Ablation of ‘Incisional’ Reentrant Atrial Tachycardia Complicating Surgery for Congenital Heart Disease
Use of Entrainment to Define a Critical Isthmus of Conduction
Background Intra-atrial reentrant tachycardia occurs frequently after surgery for congenital heart disease and is difficult to treat. We tested the hypotheses that intra-atrial reentrant tachycardia in patients who had undergone prior reparative surgery for congenital heart disease could be successfully ablated by targeting a protected isthmus of conduction bounded by natural and surgically created barriers and that entrainment techniques could be used to identify these zones.
Methods and Results Eighteen consecutive patients with 26 intra-atrial reentrant tachycardias complicating surgery for congenital heart disease (9 atrial septal defect repair, 4 Fontan, 2 Mustard, 2 Senning, and 1 Rastelli procedure) underwent electrophysiological study and ablation attempts. Mapping of activation was facilitated by the deployment of catheters with multiple electrodes. Sites for ablation were sought that demonstrated entrainment with concealed fusion and at which the postpacing interval minus the tachycardia cycle length and the stimulus to P wave minus the activation time were <30 ms. These sites were considered to be within a narrow isthmus critical to the tachycardia mechanism. Anatomic barriers bordering the critical isthmus of conduction were identified on anatomic grounds, by the presence of areas of electrical silence or by the demonstration of split potentials signifying a line of block. Success was achieved in 15 patients with 21 arrhythmias. The median number of radiofrequency applications was 5. There was a wide range of activation times at successful sites (−30 to −250 ms). At a mean duration of follow-up of 17±8 months, 11 patients were asymptomatic and 9 did not require antiarrhythmia therapy.
Conclusions Successful ablation of intra-atrial reentrant tachycardia complicating surgery for congenital heart disease may be achieved by creation of an ablative lesion in a critical isthmus of conduction bounded by anatomic barriers. This isthmus may be identified by the presence of entrainment with concealed fusion and an analysis of the relationship between the postpacing interval and the tachycardia cycle length and between the activation time and the stimulus time. Because this isthmus is invariably confined on at least one aspect by a surgical repair site that is of central importance to the tachycardia mechanism, we suggest that this type of arrhythmia be given the descriptive designation of “incisional reentry.”
Intra-atrial reentrant tachycardia occurs frequently in patients who have undergone prior corrective surgery for congenital heart disease.1 2 3 These arrhythmias are often poorly tolerated and may be life-threatening. Antiarrhythmic agents are frequently ineffective, have the potential to adversely affect myocardial function, exacerbate coexisting sinus node dysfunction, and may also be proarrhythmic. The advent of radiofrequency ablation has provided curative therapy for the majority of patients with supraventricular tachyarrhythmias but no structural heart disease.4 5 Preliminary reports suggest that radiofrequency ablation may also be applicable when atrial arrhythmias are associated with congenital heart disease.6 7 8 However, in patients with complex and heterogeneous structural abnormalities, and in whom several arrhythmia mechanisms may frequently coexist, successful ablation represents a considerable challenge.
Recent reports have described the successful ablation of atrial flutter, with application of radiofrequency lesions that modify a critical isthmus of slow conduction.6 9 10 This isthmus, which is defined at least in part by naturally occurring anatomic barriers, may be localized using the responses to pacing during tachycardia that result in entrainment with concealed fusion.6 This method of entrainment mapping was initially described for the identification of and targeting of ablative lesions to a critical isthmus during ventricular tachycardia complicating healed myocardial infarction.11 12
The purpose of the present study was to test the hypotheses that intra-atrial reentrant tachycardia in patients who had undergone prior reparative surgery for congenital heart disease could be successfully ablated by targeting a protected isthmus of conduction bounded by natural and surgically created barriers and that entrainment techniques could be used to identify these zones. We describe the electrophysiological and anatomic characteristics of successful ablation sites in a consecutive series of such patients.
