Circulation. 1996;94:407-424
(Circulation. 1996;94:407-424.)
© 1996 American Heart Association, Inc.
Role of the Tricuspid Annulus and the Eustachian Valve/Ridge on Atrial Flutter
Relevance to Catheter Ablation of the Septal Isthmus and a New Technique for Rapid Identification of Ablation Success
Hiroshi Nakagawa, MD, PhD;
Ralph Lazzara, MD;
Terrance Khastgir, MD;
Karen J. Beckman, MD;
James H. McClelland, MD;
Shinobu Imai, MD;
Jan V. Pitha, MD, PhD;
Anton E. Becker, MD;
Mauricio Arruda, MD;
Mario D. Gonzalez, MD;
Lawrence E. Widman, MD, PhD;
Michael Rome, MD;
Jeffrey Neuhauser, MD;
Xunzhang Wang, MD;
James D. Calame, RN;
Maurice D. Goudeau, RN;
Warren M. Jackman, MD
the Department of Medicine, University of Oklahoma Health Sciences Center, and the Department of Veterans Affairs Medical Center, Oklahoma City, Okla, and Academic Medical Center (A.E.B.), Amsterdam, Netherlands.Presented in part at the 14th Annual Scientific Sessions of the North American Society of Pacing and Electrophysiology, May 1993, San Diego, Calif.
Correspondence to Warren M. Jackman, MD, Department of Medicine/Cardiovascular Section, University of Oklahoma Health Sciences Center, 920 SL Young Blvd, Room 5SP300, Oklahoma City, OK 73104.
 |
Abstract
|
|---|
Background Typical atrial flutter (AFL) results from right atrial
reentry by propagation through an isthmus between the inferior
vena cava (IVC) and tricuspid annulus (TA). We postulated that
the eustachian valve and ridge (EVR) forms a line of conduction
block between the IVC and coronary sinus (CS) ostium and forms
a second isthmus (septal isthmus) between the TA and CS ostium.
Methods and Results Endocardial mapping in 30 patients with AFL demonstrated atrial activation around the TA in the counterclockwise direction (left anterior oblique projection). Double atrial potentials were recorded along the EVR in all patients during AFL. Pacing either side of the EVR during sinus rhythm also produced double potentials, which indicated fixed anatomic block across EVR. Entrainment pacing at the septal isthmus and multiple sites around the TA produced a 
return interval 
8 ms in 14 of 15 patients tested. Catheter ablation eliminated AFL in all patients by ablation of the septal isthmus in 26 patients and the posterior isthmus in 4. AFL recurred in 2 of 12 patients (mean follow-up, 33.9±16.3 months) in whom ablation success was defined by the inability to reinduce AFL, compared with none of 18 patients (mean follow-up, 10.3±8.3 months) in whom success required formation of a complete line of conduction block between the TA and the EVR, identified by CS pacing that produced atrial activation around the TA only in the counterclockwise direction and by pacing the posterior TA with only clockwise atrial activation.
Conclusions (1) The EVR forms a line of fixed conduction block between the IVC and the CS; (2) the EVR and the TA provide boundaries for the AFL reentrant circuit; and (3) verification of a complete line of block between the TA and the EVR is a more reliable criterion for long-term ablation success.
Key Words: atrial flutter • mapping • catheter ablation • radiofrequency
|
Introduction
|
|---|
Atrial flutter was described early in this century
1 and was
considered to be the result of a macroreentrant circuit in the
right atrium in the vicinity of the venae cavae.
2 3 In recent
decades, this hypothesis was supported by mapping of activation
in animal models of atrial flutter and typical atrial flutter
in humans with sequential catheter recordings and with simultaneous
multielectrode maps of atrial activation during surgery.
4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 The prevalent interpretation
of the activation maps has been that the reentrant circuit contains
a long segment that propagates anteriorly in the interatrial
septum and another that propagates posteriorly along the lateral
wall of the right atrium, but the specific course of the circuit
in its anterior and posterior aspects has been unclear. A line
of conduction block extending between the venae cavae has been
thought to separate the septal and free wall segments.
19 20 In recent years, it has been shown that the reentrant circuit
courses between the inferior vena cava and the tricuspid annulus
posteriorly and that this posterior isthmus is an opportune
site to interrupt the circuit with a lesion induced by radiofrequency
current.
21 22 23
We recently postulated the presence of a second line of conduction block extending between the inferior vena cava and the coronary sinus ostium that forces the reentrant impulse to propagate between the coronary sinus ostium and the tricuspid annulus and forms another, more narrow isthmus (septal isthmus) amenable to ablation.24 25 In this hypothesis, the reentrant atrial wavefront propagates around the tricuspid annulus, between the annulus and the inferior vena cava posteriorly (site A in Fig 1
), and arrives at the line of block between the coronary sinus ostium and the inferior vena cava (site B in Fig 1). The impulse travels anteriorly only through the region between the coronary sinus ostium and the tricuspid annulus (site C in Fig 1). Atrial activation then proceeds anteriorly along the tricuspid annulus (site D in Fig 1) and simultaneously pivots around the coronary sinus ostium, back toward the inferior vena cava (site E in Fig 1). Electrograms in the region of the line of block would be expected to exhibit two distinct atrial potentials separated by an isoelectric interval (double potentials).19 20 24 25 26 27 28 29 The first potential is generated by the arriving wavefront on the posterior side of the line of block (site B in Fig 1), whereas the second potential is generated by the returning atrial wavefront on the anterior side of the line of the block (site E in Fig 1). Electrodes positioned on the proximal (posterior) side of the block would exhibit a larger, sharper first potential and a smaller, rounded (distant-appearing) second potential, whereas electrodes positioned on the distal (anterior) side of the block would record a small first potential and a larger second potential.27 29
View larger version (28K):
[in this window]
[in a new window]
|
Figure 1. Schematic of the right atrium, as viewed in the right anterior oblique projection, illustrates the hypothesized reentrant circuit in typical atrial flutter (arrows) and the role of the eustachian valve and ridge in forming a line of conduction block between the inferior vena cava (IVC) and the coronary sinus ostium (CS). The eustachian valve/ridge and the tricuspid annulus (TA) form boundaries of a protected channel within the reentrant circuit, beginning with the posterior isthmus (between the TA and the IVC, site A) and ending with the septal isthmus (between the TA and the CS, site C). Dashed lines represent the anterior end of the tendon of Todaro, which has overlying right atrial myocardium. SVC indicates superior vena cava.
