Author + information
- Received March 4, 2005
- Revision received July 24, 2005
- Accepted August 1, 2005
- Published online December 6, 2005.
- Prashanthan Sanders, MBBS, PhD⁎,†,⁎ (, )
- Mélèze Hocini, MD⁎,†,
- Pierre Jaïs, MD⁎,†,
- Li-Fern Hsu, MBBS⁎,†,
- Yoshihide Takahashi, MD⁎,†,
- Martin Rotter, MD⁎,†,
- Christophe Scavée, MD⁎,†,
- Jean-Luc Pasquié, MD, PhD⁎,†,
- Fréderic Sacher, MD⁎,†,
- Thomas Rostock, MD⁎,†,
- Chrishan J. Nalliah, BSc⁎,†,
- Jacques Clémenty, MD⁎,† and
- Michel Haïssaguerre, MD⁎,†
- ↵⁎Reprint requests and correspondence:
Dr. Prashanthan Sanders, Hôpital Cardiologique du Haut-Lévêque, Avenue de Magellan, Bordeaux-Pessac, France.
Objectives The goal of this study was to characterize the origin of focal atrial tachycardias (AT).
Background Focal ATs originate from a small area and spread centrifugally; however, activation at the AT origin has not been characterized.
Methods Twenty patients with AT having failed prior ablation or occurring after atrial fibrillation ablation were studied. After excluding macro–re-entry, AT was mapped using a 20-pole catheter (five radiating spines; diameter 3.5 cm), performing vector mapping to identify the earliest activity followed by high-density mapping at the AT origin. Localized re-entry was considered if >85% of the tachycardia cycle length (CL) was observed within the mapping field and was confirmed by entrainment.
Results A total of 27 ATs were mapped to the pulmonary vein ostia (n = 5), and left (n = 16) and right atria (n = 6). A localized focus was evidenced at the site of origin in 19 ATs (70%), whereas in 8 (30%), localized re-entry was evidenced by 95.2 ± 4.5% of the tachycardia CL recorded within the mapping field and entrainment showed a post-pacing interval <20 ms longer than tachycardia CL (6 of 6 tested). Localized re-entry had a shorter CL (p = 0.009), slowed conduction at its origin (fractionated potential 115 ± 19 ms vs. 64 ± 22 ms, representing 49 ± 10% and 20 ± 10% of tachycardia CL, respectively; p < 0.0001), and were more often contiguous with regions of electrical silence or conduction abnormalities (88% vs. 32%; p = 0.01). In addition, mapping documented varying degrees of intra-atrial conduction block, preferential conduction (n = 5), and rapid bursts of myocardial activity (n = 1). At 11 ± 7 months, none have had recurrence of AT.
Conclusions High-density multielectrode mapping can be used to perform vector mapping to localize complex AT. It provides novel insight into the mechanisms of focal AT, distinguishing focal AT from localized re-entry.
Focal atrial tachycardias (AT) are characterized by atrial activity that originates from a small area (focus) and spreads out centrifugally (1). The mechanisms of these arrhythmias have been the subject of debate, with increased automaticity, triggered activity, and micro–re-entry all considered possibilities. Atrial tachycardias demonstrate an anatomic predilection; in the right atrium (RA), along the crista terminalis, triscuspid annulus, coronary sinus (CS) ostium, and the septum (2–5), and in the left atrium (LA), from the septum, mitral annulus, and pulmonary vein (PV) ostia (6–8). Characterization of the electrocardiographic and electrophysiologic features of tachycardia from these sites has facilitated their mapping and ablation (2,4–9).
Nevertheless, in 5% to 24% mapping is prolonged and can result in failure of ablation, particularly in patients with previous atrial surgery or ablation, or when tachycardia arises from a less well characterized region (10). Despite the frequent use of multielectrode catheters and three-dimensional mapping systems, limited information is available regarding the localized activation at the tachycardia origin. The aim of the present study was to evaluate the role of localized high-density mapping to identify and characterize the origin of focal AT.
The study comprised 20 patients undergoing ablation of AT. Patients were selected on the basis of having failed prior attempts at AT ablation (n = 8) or presenting with regular arrhythmia that were not macrore-entrant after ablation for atrial fibrillation (AF) (n = 12). Baseline characteristics of these patients are presented in Table 1.All patients gave written informed consent to the study, which was approved by the institutional clinical research and ethics committee.
Electrophysiological study was performed with the minimal use of sedation. All antiarrhythmic drugs, with the exception of amiodarone, were ceased ≥5 half-lives before the procedure.