The study group consisted of 18 patients who underwent radiofrequency ablation between April 1992 and March 1995. Patients were included if they had undergone previous corrective atrium surgery for congenital heart disease and if they met the following diagnostic criteria for intra-atrial reentrant tachycardia (excluding typical atrial flutter): an atrial cycle length between 200 and 600 ms in the absence of antiarrhythmia medications, an abnormal P-wave axis and/or morphology during tachycardia and an endocardial activation sequence inconsistent with sinus origin, initiation and termination or resetting of the tachycardia with appropriately timed premature atrial extrastimuli, demonstration of manifest entrainment, and exclusion of an accessory pathway and AV nodal reentry. Patients with only typical (type 1) atrial flutter were not included in this study. Typical atrial flutter was identified by the presence of a characteristic “sawtooth” flutter wave pattern on the surface ECG, a counterclockwise macro-reentrant loop in the right atrium, and an area of slow conduction in the low right atrium between the TA and the IVC.
Patient characteristics are shown in Table 1⇓. Mean patient age was 26.6±15.1 years. The clinical results of four of these patients were reported in part in a report of a consecutive series of patients undergoing radiofrequency ablation at our institution6 and are included here for a more detailed analysis of electrophysiological criteria for ablation and to provide long-term follow-up.
The protocol was approved by the Committee on Human Research of the University of California, San Francisco, and written informed consent was obtained from the patients and/or their parents or guardians. All patients underwent diagnostic study and radiofrequency ablation at the same procedure. Antiarrhythmia medications, other than AV nodal blocking agents, were discontinued five half-life periods before the procedure, which was performed under conscious sedation with fentanyl and midazolam.
Mapping of endocardial activation was facilitated when possible by the deployment of catheters with multiple electrodes. This included either a 20-pole halo catheter or a custom 8F 20-pole steerable catheter with 2-mm interelectrode spacing, in most patients. In addition, one or two steerable octapolar catheters (2.5-mm bipolar recording distance, 10 mm between bipoles) were positioned in the right atrium. Specific anatomic positioning of these catheters was individualized in each patient to provide simultaneous electrical recordings from a wide array of atrial sites and also from regions in close proximity to abnormal anatomic architecture. Specifically, recordings were from patients with prior atrial septal defect repair in the region of the atriotomy site and at the margins of the atrioseptal patch, patients with Fontan repair at the atrial insertion of the Fontan conduit, and patients with Mustard or Senning procedures at the border zones of the atrial baffle.
All 12 surface leads and intracardiac electrograms, filtered from 30 to 500 Hz, were recorded on a computer-based digital amplifier-recorder system with optical disk storage (Prucka Engineering).
In those patients in whom the arrhythmia was not present at the commencement of the procedure, a previously described protocol was used for arrhythmia induction.6 Only arrhythmias that had been clinically documented were targeted for ablation.
Intracardiac Mapping and Entrainment Procedures
During atrial tachycardia, initial localization was performed using an evaluation of multiple simultaneous right or systemic venous atrial electrogram recordings. At each site the local AT was evaluated in relation to earliest onset of the surface P wave. An 8F quadripolar ablation catheter with a deflectable tip (5-mm distal ablation tip, 2-mm interelectrode spacing; EP Technologies) was used for detailed atrial mapping, particularly in proximity to surgical sites such as an atrial septal defect repair or a Fontan conduit or atriotomy scars. In this respect, an important part of the preprocedure evaluation was a careful review of the congenital anatomic defect and the reparative surgical operation(s) undertaken. In addition, the most recently performed echocardiograms were reviewed.
Our aim was to identify using entrainment techniques an isthmus of slow conduction that was critical to the tachycardia circuit. Initial entrainment was performed from the high right (or systemic venous) atrium during tachycardia at cycle lengths 10 to 50 ms less than that of the tachycardia to demonstrate the criteria for manifest entrainment. All 12 leads of the surface ECG were carefully examined to detect the frequently subtle changes observed during manifest entrainment. When the surface P wave could not be clearly demonstrated (because it was partially or completely obscured by the QRS complex or the preceding T wave), then intravenous esmolol was infused cautiously in an attempt to increase the degree of AV block. In those cases in which a P wave could not be clearly demonstrated despite this maneuver, the presence or absence of fusion was evaluated using the endocardial activation sequence recorded from a minimum of eight widely separated bipolar sites. Potential target sites were identified at which local recorded activation preceded onset of the surface P wave and in which this relationship remained constant during manifest entrainment. At these sites, the rove catheter was then used for pacing to determine whether entrainment with concealed fusion was present and to characterize whether the site was within the reentrant circuit. To minimize the effects of possible decremental conduction that would be expected to prolong the ST and the PPI during entrainment, the longest cycle lengths that reliably entrained the tachycardia were analyzed. Stimulation was performed at twice diastolic threshold using a 2-ms pulse width.