|
|
Entrainment pacing also can be used to verify the line of block. Entrainment pacing (at a cycle length slightly shorter than the flutter cycle length) from a site within the reentrant circuit would produce an identical P (flutter) wave and atrial activation sequence except for a small area of antidromic activation close to the pacing site.5 30 31 32 33 34 The interval from the last pacing stimulus to the return atrial potential recorded at the pacing site would be very close (within 10 ms) to the atrial flutter cycle length.30 31 32 33 34 The response to entrainment pacing at the atrial myocardium below the line of block (identified by double potentials with a larger first potential, site B in Fig 1
) would be similar to that described for a "blind alley" connecting to the reentrant circuit in ventricular tachycardia.31 32 33 34 The atrial activation sequence during pacing would be identical to the flutter activation sequence (concealed entrainment), but the interval from the last pacing stimulus to the return atrial potential would be significantly longer than the flutter cycle length. If there were no line of block, the interval between the last pacing stimulus and the return atrial potential would be approximately equal to the flutter cycle length.
The purpose of this study was to test the hypothesis that, in typical atrial flutter, a line of conduction block exists between the coronary sinus ostium and the inferior vena cava that forces the reentrant impulse through a relatively narrow "septal isthmus" between the coronary sinus ostium and the tricuspid annulus. The probable anatomic location for this line of block would be the eustachian valve and ridge (Fig 2A). We tested this hypothesis using (1) intracardiac mapping (to identify double potentials), (2) entrainment pacing (concealed entrainment with a return interval longer than atrial flutter cycle length after pacing below the eustachian valve/ridge), and (3) radiofrequency catheter ablation of the septal isthmus to determine whether this produces complete posterior-anterior conduction block for atrial wavefronts that propagate septally between the inferior vena cava and the tricuspid annulus and eliminates atrial flutter.
View larger version (128K):
[in this window]
[in a new window]
|
Figure 2. Photographs of the right atrial septum taken in the right anterior oblique projection in two autopsy hearts to illustrate the anatomic relationship of the eustachian valve (EV) and eustachian ridge (ER) to the inferior vena cava (IVC), coronary sinus ostium (CS), and thebesian valve (ThV) in A and D. Note the continuity between the EV and the ThV in D, which might produce a continuous line of conduction block, compared with the more muscular region between the ER and the CS in A, which might allow conduction between the ER and the CS. B shows the location of the orthogonal electrode catheter extending along the EV and ER, between the IVC and the CS. C illustrates the lines of ablation across the septal isthmus (SI) and the posterior isthmus (PI). Note the shorter length and smoother surface across the SI than the PI. FO indicates fossa ovalis; TA, tricuspid annulus.
|
|
|
Methods
|
|---|
Study Population
The study population consisted of 30 patients referred for catheter
ablation of atrial flutter (Table 1
). There were 19 men and
11 women, ranging in age from 20 to 75 years (mean, 50±16
years). Atrial flutter with typical flutter wave morphology
(inverted sawtooth pattern in the inferior ECG leads) was the
predominant clinical arrhythmia. Atrial flutter was chronic/incessant
in 11 patients and paroxysmal in 19 of the 30 patients (63%).
At least one episode of atrial fibrillation had been documented
in 12 of the 30 patients (40%) and episodes of atrial tachycardia
in 2 patients. A mean of 3.5±1.4 (range, 2 to 6) antiarrhythmic
drugs had failed to prevent recurrences of the atrial flutter.
Palpitations and other symptoms that had been attributed to
atrial flutter were present for a mean of 7±6.3 years.
Atrial flutter produced syncope or presyncope in 13 patients.
Structural heart disease was present in 23 of the 30 (77%) patients.
Echocardiographic evidence of left atrial enlargement (>4
cm) and/or right atrial enlargement was present in 15 of the
30 (50%) patients. Seven of the 30 (23%) patients had previously
undergone an unsuccessful attempt at catheter ablation of the
atrial flutter at another institution. One of these patients
(No. 7) underwent an atrioventricular nodal modification procedure
(using the anterior approach) after failure to eliminate the
atrial flutter.
Electrophysiological Study Protocol
Classes I and III antiarrhythmic drugs were withdrawn at least 5 days before study and aspirin (325 mg daily) was administered 1 day before the study. After providing written informed consent, each patient underwent electrophysiological study in the fasting state under heavy sedation with fentanyl (25 to 100 µg/h) and midazolam (1 to 4 mg/h). Oxygen saturation was monitored with a pulse oximeter, and expired carbon dioxide was monitored with a capnometer. Five multipolar electrode catheters (2-mm interelectrode spacing or orthogonal electrodes) were inserted percutaneously into the right subclavian vein and the right and left femoral veins. Three of the catheters were advanced to the right atrial appendage, His bundle region, and coronary sinus. A 7F deflectable catheter with 20 electrodes spaced in 2-7-2-mm intervals (Halo catheter, Cordis Webster) was positioned around the tricuspid annulus to record atrial activation close to the lateral and posterior tricuspid annulus (TA in Fig 3). The remaining catheter was used for right atrial mapping (MAP in Fig 3). One of these catheters (or an additional catheter) was positioned in the right ventricle during the ablation portion of the procedure.
In 18 of the 30 patients, the orthogonal coronary sinus catheter was advanced from the inferior vena cava to the proximal coronary sinus to obtain recordings along the eustachian valve/ridge between the inferior vena cava and the coronary sinus ostium (Fig 2B and IVC-CS in Fig 3). A 7F deflectable catheter was used, which had 8 orthogonal electrode pairs with 1.5-mm spacing between orthogonal pairs (Cordis Webster). The shaft of this catheter extends 2 cm beyond the distal orthogonal electrode to anchor the catheter in the coronary sinus. In the remaining patients, electrograms from the region of the eustachian valve/ridge were obtained with the right atrial mapping catheter.
