Author + information
- Received October 6, 2000
- Revision received March 2, 2001
- Accepted April 12, 2001
- Published online August 1, 2001.
- ↵*Reprint requests and correspondence: Dr. Bernhard Frey, Klinik fur Innere Medizin II, Abteilung fur Kardiologie, Währinger Gürtel 18-20, A-1090 Wien, Austria
The purpose of the study was to examine the value of right- and left-sided mapping to identify the site of tachycardia origin.
Focal atrial tachycardia may originate from the vicinity of the atrioventricular node from either side of the interatrial septum.
In 16 patients undergoing radiofrequency catheter ablation of perinodal atrial tachycardia, activation mapping of the right and left side of the interatrial septum was performed.
Atrial tachycardia originated from the right side of the interatrial septum in 10 patients (group A) and from the left side in 6 patients (group B). On the right side, earliest atrial activity preceded the onset of the P-wave by 49 ± 15 ms in group A and by 38 ± 8 ms in group B (NS), and it preceded the signal recorded from the right atrial appendage by 59 ± 19 ms in group A and by 60 ± 13 ms in group B (NS). On the left side, earliest activity preceded the onset of the P-wave by 27 ± 16 ms in group A and by 51 ± 6 ms in group B (<0.01), and it preceded the signal obtained from the right atrial appendage by 38 ± 19 ms in group A and by 73 ± 9 ms in group B (<0.01). Atrial tachycardias were successfully eliminated in all patients without impairment of atrioventricular conduction. During follow-up, two patients had a recurrence of tachycardia.
Mapping of only the right side cannot exclude a left-sided origin. Therefore, mapping of both sides of the interatrial septum is required prior to ablation of focal atrial tachycardia originating from the vicinity of the atrioventricular node.
Focal atrial tachycardias can originate from any site within the right or left atrium, but they generally tend to cluster in certain anatomical areas, such as the crista terminalis or the pulmonary veins (1–6).
Recently, the apex of the triangle of Koch has been identified as a particular site of origin for focal atrial tachycardias (7,8). A substantial proportion of these tachycardias originating from the vicinity of the atrioventricular (AV) node may actually arise from the left side of the interatrial septum. Selection of the appropriate ablation site in that area is mandatory to decrease the risk of inadvertent damage to the AV node during ablation (9,10). In the present study, a comparison of right- and left-sided activation mapping of the interatrial septum was carried out in patients with focal atrial tachycardia originating from the vicinity of the AV node to delineate more precisely the optimal ablation site.
The study population consisted of 16 consecutive patients (9 women, 7 men, mean age 53 ± 20 years) with focal atrial tachycardia in whom mapping of the right atrium revealed earliest activation at the apex of the triangle of Koch. Atrial tachycardia was paroxysmal in 15 patients and incessant in 1 patient. The median duration of symptoms was six years (range 6 months to 20 years). Initial evaluation included medical history, physical examination, a standard electrocardiogram (ECG) and a 24-hour Holter ECG. Narrow QRS-complex tachycardia was documented in 15 patients, and 1 patient presented with wide QRS-complex tachycardia. Three patients also had other tachycardias: paroxysmal atrial fibrillation in two patients, and one patient who had previously undergone slow pathway ablation for control of AV nodal reentrant tachycardia. The patient characteristics are summarized in Table 1. Coronary artery disease was present in two patients, arterial hypertension was present in five patients, and one patient had left to right shunt due to an unroofed coronary sinus with a pulmonary-to-systemic flow ratio of 2.2. In eight patients, there was no evidence of organic heart disease. On echocardiography, right atrial dilation was found in six patients, left atrial dilation was found in four patients, and one patient had an aneurysm of the interatrial septum. In the remaining five patients, no structural atrial abnormality was found.
Surface 12-lead ECG recordings of atrial tachycardia were obtained in all patients during the electrophysiologic study at a recording speed of 50 and 100 mm/s. The tracings were selected for evaluation according to P-wave visibility. This was particularly the case during delivery of premature ventricular beats, which did not reset tachycardia, or during abrupt changes in anterograde AV nodal conduction. The ECGs were evaluated by consensus by three observers. The P-wave duration was measured in the inferior leads; P-waves were classified as either positive, negative, biphasic or isoelectric.
Electrophysiologic study protocol
Written, informed consent was obtained from all patients. Antiarrhythmic drug medication was discontinued at least five half-lives before the study. Electrophysiologic study was performed with the patients in the fasting state. Midazolam and fentanyl were used for sedation and analgesia.