To characterize entrainment of reentrant atrial tachycardia, we have adopted definitions as previously described for entrainment of ventricular tachycardia.11
Entrainment with manifest fusion. Entrainment demonstrating surface ECG evidence of constant fusion at a constant pacing rate and progressive fusion with incremental pacing. Fusion occurs because of atrial activation that occurs in part from the stimulus site and in part from the previous paced wave front exiting the zone of slow conduction13 (Fig 1⇓). When entrainment is manifest, the last captured P wave is entrained at the pacing cycle length but does not demonstrate fusion.12 13
Entrainment with concealed fusion. Entrainment in which there is no evidence of surface fusion and in which there is a delay between the stimulus artifact and the P wave. Fusion does not occur because pacing is from within the narrow or critical conduction isthmus and all atrial activation registered at the body surface occurs from the exit site in the same manner as during spontaneous tachycardia (Fig 1⇑). The first P wave after the last captured P wave returns at the TCL.
Postpacing interval. The time from the last stimulus artifact (distal electrode pair) to the first recorded electrogram (proximal electrode pair) at the pacing site.
Tachycardia cycle length. Taken as an average of 3 intervals after the first PPI.
Stimulus time. Time from the pacing artifact to onset of the next recorded P wave at sites demonstrating entrainment with concealed fusion. This time corresponds to the time taken for the stimulus impulse to exit orthodromically from a protected area of slow conduction.
Activation time. Time from the electrogram recorded at the pacing site to onset of the next P wave during spontaneous (nonentrained) tachycardia.
Critical isthmus sites. These sites were characterized by the presence of entrainment with concealed fusion, and where the PPI−TCL and the ST−AT were ≤30 ms. For sites with these characteristics, on the basis of previous work of entrainment in ventricular tachycardia,12 we arbitrarily subdivided into three groups: Sites were considered to be entrance when the AT expressed as a percentage of the TCL was ≥60%; to be exit when this ratio was ≤30%; and to be central when the ratio was between 30% and 60%.
Bystander sites. Sites demonstrating entrainment with concealed fusion but at which the PPI−TCL and the ST−AT were >30 ms.
Outer loop sites. Sites demonstrating entrainment with manifest fusion but at which the PPI−TCL was ≤30 ms.
Split potentials. Split potentials were defined by the demonstration of two discrete electrogram components separated by an isoelectric interval.14
Characterization of the Critical Isthmus of Conduction and Identification of Anatomic Boundaries
In all patients, attempts were made to fully characterize the zone of slow conduction and its anatomic boundaries.
When entrainment with concealed fusion was demonstrated, the extent of the critical slow zone was carefully mapped to identify entrance sites, exit sites, and bystander regions. Barriers bordering the protected isthmus were identified: (1) on anatomic and fluoroscopic grounds (eg, the IVC and SVC, the TA, or the ring of a Fontan conduit); (2) electrophysiologically by the demonstration of split potentials—entrainment techniques were also used to prove that activation timing of the discrete split potentials was markedly disparate on either side of the barrier, and the location of a barrier to conduction was inferred when small catheter movements resulted in large changes in activation timing with loss of concealed entrainment; and (3) electrophysiologically by the demonstration of areas from which an endocardial electrogram could not be recorded (“electrically silent” areas). Such areas included over an atrial septal patch, over an interatrial baffle, within a Fontan conduit, or in an area of right atrial patch augmentation. The border of these areas was identified when a small catheter movement resulted in the recording of an electrogram with sharp components (ie, not a far-field electrogram), when no electrical activity had been present before catheter movement, and when the fluoroscopic catheter location was consistent with the anatomic location of prosthetic material.