Close bipolar intracardiac electrograms (2-mm spacing or orthogonal electrodes) were recorded from each catheter with a filter bandwidth of 30 to 500 Hz and were displayed at low gain (5 to 20 mV/cm).
In patients with sinus rhythm at the onset of the procedure, atrial flutter was induced by programmed atrial stimulation with up to three extrastimuli and burst pacing at two atrial sites (right atrial appendage and posterior or posterolateral coronary sinus). If atrial flutter was not induced or was not sustained in the baseline state, isoproterenol (0.5 to 2 µg/min) was administered by continuous infusion and programmed atrial stimulation was repeated.
During atrial flutter, mapping of the right atrium and coronary sinus was performed to identify the atrial activation sequence along the tricuspid annulus, around the coronary sinus ostium, along the region between the coronary sinus ostium and the inferior vena cava (including the eustachian valve/ridge), and in the proximal coronary sinus. Entrainment pacing was performed in 15 of the 30 patients at (1) the posteroseptal right atrium between the coronary sinus ostium and the tricuspid annulus (site C in Fig 1), (2) anterior and posterior to the line of double potentials extending along the eustachian valve/ridge between the coronary sinus ostium and the inferior vena cava (sites E and B in Fig 1), and (3) at several free wall sites around the tricuspid annulus. Entrainment pacing was performed during atrial flutter at a cycle length 15 to 25 ms shorter than the flutter cycle length. The sequence of atrial activation at all electrode recording sites during entrainment pacing was compared with the atrial activation sequence at these same sites during atrial flutter. The return interval was defined as the interval from the last pacing stimulus to the return atrial potential, which was recorded at the pacing site. When amplifier saturation prevented the recording of the return atrial potential at the pacing site, the timing of the return atrial potential was estimated from the timing of return atrial potentials recorded close to the pacing site. The return interval was defined as (return interval) minus (flutter cycle length). The return interval was used as an estimation of the distance of the entrainment pacing site from the reentrant circuit.
In 15 of the 19 patients who were in sinus rhythm before ablation, atrial pacing (at long cycle lengths) was used to determine whether the line of conduction block along the eustachian valve/ridge was present in the absence of atrial flutter (fixed anatomic block versus functional block during atrial flutter). Two deflectable electrode catheters were positioned just anterior and posterior to the eustachian valve/ridge (Fig 4). Atrial pacing (cycle length >500 ms) was performed individually from the anterior and posterior catheters, and the timing of atrial activation at the opposite catheter was compared with the timing of atrial activation at the coronary sinus ostium. Later atrial activation at the opposite catheter than at the coronary sinus ostium (or the region between the tricuspid annulus and the coronary sinus ostium) would suggest the presence of fixed anatomic conduction block at the eustachian valve/ridge with propagation of the paced atrial wavefront around the coronary sinus ostium or around the anterior portion of the eustachian ridge (Fig 5). Earlier atrial activation at the catheter opposite to that at the coronary sinus ostium would indicate the presence of conduction across the eustachian valve/ridge and suggest that the block along the eustachian valve/ridge during atrial flutter is functional.
View larger version (31K):
[in this window]
[in a new window]
|
Figure 5. Schematics of the right atrium (RA) and left atrium (LA) as viewed in the left anterior oblique projection illustrate the expected pattern of atrial activation in the presence of a line of conduction block along the eustachian valve/ridge (EVR) during atrial pacing anterior to the EVR in A and posterior to the EVR in B. Ant indicates anterior pacing site; Post, posterior pacing site; and HB, His bundle. Other abbreviations as in Fig 3.
|
|
Catheter Ablation
The primary approach for radiofrequency catheter ablation of the atrial flutter was to create a line of atrial conduction block between the tricuspid annulus and the coronary sinus ostium (Line SI in Fig 2C, Fig 6A, and Fig 7A and 7B). If the eustachian valve/ridge provides a line of block between the coronary sinus ostium and inferior vena cava, that ablation should create an arc of conduction block (tricuspid annulus–coronary sinus ostium–inferior vena cava) and eliminate the atrial flutter (Fig 6A).
View larger version (74K):
[in this window]
[in a new window]
|
Figure 7. Radiographs in the right anterior oblique projection show the ablation catheter (ABL) positions during ablation along the septal isthmus, beginning at the tricuspid annulus at the level of the posterior margin of the coronary sinus ostium (A) and extending to the posteroapical margin of the coronary sinus ostium (B), completing ablation approach A. Extension of the ablation line to the eustachian ridge (C) completes ablation approach A and B. RV indicates right ventricle; other abbreviations as in previous figures.
|
|
The ablation catheter was inserted through one of the right femoral venous sheaths, and the tip electrode was positioned against the posteroseptal right atrium, close to the tricuspid annulus at the level of the posterior margin of the coronary sinus ostium. During atrial flutter, the distal bipolar electrogram at this site recorded a single atrial potential that was closer in timing to the first potential of the double potentials recorded just behind the coronary sinus ostium. This electrogram pattern was thought to represent activation in the proximal portion of the septal isthmus between the tricuspid annulus and the coronary sinus ostium. Sites that recorded atrial activation closer in timing to the second potential of the double potentials were avoided because this activation pattern could represent activation at a site distal to the exit of the septal isthmus. The tip electrode was then advanced slightly toward the right ventricle (at the level of the posterior margin of the coronary sinus ostium) until the distal bipolar electrogram recorded a low-amplitude atrial potential with a large, sharp ventricular potential, indicating a location close to the tricuspid annulus (Fig 7A). Radiofrequency current (550 to 650 kHz) then was delivered to the tip electrode at 45 to 60 V when using a 7F/4-mm tip electrode (Cordis Webster) in 20 patients and 50 to 70 V when using an 8F/8-mm tip electrode (EP Technologies) in 10 patients (Table 3). Two adhesive electrosurgical dispersive pads (both positioned over the left posterior chest) were used for the return electrode. The ablation electrode was withdrawn toward the posteroapical margin of the coronary sinus ostium in 2- to 3-mm increments every 15 to 20 seconds while radiofrequency current was continuously applied (Fig 6A and Fig 7A and 7B). When the electrode entered the posteroapical edge of the coronary sinus ostium (Fig 7B), the voltage was lowered to 45 to 50 V, and slight forward pressure (toward the tricuspid annulus) was applied to the catheter to enhance current delivery to the tissue between the coronary sinus ostium and the tricuspid annulus. When possible, radiofrequency current was delivered between the tricuspid annulus and the coronary sinus ostium as a single, continuous application. However, the application of radiofrequency current was terminated immediately in the event of an impedance rise (
10 
). In that event, two or more radiofrequency applications were required to produce the contiguous lesion. Voltage output was guided by impedance monitoring (avoiding more than 5- to 10- decreases in impedance) to reduce the incidence of impedance rise.35
If atrial flutter persisted after one or more radiofrequency applications between the tricuspid annulus and the coronary sinus ostium, radiofrequency current was applied along the posterior margin of the coronary sinus ostium and between the posterior margin of the ostium and the eustachian ridge (Figs 6B and 7C). If atrial flutter still persisted, radiofrequency current was applied along a line between the posterior or posterior paraseptal tricuspid annulus and the inferior vena cava or the eustachian ridge (Line PI in Fig 2C and Fig 6C).