Surface electrocardiogram and bipolar endocardial electrograms were continuously monitored and stored on a computer-based digital amplifier/recorder system (Bard Electrophysiology, Lowell, Massachusetts). Intracardiac electrograms were filtered from 30 to 500 Hz and measured at a sweep speed of 100 to 200 mm/s.
In the absence of spontaneous tachycardia, induction was attempted by burst pacing and programmed extrastimuli. If these were unsuccessful, isoproterenol was infused (1 to 6 μg/min). Standard electrophysiologic criteria were used to diagnose AT (11). Initial mapping was performed to exclude patients with a macrore-entrant mechanism. This was performed by positioning a quadripolar catheter (Xtrem, Ela Medical, Montrouge, France) in the CS and using a roving 4-mm tip quadripolar ablation catheter (Biosense-Webster, Diamond Bar, California/Bard Electrophysiology) to perform activation and entrainment mapping. If mapping failed to demonstrate macro–re-entry within the RA, mapping of the LA was performed. After LA access, heparin (0.5 mg/kg) was administered. Macro–re-entry was excluded by the following criteria: 1) sequential activation mapping demonstrating <50% of the tachycardia cycle length (CL); and 2) a post-pacing interval at ≥3 sites/atria being ≥40 ms longer than the tachycardia CL.
High-density mapping catheter
After excluding macro–re-entry, mapping was performed utilizing a new high-density mapping catheter (PentaRay, Biosense-Webster). This 7-F steerable catheter (180° of unidirectional flexion) has 20 electrodes distributed over five soft radiating spines (1-mm electrodes separated by 4-4-4 or 2-6-2 mm interelectrode spacing) allowing splaying of the catheter to cover a surface diameter of 3.5 cm. The spines have been nominated alphabetical nomenclature (A to E) with spines A and B being recognized by radio-opaque markers (Fig. 1A).
Vector mapping to localize the origin of tachycardia
Localization of the atrial focus was performed during tachycardia or atrial ectopy. Guided by the electrocardiographic appearance and previous mapping information, the catheter was sequentially applied to the endocardial surface in various atrial regions to allow rapid activation mapping. Mapping was performed to identify the earliest endocardial activity relative to the P-wave and/or to a fixed catheter positioned within the CS. By identifying the earliest site of activation around the circumference of the high-density catheter, vector mapping was performed, moving the catheter and applying it to the endocardium in the direction of earliest activation (outer bipoles) to identify the tachycardia origin and bracket activation, that is, demonstrating later activation in all surrounding regions (Fig. 1B).
Atrial tachycardia was considered to be localized re-entry if the mapping field demonstrated >85% of the tachycardia CL at the site of earliest activation, and was confirmed by entrainment.
The accuracy of mapping was confirmed by the termination or change in tachycardia with ablation. A change in tachycardia was defined by a change in the P-wave morphology with an alteration of the endocardial activation. Ablation was performed at the site of earliest endocardial activation relative to the P-wave when tachycardia was bracketed and at the most fractionated site when the entire tachycardia CL was identified. Ablation used continuous temperature feedback control of power output to achieve a target temperature of 50°C to 60°C, for a delivered power of 30 to 50 W. The acute procedural success was defined by the absence of tachycardia or ectopy in the 30 min after ablation despite infusion of isoproterenol and burst atrial pacing.
Continuous variables are reported as mean ± SD or median and range. Comparison between groups was performed with the Student ttest or when data was not normally distributed (Shapiro-Wilk test), the Wilcoxon rank sum test. Categorical variables are reported as number and percentage, and compared using the Fisher exact test. Statistical significance was established at p < 0.05.
Table 1shows the baseline characteristics according to the presence of prior AF ablation. Patients with AT and no AF were younger (p = 0.02; Student ttest) and had smaller atrial chambers (p = 0.05; Student ttest). They had AT for significantly longer durations (p = 0.0005; Wilcoxon rank sum test). All patients with AF had undergone PV electrical isolation and additional linear ablation as previously described (12–14). These patients presented with AT 3.1 ± 2.1 months after AF ablation.
Localization of AT
Twenty-seven different ATs were mapped and ablated during repetitive bursts or sustained AT (median CL 262 ms, range 190 to 550 ms, 95% confidence interval [CI] 246 to 425 ms). In 18 patients a single tachycardia was ablated and in 2 patients, 1 without AF and 1 after AF ablation; AT originated from six and three separate sites, respectively. Figure 2demonstrates a schematic representation of the AT origin. In patients with AT without AF, the tachycardia origin was localized to the PV ostia (n = 3), anterior LA (n = 4), posterior or roof of the LA (n = 2), LA appendage (n = 1), RA appendage (n = 2), and high-lateral RA (n = 1). One focus located along the anterior LA and one in the high-lateral RA were contiguous with regions of spontaneous electrical silence. In patients with AT after AF ablation, the tachycardia origin was localized to the PV ostia (n = 2), anterior or septal LA (n = 5), posterior LA or roof (n = 3), inferior LA (n = 1), posterior RA or crista terminalis (n = 2), and anterior RA (n = 1); 9 of 14 ATs occurred in close proximity to previous ablations and a further 2 arose from regions demonstrating marked conduction abnormality with spontaneous electrical silence (scar) or double potentials along the posterior RA and the anterior LA, respectively.