Multielectrode catheters were inserted percutaneously into the femoral veins and were positioned at the right atrial appendage, the inferior aspect of the tricuspid annulus and the right ventricular outflow tract, and adjacent to the AV junction, to record the His-bundle deflection. Electrograms were stored using a digital recording system (Bard Electrophysiology, Tewksbury, Massachusetts). Both AV conduction and refractoriness in the anterograde and retrograde direction were evaluated by incremental atrial and ventricular pacing and atrial and ventricular extrastimulus technique. Atrial single and double premature extrastimuli as well as incremental pacing and burst pacing were used to induce tachycardia. In addition, Orciprenaline (0.2 μg/kg/min) was given in three patients to facilitate induction of tachycardia.
Accessory pathways, AV nodal reentry, sinoatrial reentrant tachycardia and atrial flutter (cycle lengths <280 ms) were excluded as diagnostic possibilities. The following criteria were used to delineate the tachycardia mechanism and confirm the diagnosis of atrial tachycardia: atrial activation sequence during tachycardia different from that during sinus rhythm; atrial activation sequence during tachycardia different from retrograde atrial activation sequence during ventricular stimulation; inability to advance atrial activation by ventricular premature beats delivered during tachycardia at a time of His-bundle refractoriness; absent or weak ventriculoatrial (VA) conduction; tachycardia induction and maintenance independent of AV nodal conduction; presence of anterograde AV block during tachycardia; and a change in anterograde AV nodal conduction during tachycardia from the fast to the slow AV nodal pathway without a corresponding change in tachycardia cycle length.
Radiofrequency catheter ablation
A quadripolar 4-mm tip catheter (Cordis Webster, Baldwin Park, California) was used for mapping and delivery of radiofrequency energy. During atrial tachycardia, endocardial activation mapping was performed in reference to the signal recorded from the right atrial appendage. The right side of the interatrial septum was mapped first, and the site of earliest local atrial activation was identified. Subsequently, the left side of the interatrial septum was mapped using a trans-septal approach (trans-septal 8F sheath, Daig, Minnetonka, Minnesota) in 15 patients or via an atrial septal defect in 1 patient. The site of the earliest local atrial activation was targeted for ablation. Radiofrequency energy (Radionics, Burlington, Massachusetts) was delivered between the tip electrode and a cutaneous patch electrode positioned under the left scapula. Radiofrequency energy was applied either during tachycardia or during sinus rhythm. The AV conduction was carefully monitored during energy application. Energy was initially set at 10 W and was increased slowly in a titrated manner to a maximum of 30 W (7,8). Application of energy was ceased upon occurrence of junctional rhythm. At these sites, energy was delivered only after demonstration of unchanged AV nodal conduction properties and unchanged inducibility of tachycardia. Radiographs of the catheter position at the successful ablation sites were obtained. The end point of ablation was the abolition of spontaneous tachycardia and/or the inability to induce tachycardia with programmed stimulation or burst pacing or with Orciprenaline infusion.
Patients were monitored for 24 to 48 h after ablation. Patients in whom radiofrequency energy had been applied to the left side of the interatrial septum were treated with intravenous heparin for 48 h, keeping the activated partial thromboplastin time at two to three times the basal value. Aspirin, 500 mg/day, was given for six weeks. Patients were followed clinically, including the recording of a standard 12-lead ECG.
All data are reported as mean ± 1 SD or as median (range) as appropriate. The Student ttest was used for group comparison of continuous variables.
On arrival of patients in the electrophysiological laboratory, tachycardia occurred spontaneously in two patients. Tachycardia could be initiated by programmed atrial stimulation or burst pacing in 11 patients, and with Orciprenaline infusion in the remaining three patients. Tachycardia had a mean cycle length of 440 ± 56 ms and showed a marked cycle length variability.
The diagnosis of atrial tachycardia was confirmed according to the criteria outlined in the Methods section. In detail, VA conduction was absent or weak in five patients. Atrial activation sequence during tachycardia differed from retrograde atrial activation sequence during incremental ventricular stimulation in all 11 patients with VA conduction during incremental ventricular pacing. In seven patients with VA conduction and sustained atrial tachycardia, the atrial activation could not be advanced by ventricular premature beats delivered during tachycardia at a time of His-bundle refractoriness. Both tachycardia induction and maintenance were independent of AV nodal conduction in all 16 patients, and anterograde AV block during tachycardia was present in 1 patient. A change in anterograde AV nodal conduction during tachycardia from the fast to the slow AV nodal pathway without a corresponding change in tachycardia cycle length was observed in three patients.
In the 11 patients with VA conduction during incremental ventricular pacing, subtle differences were seen in atrial activation between atrial tachycardia and retrograde AV nodal conduction (Fig. 1). The time from local atrial activation recorded by the His catheter to activation of the right atrial appendage differed by 6 ± 4 ms. Furthermore, the time to the atrial activation in the region of the right inferoseptal aspect of the tricuspid annulus differed by 12 ± 10 ms.