This sequential approach, beginning with a lesion between the tricuspid annulus and the coronary sinus ostium, was used in 27 of the 30 patients. In the remaining 3 patients (Nos. 7, 19, and 24), radiofrequency current was only delivered between the posterior tricuspid annulus and the inferior vena cava or the eustachian valve/ridge (Fig 6C). In 2 of these patients (Nos. 7 and 19) it was thought that elimination of the posterior input to the AV node (slow AV nodal pathway) by ablation of the septal isthmus might produce AV block. One of these patients (No. 7) had previously undergone an AV nodal modification procedure using the anterior approach, which might have eliminated the anterior inputs to the AV node (fast AV nodal pathway). The other patient (No. 19) had an A-H interval of 205 ms during sinus rhythm (and no retrograde AV nodal conduction) after a myomectomy for hypertrophic obstructive cardiomyopathy, which might have indicated the absence of conduction over the fast AV nodal pathway (anterior inputs to the AV node). The third patient (No. 24) had a persistent left superior vena cava inserting into the great cardiac vein. This was associated with a giant coronary sinus ostium that displaced the eustachian ridge to approximately 4 cm from the tricuspid annulus. In this patient, we believed that a continuous lesion could be created more reliably through the posterior isthmus than through the septal isthmus.
Criteria for Successful Ablation and Termination of the Ablation Procedure
In 12 patients, ablation success was defined by (1) the termination of atrial flutter during an application of radiofrequency current due to conduction block within the reentrant circuit at the ablation site and (2) the inability to reinduce atrial flutter for a period of at least 30 minutes by programmed stimulation of the right and left atria (from the posterior or posterolateral coronary sinus), including extensive burst pacing (noninduction criteria). Isoproterenol was used in the postablation testing in patients who required isoproterenol for induction of atrial flutter before ablation at a dose exceeding the preablation dose.
In 18 patients, ablation success was defined by (1) the noninduction criteria and (2) the demonstration of a line of bidirectional conduction block between the tricuspid annulus and the eustachian valve/ridge (line of block criteria). We verified complete conduction block between the tricuspid annulus and the eustachian ridge by pacing the right atrium adjacent to the posterior tricuspid annulus (posterior to the ablation line) and noting that atrial activation propagates around the tricuspid annulus in the clockwise direction (as viewed in the left anterior oblique projection), with late atrial activation at the right anterior septum (His bundle electrogram) and even later atrial activation in the proximal coronary sinus, even though the electrodes in the proximal coronary sinus are anatomically close to the pacing site (Fig 8A and 8B), and by pacing the left atrium from the posterior coronary sinus and noting that atrial activation propagates around the tricuspid annulus in the counterclockwise direction as viewed in the left anterior oblique projection, with early atrial activation at the anterior septum (His bundle electrogram) and late atrial activation at the tricuspid annulus immediately posterior to the ablation line (Fig 8C and 8D).
View larger version (37K):
[in this window]
[in a new window]
|
Figure 8. Schematics illustrate the pacing technique used to verify a complete arc of conduction block after ablation of the septal isthmus (gray lines in B and D). A, Right atrial pacing adjacent to the posterior tricuspid annulus before ablation results in atrial activation around the tricuspid annulus in both the clockwise and counterclockwise directions with relatively early atrial activation recorded in the His bundle and proximal coronary sinus electrograms. B, After ablation of the septal isthmus, producing a complete arc of conduction block from the tricuspid annulus to the CS ostium, eustachian ridge, and IVC. Right atrial pacing adjacent to the posterior tricuspid annulus results in atrial activation around the tricuspid annulus only in the clockwise direction with late atrial activation recorded in the HB electrogram and even later activation recorded from the proximal CS. C, Left atrial pacing from the proximal CS before ablation results in right atrial activation from the septum in both the clockwise and counterclockwise directions. D, After ablation, left atrial pacing from the proximal CS produces right atrial activation only in the counterclockwise direction with the latest atrial activation recorded immediately posterior to the ablation line. Abbreviations as in previous figures. S indicates atrial pacing site.
|
|
Postablation Management
Patients were electrocardiographically monitored until hospital discharge on the second day after ablation. A transesophageal echocardiogram was obtained on the day after ablation to exclude a thrombus at the ablation sites, pericardial effusion, and tricuspid valve injury. The patients received aspirin (325 mg daily) for 6 weeks. No patient received antiarrhythmic drug therapy after ablation until recurrence of atrial flutter or atrial fibrillation. Patients were followed by the investigators or by the referring physician, and follow-up information was confirmed by telephone at the time of writing.