All ATs were successfully ablated with 7.6 ± 4.8 min (duration to eliminate and consolidate) of radiofrequency energy application per site. This was performed with fluoroscopic and procedural durations of 41.2 ± 19.4 min and 135.9 ± 36.2 min, respectively. No complications were observed with the use of the multispine catheter.
In two patients with AT after AF ablation, AT recurred one and two days after ablation, respectively, and re-ablation was performed successfully at the same sites as the initial ablation. At 11 ± 7 months after the last procedure, one patient has required further ablation for AF, but there have been no further recurrences of AT.
Focal AT versus localized re-entry
In all 27 ATs mapped in this study, atrial activation was observed to be centrifugal arising from a distinct region of the atria. In 19 (70%, 95% CI 50% to 86%), the origin of the AT demonstrated radial activation from a focal site (Fig. 1B), including 5 with activation commencing from the center of the mapping field (Fig. 3).Such focal AT demonstrated earliest activation 48.5 ± 24.7 ms (range 16 to 98 ms, 95% CI 37.4 to 59.6 ms) ahead of the P-wave, and the mapping catheter recorded the first 94.0 ± 37.0 ms of tachycardia (range 40 to 170 ms, 95% CI 76.8 to 111.2 ms) representing 26.1 ± 16.1% of the tachycardia CL (range 7.7% to 69.4%). Ablation at this site resulted in AT termination in all cases.
In the remaining eight ATs (30%, 95% CI 14% to 50%), mapping at the site of origin demonstrated 95.2 ± 4.5% (95% CI 92.1% to 98.4%) of the tachycardia CL (Figs. 4 to 6).⇓⇓⇓Four of these patients required programmed extrastimuli to induce arrhythmia, while in the remaining, AT occurred spontaneously. In six, entrainment could be performed from one to three sites within this region and demonstrated a post-pacing interval that was <20 ms longer than tachycardia CL (Fig. 5). In the remaining two, entrainment could not be assessed due to inability to capture the atrium at this site. Seven of these ATs were observed in patients with a history of AF; four in proximity to previously ablated regions (Figs. 2 and 4) and three occurring along the anterior or septal LA, an area not previously targeted for ablation (Figs. 2 and 5). In all these patients, the site of origin demonstrated prolonged fractionated electrograms (Figs. 4 to 6). Such local re-entrant activation, displaying the entire tachycardia CL, was observed in only one patient in the absence of a history of AF. This 45-year-old man had no known structural heart disease or systemic ailments; however, during previous mapping (before ablation), he was observed to have a large area of spontaneous low voltage and electrical silence along the lateral RA (Fig. 6A). For the current study, the region of electrical silence could readily be identified with activation mapping localizing the AT to an area cranial and contiguous with this region (Figs. 6B and 6C). While the entire tachycardia CL was recorded locally within the field of mapping (with centrifugal activation to the remaining atria), entrainment could not be performed due to the very low voltage amplitude (0.03 to 0.07 mV). In patients who demonstrated a localized re-entrant arrhythmia, ablation was performed at the site with the longest fractionated signal determined by the mapping catheter, which had an electrogram duration of 114.5 ± 18.6 ms (range 88 to 150 ms, 95% CI 102 to 127 ms) representing 48.9 ± 10.2% of the tachycardia CL. Ablation at this site terminated tachycardia in all.
Table 2shows the distinctive features of localized re-entry compared to focal AT. Compared to patients with focal AT, patients with localized re-entry had a shorter tachycardia CL (p = 0.009; Wilcoxon rank sum test), evidence of slowed conduction at the origin of tachycardia (p < 0.0001; Student ttest), tachycardia arose more often contiguous with regions of electrical silence or double potentials (p = 0.01; Fisher exact test), and had a history of AF (p = 0.03; Fisher exact test).