Ablation was performed between the distal electrode of the roving catheter and a large surface-area cutaneous electrode with 550-kHz unmodulated radiofrequency current from a generator (EP Technologies). In more recent cases, radiofrequency power was continuously adjusted to obtain a catheter-tip temperature of 60°C to 75°C.
Radiofrequency energy was only delivered to locations meeting the electrophysiological criteria for critical conduction sites based on the entrainment techniques described above.
Radiofrequency lesions were applied in an attempt to bridge two barriers, either surgical or naturally occurring. For patients in whom entrainment with concealed fusion could be demonstrated from more than one region (eg, in patient 8, both between the atriotomy and the IVC and also between the atriotomy and the SVC), radiofrequency energy was first targeted to the area at which the isthmus was considered to be anatomically narrowest and/or at which catheter access was most easily achieved. In addition, rather than attempting to target a single site, radiofrequency energy was applied to create a lesion bridging the two barriers. This was performed either with sequential radiofrequency applications or by slowly “dragging” the catheter during the application.
After termination of tachycardia with ablation, reinduction of the arrhythmia was attempted using the same pacing paradigm as described above. Successful ablation was defined as termination of tachycardia during radiofrequency application with an inability to reinduce the arrhythmia.
Values are expressed as mean±SD. Statistical comparisons were performed using the Student’s t test, paired or unpaired as appropriate. Statistical significance was accepted at the P<.05 level.
Eighteen patients (age range, 3 to 55 years; mean age, 26.6±15.1 years) underwent ablation of 26 atrial tachycardias. Patients had failed a mean of 3.5±1.4 antiarrhythmic medications and had been symptomatic for a mean of 7.3±5.2 years.
All patients had undergone prior surgery for congenital heart disease, the details of which are included in Table 1⇑.
Successful ablation was achieved in 15 of 18 patients and 21 of 26 arrhythmias. The median number of radiofrequency applications was 5 (range, 3 to 15). Characteristics and sites of successful ablation are shown in Table 2⇓.
There was a wide range in mean AT at successful sites (range, −30 to −250 ms; mean −104.0±66.1 ms, Fig 2A⇓). The mean AT at successful sites, expressed as a percentage of the TCL, was 34.9±23.9% and ranged from 10.2% to 83.3%. There were 11 sites at which the AT expressed as a percentage of the TCL was <30% (exit), 6 between 30% and 60% (central), and 4 >60% (entrance).
Entrainment Mapping and Ablation Sites
In all patients, manifest entrainment was initially achieved, usually from sites distant to surgically created barriers (Fig 3A⇓). Pacing from these sites demonstrated constant and progressive surface and endocardial fusions, with the first apparent postpacing beat being entrained at the pacing rate but not demonstrating fusion. Entrainment with concealed fusion (Fig 3B⇓) was achieved in all 21 of the successfully ablated tachycardias and was most reliably demonstrated in close proximity to naturally occurring and surgically created barriers. All of these sites also demonstrated fractionation of the local electrogram.
Sites at which entrainment with concealed fusion could be achieved and successful ablation performed included the following (sites of successful ablation in patients with prior repair of an atrial septal defect are also shown schematically in Fig 4⇓).
(1) The area between the inferior border of the atriotomy scar and the TA or between the lower end of the atriotomy and the IVC—in these patients concealed entrainment could be demonstrated over a considerable distance along an extensive isthmus between the TA and the atriotomy. STs along this isthmus varied by up to 100 ms (Fig 5⇓).
Interestingly, when pacing after successful ablation from some successful sites at which a long ST was demonstrated during concealed entrainment before ablation, a short ST was then noted along with a marked change in P-wave axis and morphology.
(2) In one patient, concealed entrainment was demonstrated in the isthmus between the IVC and the TA, successful ablation was achieved with a linear lesion between the IVC and the posterolateral TA; (3) in the area between the atriotomy and the SVC in one patient; (4) near the border of the atrial septal patch and between it and the TA; (5) in patients with Fontan repairs, in the region near the base of the AV connection or conduit, between it and either a suture line, or the anterolateral TA; (6) in two patients who had undergone Mustard procedures, between the baffle and either the ostium of the coronary sinus or the TA; and (7) in one patient who had undergone Senning repair, between the atrial baffle and the TA.