Statistical Analysis
Data are listed as mean±SD. The significance of the difference between the return intervals at the various entrainment pacing sites was assessed by ANOVA, with Scheffe's method for pairwise comparisons. A 
2 test was used to determine the significance of the difference in atrial flutter recurrence between the two criteria for successful ablation and the significance of the difference in the clinical occurrence of atrial fibrillation after ablation between the presence or absence of structural heart disease, atrial enlargement, and previously documented atrial fibrillation. The number of applications of radiofrequency current was compared between the patients with or without previous ablation failure and the presence or absence of an episode of atrial fibrillation after ablation using a two-tailed, unpaired t test. A value of P<.05 was considered statistically significant.
|
Results
|
|---|
Atrial flutter was present at the onset of the electrophysiological
study in 11 of the 30 patients. The ECG exhibited the pattern
of typical atrial flutter with a negative flutter wave in the
inferior leads in 10 of these 11 patients. Reverse typical atrial
flutter (defined later) with a positive flutter wave in the
inferior leads was present in the remaining patient (No. 20),
who had had clinical episodes of both typical and reverse typical
flutter. In the other 19 patients, typical atrial flutter was
induced by extrastimuli or burst atrial pacing from the right
atrial appendage and/or the coronary sinus (Fig 9A
). Isoproterenol
(1 to 2 µg/min) was required for sustained atrial flutter
in 6 patients (Table 2
). In 4 patients (Nos. 22 and 25 through
27), sustained typical atrial flutter could not be maintained
because of spontaneous termination or recurrent spontaneous
degeneration to atrial fibrillation. Episodes of reverse typical
atrial flutter also were induced in 7 patients before ablation.
Episodes of reverse typical atrial flutter were induced in 3
additional patients after elimination of typical atrial flutter
by ablation. Therefore, typical atrial flutter was studied in
29 patients and reverse typical atrial flutter in 10 patients.
The atrial cycle length during typical atrial flutter was 242±52
ms (Table 2
). The atrial cycle length of reverse typical atrial
flutter before ablation (7 patients) was 204±23 ms. The
atrial cycle length of reverse typical atrial flutter induced
only after the elimination of typical atrial flutter by ablation
(3 additional patients) was 235±59 ms (Table 2
). Sustained
atrial fibrillation (
60 seconds) was induced by programmed atrial
stimulation in 10 patients before ablation and 2 patients after
ablation of atrial flutter (Table 3
).
View larger version (44K):
[in this window]
[in a new window]
|
Figure 9. Induction of atrial flutter by burst atrial pacing from the coronary sinus. Tracings from the top are ECG lead II and bipolar electrograms recorded from the right atrial appendage (RAA), His bundle region (HB), lateral (TA9,10) to posterior tricuspid annulus (TA5), mapping catheter with distal pair of electrodes positioned at the anterior end of the septal isthmus (MAPd) and proximal pair of electrodes positioned just behind the coronary sinus ostium overlying the eustachian ridge (MAPp), and the posterior (CSp) to posterolateral coronary sinus (CSd). A, During sinus rhythm, three extrastimuli (S2, S3, S4) were delivered to the posterolateral coronary sinus. S3 and S4 activated the posterior left atrium (CS electrograms) and the right atrium anterior to the septal isthmus (MAPd and HB electrograms), but conduction block within the septal isthmus resulted in selective activation of the right atrium around the tricuspid annulus in the counterclockwise direction as viewed in the left anterior oblique projection, with early atrial activation recorded in the HB electrogram (A) followed by atrial activation in the TA electrograms from the lateral (TA9) to posterior (TA5) initiating typical atrial flutter. B, Electrograms during typical atrial flutter demonstrate atrial activation around the TA in the counterclockwise direction with secondary left atrial activation recorded progressively from CSp to CSd electrograms. Note the double potentials recorded in MAPp electrograms recorded just behind the CS ostium.
|
|
Endocardial Mapping
Recordings along the tricuspid annulus during typical atrial flutter showed atrial activation propagating around the tricuspid annulus in the counterclockwise direction as viewed in the left anterior oblique projection (anteriorly along the septum and posteriorly along the free wall) in all 29 patients (Figs 9 and 10). Left atrial activation, recorded from the proximal coronary sinus, occurred soon after the timing of atrial activation in the septal isthmus (Fig 9B). Left atrial activation was recorded at progressively later times at more distal sites in the coronary sinus, consistent with activation of the left atrium in the counterclockwise direction as viewed in the left anterior oblique projection. During reverse typical atrial flutter, atrial activation propagated around the tricuspid annulus in the clockwise direction as viewed in the left anterior oblique projection (Fig 11).
View larger version (35K):
[in this window]
[in a new window]
|
Figure 10. Double potentials recorded along the eustachian valve/ridge during typical atrial flutter. Catheter positions are shown in Fig 3. A, During typical atrial flutter, counterclockwise atrial activation around the tricuspid annulus is manifested by atrial activation in the HB electrograms followed by activation in the TA electrograms from lateral (TA5) to posterior (TA1) and then by activation at the septal isthmus (MAPd). The IVC-CS electrograms, recorded along the eustachian valve/ridge, show double potentials, which indicate a line of conduction block. The first potentials in the IVC-CS7 to IVC-CS3 electrograms (bold downward arrows) were recorded before the timing of atrial activation in the septal isthmus (MAPd and vertical dashed line) and resulted from activation along the posterior/inferior side of the eustachian valve/ridge. The second potential (upward arrows) was recorded after atrial activation in the septal isthmus and resulted from atrial activation on the anterior/superior side of the eustachian valve/ridge. The interval between the two potentials is shortest at the anterior end of the line of conduction block (IVC-CS3) and largest at the posterior end close to the inferior vena cava (IVC-CS7), consistent with pivoting of the wavefront around the anterior end of the eustachian ridge and the CS ostium. B, During sinus rhythm, electrograms along the eustachian valve/ridge (IVC-CS7 to IVC-CS3) did not exhibit distinct double potentials separated by an isoelectric interval. Abbreviations as in previous figures.
|
|
Electrograms recorded from the septal isthmus, between the coronary sinus ostium and the tricuspid annulus, exhibited wide atrial potentials that often had multiple components but not discrete double potentials separated by isoelectrical interval (MAPd electrogram in Fig 9B). Conduction block in the septal isthmus was responsible for the spontaneous termination of typical atrial flutter in 11 patients, which suggests that the septal isthmus may have a low safety factor for impulse propagation. Conduction block at this site in the anterior-to-posterior direction was responsible for the induction of typical atrial flutter by burst pacing from the coronary sinus in 6 patients (Fig 9A).