Complex intra-atrial conduction
Intra-atrial conduction block at the site of origin of AT was observed in five patients, one without previous AF ablation. The degree of intra-atrial conduction block observed was variable ranging from occasional concealed beats to regions demonstrating dissociated rhythms (Figs. 7Aand 7B). In the same patient as in Figure 7A, during mapping of the AT, an area of electrical silence was observed in the low-lateral RA. This region, despite being surrounded by areas of electrical silence, contained a small pocket of atrial myocardium (0.5 × 0.5 cm) demonstrating high-frequency activity (mean CL 83 ms) recorded on a single 5-mm bipole (Fig. 7B). Ablation was performed at this site and locally eliminated all electrical activity but had no effect on the AT, suggesting that this island of tissue may have been sequestrated from the remaining atria.
In addition, at the site of origin of the AT, preferential conduction was observed from origin-to-exit or breakout to the remaining atria in one patient (Fig. 8).In three further cases, earliest atrial activation preceded the P-wave by ≥80 ms with a gap in the locally mapped activation and a significant delay to the surrounding atria, presumably due to the presence of a preferential isthmus of conduction.
This study presents new information regarding the feasibility of high-density contact mapping during clinical electrophysiological procedures and the mechanisms of “focal” AT.
First, it demonstrates the clinical utility of multielectrode localized mapping to perform vector mapping to rapidly identify the source of AT, particularly those localized to the LA and occurring after prior ineffective ablation.
Second, high-density mapping at the site of origin of tachycardia provided novel insights into the mechanisms of these apparently “focal” tachycardias that activated the atria in a centrifugal manner; in 30% the entire CL was identified within the localized mapping field with entrainment confirming a re-entrant mechanism.
Finally, evidence of tiny regions of high-frequency atrial activity, and complex local and intra-atrial conduction block were observed to manifest in misleading electrocardiographic features.
Prior studies mapping AT
Atrial tachycardias demonstrate a characteristic anatomic predilection to endocardial structures in both the RA and LA (2–8). Knowledge of the endocardial anatomy, in concert with the specific features and activation patterns, have facilitated their mapping and ablation. While the association with atrial anatomy is observed in most cases, the mapping of AT arising from other sites is more difficult and is compounded by limited experience in mapping these less frequent arrhythmias. In addition, the incidence of such arrhythmias has burgeoned as techniques for the ablation of AF are applied to greater numbers of patients, further augmenting the need for mapping techniques for such sources that occur frequently within the LA. Although LA tachycardias may be recognized by the characteristic electrocardiogram or intracardiac activation sequence (9,15,16), subsequent localization of their origin requires detailed mapping, and reports suggest a greater prevalence of procedural failure (17).
Regular atrial arrhythmias have been reported variably after surgical or catheter ablation procedures for AF. Although patients undergoing atrial substrate ablation are more likely to develop these arrhythmias (13,14,18), they have also been described in the setting of PV ablation alone (19–21). The majority of these arrhythmias have a macrore-entrant mechanism, utilizing regions of incomplete or recovered ablation lesions to sustain arrhythmia. A smaller proportion of patients have a focal mechanism for their arrhythmias (18,21).
The mapping catheter used in the current study provided a localized density of mapping within the LA that has previously not been available. It allowed rapid activation mapping of the atria to facilitate identifying the tachycardia focus that was often localized to regions of conduction abnormalities. Whether these conduction abnormalities were pre-existing or a consequence of prior ablation is not known; however, in patients after AF ablation, AT originated from sites both in proximity to and in regions not previously targeted by ablation. Interestingly, while current classification of these arrhythmias suggests that they were “focal,” demonstrating centrifugal activation of the atria, we observed that some had a re-entrant mechanism.
Focal versus localized re-entry as the mechanism of AT
Current consensus for the classification of AT suggest that a focal AT demonstrates atrial activation commencing from a localized region and spreading centrifugally (11). In contrast, macro–re-entry is defined by activation that can be recorded over the entire tachycardia CL around a “large” central obstacle, which is generally several centimeters in diameter. These latter arrhythmias exhibit less CL variation, and entrainment maneuvers performed at sites separated by at least 2 cm are consistent with a re-entrant mechanism (11).
In the current cohort of patients, all of whom displayed centrifugal activation from a localized region, 30% demonstrated the entire tachycardia CL within the localized region at the AT origin. Half of these patients had arrhythmia induced by programmed extrastimuli, and a re-entrant mechanism was confirmed by entrainment. The local electrogram at the site of origin displayed prolonged fractionated signals, which were contiguous with regions of electrical silence. While this probably constitutes what has been referred to as micro–re-entry, these observations herald the need for further classifying these localized but re-entrant arrhythmias, which are likely to be increasingly recognized with improving mapping technologies.
