other anatomic sites to conduction characteristics. These observations may have relevance to understanding the mechanism of more complex arrhythmias.
COMPETENCY IN MEDICAL KNOWLEDGE 1: The posterior line of block in CTI-dependent macro–re-entrant atrial tachycardia was most commonly located in the sinus venosus region of the posterior right atrium running linearly from the SVC to the IVC. The crista terminalis was less frequently the anatomic site of posterior conduction block.
COMPETENCY IN MEDICAL KNOWLEDGE 2: The combination of anatomic regions of conduction slowing and block together with localized zones of low voltage and slow conduction appear to represent the substrate necessary for arrhythmia maintenance in right atrial macro–re-entrant tachycardia.
TRANSLATIONAL OUTLOOK 1: The study provides new insights into the nature of right atrial macro–re-entrant circuits. These observations may have relevance to understanding the mechanism of more complex arrhythmias.
TRANSLATIONAL OUTLOOK 2: In more than 70% of cases of CTI-dependent macro–re-entrant atrial tachycardia, a zone of conduction slowing or block was observed at the limbus of the fossa ovalis. To our knowledge, conduction slowing at the limbus has not previously been described, but its relevance to the arrhythmia mechanism remains to be demonstrated.
Dr. Pathik is a recipient of the Postgraduate Research Scholarship from the National Health and Medical Research Council of Australia (NHMRC) and the National Heart Foundation of Australia. Dr. Lee is supported by an Early Career Fellowship from the NHMRC. Profs. Kistler and Kalman are both supported by a Practitioner Fellowship from the NHMRC. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
All authors attest they are in compliance with human studies committees and animal welfare regulations of the authors’ institutions and Food and Drug Administration guidelines, including patient consent where appropriate. For more information, visit the JACC: Clinical Electrophysiology author instructions page.
- Abbreviations and Acronyms
- coronary sinus
- cavotricuspid isthmus
- interquartile range
- inferior vena cava
- superior vena cava
- tachycardia cycle length
- Received October 24, 2016.
- Revision received January 4, 2017.
- Accepted January 4, 2017.
- 2017 American College of Cardiology Foundation
- Jolly W.A.,
- Ritchie W.T.
- Waldo A.L.,
- MacLean W.A.,
- Karp R.B.,
- et al.
- Waldo A.L.
- Saoudi N.,
- Atallah G.,
- Kirkorian G.,
- Touboul P.
- Touboul P.,
- Saoudi N.,
- Atallah G.,
- Kirkorian G.
- Kalman J.M.,
- Olgin J.E.,
- Saxon L.A.,
- Fisher W.G.,
- Lee R.J.,
- Lesh M.D.
- Kalman J.M.,
- Olgin J.E.,
- Saxon L.A.,
- Lee R.J.,
- Scheinman M.M.,
- Lesh M.D.
- Olgin J.E.,
- Kalman J.M.,
- Fitzpatrick A.P.,
- Lesh M.D.
- Schilling R.J.,
- Peters N.S.,
- Goldberger J.,
- Kadish A.H.,
- Davies D.W.
- Shah D.C.,
- Jaïs P.,
- Haïssaguerre M.,
- et al.
- Olshansky B.,
- Okumura K.,
- Hess P.G.,
- Waldo A.L.
- Tai C.T.,
- Chen S.A.,
- Chiang C.E.,
- et al.
- Bollmann A.,
- Hilbert S.,
- John S.,
- Kosiuk J.,
- Hindricks G.
- Nakagawa H.,
- Ikeda A.,
- Sharma T.,
- Lazzara R.,
- Jackman W.M.
- Olshansky B.,
- Okumura K.,
- Henthorn R.W.,
- Waldo A.L.
- Olgin J.E.,
- Kalman J.M.,
- Saxon L.A.,
- Lee R.J.,
- Lesh M.D.
- Friedman P.A.,
- Luria D.,
- Alexis M.,
- et al.
- Tai C.T.,
- Huang J.L.,
- Lin Y.K.,
- et al.
- Joshi S.D.,
- Chawre H.K.,
- Joshi S.S.