Double Potentials
Distinct double potentials separated by an isoelectrical interval (consistent with conduction block) were recorded along the eustachian valve/ridge from the anterior/superior margin of the coronary sinus ostium to the inferior vena cava in all 29 patients during typical atrial flutter (Figs 1, 3, and 10). The first of the two potentials was large and sharp and the second potential was small and rounded (distant) when the recording electrodes were positioned posterior/inferior to the line of equal amplitude double potentials, indicating a location proximal to (below) the line of block (MAP-A electrogram in Fig 12). The second of the two potentials was large and sharp and the first potential was small and distant when the recording catheter was positioned anterior/superior or distal to (above) the line of equal amplitude potentials (MAP-B electrogram in Fig 12), which suggests a location distal to (above) the line of block (MAP-C electrogram in Fig 12). Atrial activation in the septal isthmus between the tricuspid annulus and the coronary sinus ostium (site C in Fig 1) followed the timing of the first potential of the double potentials and preceded the second potential of the double potentials recorded along the eustachian valve/ridge (MAP electrogram in Fig 10). The timing of the second potential became progressively later from the coronary sinus end to the inferior vena cava end of the eustachian valve/ridge. The greatest interval between the two potentials (91±17 ms, Table 2) was consistently recorded near the inferior vena cava end of the eustachian valve/ridge. The second potential of the two potentials was usually recorded after atrial activation in the His bundle electrogram, which suggests that the second potential may represent activation outside of the reentrant circuit.
View larger version (24K):
[in this window]
[in a new window]
|
Figure 12. Characteristics of the double potentials recorded along the eustachian valve/ridge. The bottom three tracings are composed electrograms recorded at three sites shown schematically on the right (A, B, and C). MAP-B was recorded on the eustachian ridge. The two potentials of the double potential are approximately equal in amplitude. MAP-A was recorded on the posterior/inferior side of the eustachian ridge and exhibited a large, sharp first potential and a small, rounded second potential (representing far field activation on the anterior/superior side of the eustachian ridge). MAP-C was recorded anterior/superior to the eustachian ridge. The second potential is large and sharp, whereas the first potential is small and rounded (far field activation from the posterior/inferior side of the eustachian ridge). Abbreviations as in previous figures.
|
|
Double potentials also were recorded along the eustachian valve/ridge during reverse typical atrial flutter. The order of the two potentials was reversed (compared with typical flutter), with the first potential resulting from atrial activation anterior to (above) the eustachian valve/ridge and the second potential resulting from activation posterior to (below) the eustachian valve/ridge (Fig 11).
Double potentials were not recorded along the eustachian valve/ridge during sinus rhythm (Fig 10B). However, double potentials were consistently elicited by right atrial pacing at long cycle lengths just anterior or posterior to the eustachian valve/ridge, with activation occurring in the septal isthmus before activation of the opposite side of the eustachian valve/ridge (Figs 4 and 13).
View larger version (31K):
[in this window]
[in a new window]
|
Figure 13. Evidence obtained by atrial pacing for fixed conduction block across the eustachian valve/ridge. Catheter positions are shown in Fig 4. The anterior (Ant) and posterior (Post) catheters are positioned just anterior and posterior to the eustachian valve/ridge, respectively. A, During atrial pacing from the distal pair of electrodes on the Ant catheter at a cycle length of 600 ms, the proximal pair of electrodes on the Post catheter (Postp, located close to the eustachian valve/ridge) recorded two distinct potentials separated by an isoelectric interval. The first potential resulted from atrial activation anterior to the eustachian valve/ridge and the second (delayed) potential resulted from atrial activation posterior to the eustachian valve/ridge. The distal pair of electrodes on the Post catheter (Postd) were located further from the eustachian valve/ridge and recorded only a single delayed potential. The IVC-CS4 electrogram is recorded at the anterior margin of the coronary sinus ostium and shows atrial activation midway in timing between the anterior and posterior potentials recorded along the eustachian valve/ridge. B, During atrial pacing from the distal pair of electrodes on the Post catheter at a cycle length of 600 ms, the distal pair of electrodes on the Ant catheter (Antd, located close to the eustachian valve/ridge) recorded two atrial potentials separated by an isoelectric interval. The first potential resulted from atrial activation posterior to the eustachian valve/ridge, while the second (delayed) potential resulted from atrial activation anterior to the eustachian valve/ridge. During this recording, the IVC-CS5 electrodes were located at the anterior margin of the coronary sinus ostium and recorded atrial activation midway in timing between posterior and anterior potentials recorded along the eustachian valve/ridge. The presence of double potentials along the eustachian valve/ridge during atrial pacing at a long cycle length provides strong evidence for fixed (not functional) conduction block. C, Distinct double potentials (separated by an isoelectric interval) were not recorded during sinus rhythm, indicating that atrial activation was occurring nearly simultaneously on both sides of the eustachian valve/ridge. Abbreviations as in previous figures.
|
|
Entrainment Pacing
Entrainment pacing (cycle length, 15 to 25 ms shorter than the flutter cycle length) at the septal isthmus, between the tricuspid annulus and the coronary sinus ostium, produced an atrial activation sequence that was identical to the flutter in all recorded electrograms (concealed entrainment) in each of the 15 patients tested. The return interval (return interval minus flutter cycle length) at this pacing site was 0 to 15 ms (mean, 4.2±4.3 ms) and 8 ms in 14 of the 15 patients (Fig 14A and Table 2).
View larger version (43K):
[in this window]
[in a new window]
|
Figure 14. Entrainment pacing from the septal isthmus and anterior and posterior to the eustachian valve/ridge during typical atrial flutter. A, MAP electrograms were recorded from the septal isthmus between the tricuspid annulus and the coronary sinus ostium. The left tracing was recorded immediately before the onset of entrainment pacing. The tracing at the right shows the last 4 complexes of entrainment pacing at a cycle length of 215 ms from the distal pair of electrodes on the MAP catheter. The atrial activation sequence during entrainment pacing was identical to the atrial flutter. The return interval was 230 ms, equal to the atrial flutter cycle length ( return interval=0 ms). B, The MAP catheter was positioned just posterior to the eustachian valve/ridge, and the MAPd electrogram during atrial flutter shows double potentials with a large, sharp first potential (left). Entrainment pacing at this site produced an atrial activation sequence around the tricuspid annulus that was identical to the atrial flutter activation sequence but was associated with a long return interval of 260 ms ( return interval=30 ms), which indicated that the pacing electrodes were located some distance from the reentrant circuit ("blind alley"). C, The MAP catheter was positioned anterior to the eustachian valve/ridge, and the MAPd electrogram during atrial flutter shows double potentials, with a small (distant) first potential and a large second potential (left). Entrainment pacing at this site resulted in a longer return interval of 270 ms ( return interval=40 ms), indicating that the pacing site was located further from the reentrant circuit. Abbreviations as in previous figures.