High-density mapping technologies for AT
An expanding array of mapping technologies have been developed to provide a greater density of mapping of AT and to facilitate their three-dimensional localization for ablation. Electroanatomic mapping may identify critical isthmuses and structural abnormalities that provide the nidus for AT (22,23), but is limited by the requirement for relative temporal stability of the arrhythmia.
Noncontact mapping provides simultaneous virtual electrograms of the chamber with the advantage of being able to map nonsustained focal AT. It has allowed the recognition of regions of preferential conduction from origin-to-exitor breakout pointto the remaining atria, as was observed in the current series using contact electrograms (24). However, while providing a global impression of activation, virtual unipolar electrograms are susceptible to errors towards the extremities of the mapping field, and the central array can prove an impediment to navigation of a mapping catheter.
Basket catheter mapping has the ability to perform rapid simultaneous contact mapping of the chamber for the mapping of focal AT (25). While it provides a global density of mapping, its localized resolution is limited. In contrast, the catheter utilized in the current study allows splaying of the spines against the endocardial surface to achieve high-density contact mapping to accurately localize and characterize the origin of focal ATs.
The ability to position multiple electrodes in a circumscribed part of the endocardium is particularly advantageous for use within the LA and for mapping complex focal AT. In this setting, the catheter facilitated the rapid identification and ablation of focal AT without complications. It is likely that a similar utility would be observed in patients undergoing de novo ablation of AT and potentially for atrial sources of activity that may initiate and maintain AF.
Finally, the differentiation of focal from re-entrant mechanisms and documenting their electrocardiographic and electrophysiologic features may facilitate a better understanding of these conditions and allow evaluation of their electropharmacologic properties.
While this study performed detailed high-density mapping to localize and characterize the origin of AT, it was not designed to reproduce the electropharmacologic properties of these arrhythmias that have been explored previously (1).
High-density localized contact mapping facilitates the rapid identification of the origin of complex AT. Characterization of their origin demonstrates that some focal tachycardias can result from localized atrial re-entry.
PentaRay catheters were developed and provided by Biosense-Webster. Drs. Sanders, Jaïs, Hsu, and Haïssaguerre report having served on the advisory board of and having received lecture fees from Biosense-Webster. Dr. Sanders is supported by the Neil Hamilton Fairley Fellowship from the National Health and Medical Research Council of Australia and the Ralph Reader Fellowship from the National Heart Foundation of Australia. Dr. Rotter is supported by the Swiss National Foundation for Scientific Research, Bern, Switzerland. Dr. Rostock is supported by the German Cardiac Society. Mr. Nalliah is supported by the National Heart Foundation of Australia.
Presented in part at the Heart Rhythm Society’s 25th Annual Scientific Sessions, San Francisco, California, in 2004, and published in abstract form (Heart Rhythm 2004;1:S19).
- Abbreviations and Acronyms
- atrial fibrillation
- atrial tachycardia
- confidence interval
- cycle length
- coronary sinus
- left atrial/atrium
- pulmonary vein
- right atrial/atrium
- Received March 4, 2005.
- Revision received July 24, 2005.
- Accepted August 1, 2005.
- American College of Cardiology Foundation
- Chen S.A.,
- Chiang C.E.,
- Yang C.J.,
- et al.
- Kalman J.M.,
- Olgin J.E.,
- Karch M.R.,
- Hamdan M.,
- Lee R.J.,
- Lesh M.D.
- Marrouche N.F.,
- SippensGroenewegen A.,
- Yang Y.,
- Dibs S.,
- Scheinman M.M.
- Kistler P.M.,
- Sanders P.,
- Fynn S.P.,
- et al.
- Kistler P.M.,
- Sanders P.,
- Hussin A.,
- et al.
- Tang C.W.,
- Scheinman M.M.,
- Van Hare G.F.,
- et al.
- Saoudi N.,
- Cosio F.,
- Waldo A.,
- et al.
- Jais P.,
- Hocini M.,
- Hsu L.F.,
- et al.
- Soejima K.,
- Stevenson W.G.,
- Delacretaz E.,
- Brunckhorst C.B.,
- Maisel W.H.,
- Friedman P.L.
- Bochoeyer A.,
- Yang Y.,
- Cheng J.,
- et al.
- Thomas S.P.,
- Nunn G.R.,
- Nicholson I.A.,
- et al.
- Gerstenfeld E.P.,
- Callans D.J.,
- Dixit S.,
- et al.
- Nakagawa H.,
- Shah N.,
- Matsudaira K.,
- et al.
- Jais P.,
- Shah D.C.,
- Haissaguerre M.,
- et al.
- Higa S.,
- Tai C.T.,
- Lin Y.J.,
- et al.
- Schmitt C.,
- Zrenner B.,
- Schneider M.,
- et al.