Anterograde dual AV nodal pathway physiology was found in 11 patients. Six patients demonstrated retrograde dual AV nodal pathway physiology. However, AV nodal reentrant tachycardia was not inducible in any patient. In two patients, nonsustained atrial tachycardia with a different atrial activation sequence occurred during Orciprenaline provocation or was induced by atrial burst pacing, respectively. Neither tachycardia was targeted for ablation.
Right- and left-sided mapping of the interatrial septum
Mapping of the right atrium localized the earliest activation next to the AV node in all patients. Atrial tachycardia originated from the right side of the interatrial septum in 10 patients (group A) and from the left side in 6 others (group B). The time spans from earliest right atrial activity to the onset of P-wave and to the signal recorded from the right atrial appendage were quite similar in both patient groups and did not allow us to distinguish between right- and left-sided tachycardias (Table 2). Subsequent mapping of the left aspect of the anterior interatrial septum revealed earliest activation at a site that corresponded with the site of earliest activation on the right side on fluoroscopy. On the left side, timing of the earliest atrial activation differed significantly between the two groups (Table 2). For both aspects of the septum, the time from the earliest atrial activity to activation of the right atrial appendage was calculated. The absolute value of the difference between these two times (i.e., the time between the earliest atrial activation on the right and left side) was 22 ± 10 ms in group A and 13 ± 11 ms in group B (NS).
The P-wave configuration
On the 12-lead ECG, P-wave duration during sinus rhythm was 113 ± 31 ms. The P-wave duration during tachycardia was shorter (89 ± 27 ms, p < 0.05) and did not differ between group A (96 ± 14 ms) and group B (80 ± 38 ms, NS). The P-wave morphology is shown in Table 3. Additionally, P-waves had an inferior axis in five patients, an intermediate axis in five others and a superior axis in six patients. A negative P-wave in aVL (augmented limb lead) was found in three patients in group B. In lead V1, a monophasic positive P-wave was found in all six patients in group B, whereas in group A, the P-wave was either isoelectric or biphasic in all patients (Fig. 2).
Radiofrequency catheter ablation
Radiofrequency current was delivered only at the site of earliest atrial activation after mapping of both aspects of the septum. The AV ratio of the local electrograms at the successful ablation sites was 1.4 ± 1.2 in group A and 2.9 ± 2.0 in group B (NS). In group A, elimination of atrial tachycardia required 10 ± 6 applications of radiofrequency energy for 48 ± 14 s with 18 ± 7 W (Fig. 3). In group B, elimination of atrial tachycardia required 3 ± 3 applications of radiofrequency energy for 73 ± 29 s with 24 ± 7 W (Fig. 3). A His-bundle potential was recorded at the successful ablation site in 6 of 10 patients in group A, whereas a His-bundle potential was not recorded at any successful ablation site in group B patients. Furthermore, junctional rhythm during energy application occurred in 6 of 10 patients in group A but was not observed in any patient of group B.
After successful ablation, anterograde AV nodal conduction was not impaired in any patient. Dual AV nodal pathway physiology remained unchanged in 11 of 16 patients. The AH interval during sinus rhythm remained unchanged (76 ± 22 ms preablation vs. 77 ± 21 ms postablation, NS). The anterograde Wenckebach cycle length of the AV node decreased from 360 ± 118 ms prior to ablation to 319 ± 94 ms postablation (p < 0.05). Elimination of tachycardia originating from the right side of the septum was accompanied by abolition of retrograde fast AV nodal conduction in one patient. A distinct His potential was recorded at the successful ablation site. In the remaining nine patients, retrograde AV nodal conduction remained undamaged. In patients with VA conduction, the retrograde Wenckebach cycle length of the AV node decreased from 469 ± 330 ms prior to ablation to a postablation value of 314 ± 79 ms (p < 0.05).
Median duration of follow-up was eight months (range 1 to 68 months). Atrial tachycardia recurred in one patient of group A and in one patient of group B within three and seven months, respectively. In both patients, tachycardia could be treated effectively with beta-blockers. The remaining 14 patients had no recurrence of tachycardia. Late AV nodal conduction disturbances were not observed in any patient.
The present study comprised patients with focal atrial tachycardia in whom right atrial mapping revealed earliest right atrial activation in the vicinity of the AV node. Because selection of the appropriate ablation site in that area is mandatory to decrease the risk of inadvertent damage to the AV conduction system during ablation (7–10), a differentiation between a right-sided and a left-sided tachycardia origin is required. Earliest right atrial activation occurred at the apex of the triangle of Koch in all patients regardless of a right or left site of tachycardia origin.