|
|
Entrainment pacing just posterior to the eustachian valve/ridge, which corresponds to site B in Fig 1, produced an atrial activation sequence identical to the flutter but with a return interval of 8 to 50 ms (mean, 30±12 ms) and 15 ms in 14 of the 15 patients (Fig 14B and Table 2). Entrainment pacing just anterior to the eustachian valve/ridge (site E in Fig 1) produced a return interval of 15 to 60 ms (mean, 37±13 ms) (Fig 14C and Table 2).
Entrainment pacing from sites along the right atrial free wall adjacent to the tricuspid annulus produced some alternation of the atrial activation sequence close to the pacing site and alteration of the P wave. However, the return interval was 8 ms in 14 of the 15 patients (Fig 15 and Table 2). In the remaining patient (No. 23), the return interval was 15 ms at all sites tested around the tricuspid annulus and at the septal isthmus.
Catheter Ablation
In 27 of the 30 patients, ablation was initiated with applications of radiofrequency current delivered along a line between the tricuspid annulus and the posteroapical margin of the coronary sinus ostium, as illustrated in Fig 6A (approach A). Ablation only in this region eliminated atrial flutter in 14 of the 27 patients with 1 to 16 (median, 2) applications of radiofrequency current (column "A" in Table 3). Extending the ablation line to the eustachian ridge (including the posterior margin of the coronary sinus ostium), as illustrated in Fig 6B (approach A and B), eliminated the atrial flutter in 12 additional patients (column "A and B" in Table 3). In 2 of these 12 patients (Nos. 9 and 10), typical atrial flutter was eliminated by ablation approach A, but reverse typical atrial flutter then was induced by programmed atrial stimulation. The reverse typical atrial flutter was eliminated by extending the ablation line to the eustachian ridge. Therefore, ablation using approach A or approach A and B eliminated typical and reverse typical atrial flutter in 26 of the 27 patients with 1 to 21 (median, 3; mean, 5.8±5.5) applications of radiofrequency current. The one remaining patient (No. 16) required additional ablation between the tricuspid annulus and the inferior vena cava to eliminate atrial flutter (column "A, B, and C" in Table 3). This patient had a large coronary sinus ostium that was located more anteriorly than usual, and a His bundle potential was recorded at the anterior margin of the coronary sinus ostium. The unusual anatomy and our concern about the possibility of producing heart block significantly limited attempts to create a line of block between the tricuspid annulus and the coronary sinus ostium.
In 3 patients with either a high risk of AV block (patients 7 and 19) or a giant coronary sinus ostium caused by a persistent left superior vena cava inserting into the great cardiac vein (patient 24), ablation of atrial flutter was performed by delivery of radiofrequency current only between the posterior or posterior-paraseptal tricuspid annulus and the inferior vena cava or the eustachian ridge (Fig 6C and column "C Only" in Table 3). Five to 12 applications of radiofrequency current were required to eliminate atrial flutter.
The number of applications of radiofrequency current required to eliminate atrial flutter was not significantly different for 7 patients with previous ablation failure (mean, 6.6±5.9) compared with 23 patients without previous ablation procedures (mean, 6.7±6.2; Tables 1 and 3).
Criteria for Ablation Success
In 12 patients, ablation was considered successful and the procedure was terminated when an application of radiofrequency current terminated the atrial flutter (Fig 16) and neither typical nor reverse typical atrial flutter was induced by programmed atrial stimulation (noninduction criteria, Table 3). In the remaining 18 patients, ablation was not considered successful until the noninduction criteria were met and a line of complete bidirectional conduction block was produced between the posteroseptal tricuspid annulus and the eustachian valve/ridge (line of block criteria, Table 3). The completion of the line of conduction block was verified by right atrial pacing adjacent to the posterior paraseptal tricuspid annulus (posterior to the ablation line) and by pacing the left atrium from the posterior coronary sinus (equivalent to a site anterior to the ablation line). During right atrial pacing adjacent to the posterior tricuspid annulus, a contiguous line of block eliminated early anterior and leftward propagation of the atrial impulse, manifested by propagation of the atrial impulse around the tricuspid annulus in the clockwise direction (as viewed in the left anterior oblique projection) with late activation at the anterior septum (His bundle electrogram) and even later activation of the posterior left atrium recorded from the proximal coronary sinus (Fig 8A and 8B and Fig 17A, 17B, and 17E). A more striking shift in the pattern of atrial activation around the tricuspid annulus was observed during coronary sinus pacing. Before ablation, coronary sinus pacing resulted in activation of the right atrium in both the anterior and posterior directions, producing activation around the tricuspid annulus in both the counterclockwise and clockwise directions (Figs 8C and 17C). After completion of the line of conduction block, the right atrium was activated only in the anterior direction, which resulted in activation around the tricuspid annulus in only the counterclockwise direction (Figs 8D and 17D). The activation time at the posteroseptal right atrium (posterior to the ablation line) during coronary sinus pacing shifted from the earliest right atrial activation time before ablation to the latest time after ablation (Fig 17C and 17D).