At sites at which entrainment with concealed fusion was present, analysis of the PPI minus the TCL and of the ST minus the AT demonstrated the following regions: (1) areas at which the difference between the PPI and the TCL and the difference between the ST and the AT were both <30 ms (Fig 6B⇓ and 6C⇓). At these sites the mean PPI was 302.8±52.2, and the mean TCL was 302.0±54.6 (P=NS, paired Student’s t test). Thus, the mean PPI minus the TCL at successful sites was 0.8±12.7 ms (Fig 2B⇑). The mean ST minus the AT at successful sites was 3.4±10.2 ms (P=NS, paired Student’s t test) (Fig 2C⇑). As noted above, sites that demonstrated entrainment with concealed fusion could be entrance (Fig 6B⇓), central, or exit (Fig 6C⇓).
(2) Areas at which the PPI minus the TCL and the ST minus the AT were >30 ms (Fig 6D⇑), despite the presence of entrainment with concealed fusion. These sites were considered to be in bystander areas not critical to the circuit, and therefore radiofrequency energy was not applied.
In all patients, after identification of a critical isthmus of conduction, radiofrequency ablative lesions were targeted in an attempt to bridge naturally occurring or surgically created barriers on either side of this isthmus. Successful sites of ablation are shown in Table 2⇑, and all of these sites demonstrated the characteristics of sites critical to the circuit (ie, early activation, concealed entrainment, PPI minus TCL <30 ms and ST minus AT <30 ms).
In all patients, anatomic sites were mapped at which discrete split potentials could be recorded during tachycardia. These sites included along the ridge of an atriotomy scar in 16 patients, near the base of the Fontan conduit in 2 patients, and at the border region of the atrial septal patch or baffle in 5 patients.
Areas at which split potentials could be recorded were considered to possibly represent one of the barriers that a radiofrequency lesion must bridge to successfully terminate the tachycardia. When split potentials were found, attempts were again made to demonstrate entrainment with concealed fusion to show that one component arose from an isthmus critical to the maintenance of the tachycardia (Fig 6E⇑). In most patients, split potentials could be recorded over a distance of up to 2 to 3 cm with the rove catheter or 2 to 3 bipoles on the multipolar recording catheters (Fig 6E⇑).
In two patients we observed the phenomenon of complete intra-atrial block with two dissociated atrial rhythms occurring simultaneously. This included sinus rhythm with atrial tachycardia in one patient, and atrial fibrillation in one part of the atrium simultaneous with atrial tachycardia occurring elsewhere in another. An additional patient also demonstrated atrial tachycardia with Wenkebach conduction to another atrial site. Intra-atrial block was most frequently observed after partially successful ablation and was terminated once ablation was complete.
In 3 of 18 (17%) patients ablation was unsuccessful. In 1 patient with a Fontan repair (patient 17), sinus rhythm was present at the start of the study. Reentrant atrial tachycardia was easily inducible but was very poorly tolerated hemodynamically owing to rapid one-to-one AV nodal conduction. Pharmacological attempts to increase the level of AV block were unsuccessful. In addition, one-to-one AV nodal conduction caused the P wave to be obscured by the preceding T wave, therefore preventing analysis of activation timing and manifest versus concealed entrainment. In 1 patient with a Fontan type repair (patient 16), complete surgical exclusion of the right atrium had been performed. Catheter access to the right atrium could only be achieved by a retrograde approach (transaortic, transmitral, and across an atrial septal defect). This severely limited the ability to manipulate the ablation catheter. In 1 patient who had undergone a Senning repair (patient 6), ablation of the initially induced atrial tachycardia was successful but with immediate onset of a morphologically different slower atrial tachycardia. This was mapped to the pulmonary venous atrium by retrograde catheter approach but manipulation here was very difficult, and an adequate candidate site was not identified.