Right and left atrial activation mapping
Right atrial activation mapping in reference to a right atrial signal could not differentiate between a right-sided or a left-sided tachycardia. This implies that left-sided tachycardias exhibit a direct trans-septal impulse conduction leading to this perinodal right-sided breakthrough. The interatrial septum in the region of the AV node provides for a direct interatrial electrical connection because it is composed mainly of muscular fibers (11,12). In humans, trans-septal conduction in this area has not been investigated; however, transseptal conduction in this region has been demonstrated in dogs, using simultaneous multisite mapping (13).
Right atrial activation mapping in reference to the onset of P-wave did not turn out to be helpful either. There was a wide range in the time from earliest right atrial activation to the onset of P-wave, with significant overlap between right-sided and left-sided tachycardias that compares well with other reports in the literature (1–8). Considering this time range in relation to the relatively short trans-septal conduction time in patients with left-sided tachycardias, it is understandable that right atrial activation mapping was not able to identify the site of tachycardia origin.
Left-sided atrial activation mapping in reference to a right atrial signal revealed significant differences between right- and left-sided tachycardias. A right atrial electrical signal constitutes an appropriate reference for left atrial activation mapping because, in that case, duration of transseptal conduction allows for discrimination between a right- and left-sided origin. Vice versa, it may be possible that a left atrial reference signal might turn out to be useful for right atrial activation mapping.
In addition, left-sided activation mapping in reference to the onset of P-wave revealed significant differences between right- and left-sided tachycardias because of somewhat longer, although not statistically significant, trans-septal conduction times of right atrial tachycardias.
Radiofrequency catheter ablation
Radiofrequency catheter ablation of atrial tachycardias in the region of the AV node carries the risk of postablation AV block (7–10). Both delivery of radiofrequency energy in a titrated manner (14)and immediate discontinuation of energy delivery upon occurrence of an accelerated junctional rhythm (15)have been advocated to minimize the risk of damage to the anterograde AV conduction. On the left side of the interatrial septum, application of energy is less likely to alter AV conduction, because the distance to the compact AV node is greater when compared to the right side (16). In the present study, the higher risk of energy delivery on the right side of the septum is illustrated by the fact that in one patient retrograde fast AV nodal pathway conduction was abolished during ablation of a right-sided tachycardia.
The value of P-wave morphology to predict the site of origin of focal atrial tachycardia was evaluated in only a few studies. Chen et al. (17)analyzed P-wave morphology in 49 patients with right atrial septal tachycardia. All tachycardias originating from the anteroseptal and midseptal region exhibited a biphasic or negative polarity in lead V1. Tang et al. (18)analyzed P-wave morphology in 31 patients with focal atrial tachycardias of any location. Of interest, lead V1was most helpful in distinguishing right atrial from left atrial foci. The sensitivity and specificity of a monophasic, positive P-wave in lead V1to predict a left atrial focus were 93% and 88%, respectively. In the present study, a monophasic positive P-wave in V1correctly predicted tachycardia origin on the left side of the interatrial septum in all patients and was not found in any patient with right-sided tachycardia origin. As previously reported (19), the duration of P-waves was shorter during atrial tachycardia originating from the septum when compared to sinus rhythm. However, P-wave duration did not allow us to distinguish a left septal from a right septal site of tachycardia origin.
Focal atrial tachycardia may originate from either side of the interatrial septum in the vicinity of the AV node. A monophasic positive P-wave in lead V1during atrial tachycardia suggests an origin located on the left side of the interatrial septum. Importantly, mapping of only the right side cannot exclude a left-sided origin. Furthermore, a right-sided ablation attempt in the vicinity of the AV node carries a high risk of AV block, particularly when targeting a left-sided tachycardia. Mapping the left side of the interatrial septum appears therefore mandatory prior to ablation. Thus, a trans-septal approach is required, which has been shown to be both safe and effective (20–22).
- Received October 6, 2000.
- Revision received March 2, 2001.
- Accepted April 12, 2001.
- American College of Cardiology
- Kay G.N,
- Chong F,
- Epstein A.E,
- et al.
- Tracy C.M,
- Swartz J.F,
- Fletcher R.D,
- et al.
- Lesh M.D,
- VanHare G.F,
- Epstein L.M,
- et al.
- Chen S.A,
- Chiang C.E,
- Yang C.J,
- et al.
- Salerno J.A,
- De Ponti R,
- Storti C,
- et al.
- Hwang C,
- Peterson L,
- Uluave P.S,
- et al.
- Wang K,
- Ho Y.H,
- Gibson D.G,
- Anderson R.H
- Sun H,
- Velipasaoglu E.O,
- Wu D.E,
- et al.
- Haissaguerre M,
- Marcus F,
- Poquet F,
- et al.
- Dean J.W,
- Ho S.Y,
- Rowland E,
- et al.
- Tang C.W,
- Scheinman M.M,
- VanHare G.F,
- et al.
- Lesh M.D,
- VanHare G.F,
- Scheinmann M.M,
- et al.
- Linker N.J,
- Fitzpatrick A.P