View larger version (37K):
[in this window]
[in a new window]
|
Figure 16. Radiofrequency catheter ablation of typical atrial flutter using approach A. A, Tracings at the bottom are the current and voltage output of the radiofrequency generator. This figure shows the second application of radiofrequency current. The first application of radiofrequency current was delivered while the ablation catheter was slowly moved from the tricuspid annulus to the midpoint between the tricuspid annulus and the posteroapical margin of the coronary sinus ostium. The second application of radiofrequency current, shown in this figure, was delivered while the ablation electrode was moved from the end point of the first radiofrequency application to the posteroapical margin of the coronary sinus ostium at 51 V (30 W). B, Termination of atrial flutter 18 seconds after the onset of the second application of radiofrequency current, when the ablation electrode was maneuvered into the posteroapical margin of the coronary sinus ostium. Atrial flutter was terminated with conduction block in the septal isthmus, manifested by the absence of atrial activation after activation at the posterior tricuspid annulus (TA5 electrogram). Abbreviations as in previous figures.
|
|
View larger version (184K):
[in this window]
[in a new window]
|
Figure 17. Electrograms illustrate the line of block criteria for ablation success. Catheter positions are shown in E. A, Before ablation, right atrial pacing adjacent to the posterior tricuspid annulus (TA3 bipolar electrode) resulted in atrial activation around the tricuspid annulus in the clockwise direction (shorter upward arrow indicating activation from TA4 to TA8) and in the counterclockwise direction (represented by the downward arrow indicating propagation to the MAP electrogram recorded at the septal isthmus, followed by the long upward arrow labeled "Septal" and indicating propagation to the HB electrogram recording site at the anterior septum). The proximal coronary sinus (IVC-CS3 to IVC-CS1 electrograms) was activated relatively early after the pacing stimulus. B, After ablation of the septal isthmus, pacing adjacent to the posterior paraseptal tricuspid annulus (TA1 bipolar electrode) resulted in atrial activation propagating around the tricuspid annulus only in the clockwise direction, with a marked delay in the timing of atrial activation in the HB electrogram (relative to the RAA and TA8 electrograms) and even later activation recorded in the proximal coronary sinus (IVC-CS electrograms). Note the double potentials in the MAP electrogram recorded at the septal isthmus, indicating atrial activation on both sides of the line of conduction block produced by ablation. C, Pacing the left atrium from the posterior coronary sinus before ablation resulted in right atrial activation around the tricuspid annulus in both the clockwise direction (short upward arrow with serial atrial activation in TA1 to TA6 electrograms) and the counterclockwise direction (long upward arrow) with early atrial activation recorded in the HB electrogram followed by atrial activation at the anterior (TA8) and anterolateral tricuspid annulus (TA7). D, After ablation of the septal isthmus, left atrial pacing via the posterior coronary sinus resulted in atrial activation propagating around the tricuspid annulus only in the counterclockwise direction, with early atrial activation in the HB electrograms and later activation in the anterior (TA8 electrogram), lateral (TA5 electrogram), and posterior tricuspid annulus (TA2 electrogram). Latest atrial activation was recorded at the posterior paraseptal tricuspid annulus (TA1 electrogram) just posterior to the line of block across the septal isthmus. Note the two atrial potentials (double potentials) in the MAP electrogram. The findings in B and D confirm the presence of a complete arc of conduction block (extending from the tricuspid annulus to the coronary sinus ostium to the inferior vena cava), fulfilling the line of block criteria for ablation success. Abbreviations as in previous figures.
|
|
The presence of a complete line of conduction block was examined in 1 of the 3 patients in whom reverse typical atrial flutter was induced after elimination of typical atrial flutter by ablation (patient 9 in Table 3). Conduction across the ablation region was still present, manifested by early activation of the posterior right atrium adjacent to the tricuspid annulus during coronary sinus pacing. Extending the ablation line to the eustachian ridge was associated with the development of a complete line of conduction block and elimination of reverse typical atrial flutter.
In 15 of the 18 patients in whom line of block criteria were used to define ablation success, a line of conduction block was present as soon as the inducibility of typical and reverse typical atrial flutter was eliminated. In the remaining 3 patients (Nos. 28, 29, and 30), atrial flutter was terminated by the third to eighth application of radiofrequency current and neither forms of atrial flutter could be induced, but some degree of conduction across the ablation region was still present during pacing from the posterior right atrium (adjacent to the tricuspid annulus) or coronary sinus. In these 3 patients, the defect in the line of conduction block was found by pacing from the coronary sinus and mapping just posterior to the ablation line to locate an early atrial potential (Fig 18A and 18C). Ablation at that site was followed by a shift in the atrial activation sequence along the posterior tricuspid annulus from the clockwise direction to the counterclockwise direction as viewed in the left anterior oblique projection (Fig 18B, 18D, and 18E). Importantly, neither typical nor reverse typical atrial flutter was induced after the completion of the line of conduction block in any of the 18 patients.
View larger version (41K):
[in this window]
[in a new window]
|
Figure 18. Use of mapping to identify the defect in an incomplete ablation line across the septal isthmus. A, After 14 applications of radiofrequency current across the septal isthmus, the posterior margin of the coronary sinus ostium, and between the coronary sinus ostium and the eustachian ridge (ER), left atrial pacing from the posterior coronary sinus showed delayed (but not blocked) conduction to the posterior tricuspid annulus (TA1 electrogram). Mapping posterior to the ablation line with the ablation catheter (ABL electrogram) identified an early atrial potential (A) posterior to the septal isthmus, indicating the site of the defect in the ablation line, as shown schematically by the arrow in C. E, Ablation at this site resulted in an abrupt shift in the atrial activation sequence around the tricuspid annulus, with an increase in the stimulus-atrial (S-A) interval at the posterior tricuspid annulus (TA1 electrogram) from 145 to 180 ms, indicating the development of complete block between the tricuspid annulus and the eustachian ridge. B, After ablation, pacing from the posterior coronary sinus results in atrial activation around the tricuspid annulus only in the counterclockwise direction with early atrial activation recorded in the HB electrograms and later activation at the posterior tricuspid annulus (TA1 electrogram). Atrial activation was recorded latest in the ABL electrogram (small downward arrow), indicating completion of the line of conduction block as illustrated schematically in D. Abbreviations as in previous figures.
|
|
In 1 of the 12 patients (No. 7) in whom the noninduction criteria were used to define ablation success, conduction across the ablation line was still present after the fifth application of radiofrequency current despite noninducibility of atrial flutter. No further applications of radiofrequency current were delivered.
In 4 patients, ablation could not be performed during atrial flutter because of either frequent spontaneous termination of atrial flutter (patients 25 and 27) or frequent spontaneous conversion of atrial flutter to atrial fibrillation (patients 22 and 26). Ablation was performed during left atrial pacing from the posterior coronary sinus until a line of conduction block was evident by the abrupt delay in the timing of atrial
Top
Abstract
Introduction