In the 15 patients in whom ablation was successful, the mean duration of follow-up is 17±8 months (range, 1 to 34 months). Nine patients are asymptomatic and require no antiarrhythmic therapy. Two patients are asymptomatic when on antiarrhythmic regimens that previously were ineffective. It is not known whether the tachycardia being treated is the same as the one targeted for ablation. Two patients have had recurrence on antiarrhythmic therapy but are symptomatically improved. In 1 of these patients, recurrent atrial tachycardia is of different morphology to that successfully ablated and is of considerably longer cycle length. One patient had recurrence of a clinically significant arrhythmia despite ongoing antiarrhythmic therapy. One patient developed sustained atrial fibrillation 4 months after the ablative procedure, without evidence of recurrence of atrial tachycardia.
This study describes an electrophysiological technique for identifying an isthmus of conduction that is critical to the maintenance of reentrant atrial tachycardia complicating reparative surgery for congenital heart disease. Because this isthmus is invariably confined on at least one aspect by a surgical repair site that is of central importance to the tachycardia mechanism, we have elected to designate this type of arrhythmia as “incisional reentry.” With the use of entrainment with concealed fusion at sites of early activation, we were able to target sites for successful radiofrequency ablation of 21 of 26 (81%) incisional reentrant tachycardias in 15 of 18 (83%) consecutive patients. Radiofrequency lesions were applied in a regional fashion to bridge anatomic barriers bordering the critical isthmus of conduction.
Observations on Entrainment Mapping: Identification of a Protected Isthmus That Is Critical to the Tachycardia Circuit
The model of a macro-reentrant tachycardia circuit that includes a central common pathway, outer loops, bystander regions, and entrance and exit sites has been well described in reentrant ventricular tachycardia complicating myocardial scarring.11 This is the first comprehensive report to demonstrate that a similar concept may be applicable in patients with atrial tachycardia complicating prior corrective surgery for congenital heart disease.
In all patients, the criteria for entrainment with manifest fusion could be demonstrated, proving a reentrant mechanism of the tachycardia. At some of these sites, in the presence of manifest entrainment, the PPI was equal to the TCL, consistent with the hypothesis that although the site is not within the critical isthmus of slow conduction it is nevertheless within the tachycardia circuit and is consistent with a so-called outer loop site11 12 (Fig 1⇑). Application of radiofrequency energy at these sites would not be expected to terminate tachycardia because the reentrant wave front would simply continue around the lesion.
Entrainment with concealed fusion was also easily achieved, and sites with the following characteristics were found: (1) The ST minus the AT and the PPI minus the TCL were both <30 ms (Figs 1⇑, 6B⇑, and 6C⇑). These were considered to be within the critical isthmus of slow conduction and demonstrated a wide range of ATs. Successful ablation was achieved at sites with these characteristics. The observation that tachycardia termination did not necessarily occur when radiofrequency energy was applied at these sites may reflect the considerable width of the isthmus and the necessity of creating several discrete lesions (or a long continuous one) to successfully bridge this zone.
(2) The ST minus the AT and the PPI minus the TCL were >30 ms. These sites, although within a relatively protected zone (and hence not demonstrating surface fusion), were consistent with bystander regions not critical to tachycardia maintenance. At these sites, the ST would be increased by the conduction time out of the blind loop, and the AT reduced by the conduction time into the blind loop (Figs 1⇑ and 6D⇑). ST would therefore exceed AT. Similarly, the PPI would be expected to exceed the TCL by the sum of the conduction times into and out of the blind loop. Radiofrequency energy was not applied at sites with these characteristics. Similar to the situation in ventricular tachycardia complicating coronary artery disease,11 patients with underlying congenital heart disease, previous (often multiple) atrial surgeries, and chronically abnormal hemodynamics will have considerable areas of (atrial) scarring and slow conduction, not all of which will necessarily be critical to maintenance of a tachycardia circuit.
To demonstrate the presence of entrainment with concealed fusion, it is of paramount importance to clearly visualize the P wave. Furthermore, to measure the ST and activation intervals, the onset of the P wave must also be clearly discernible. To this end, careful attention was given to the P-wave morphology in all 12 ECG leads during tachycardia and during pacing that resulted in entrainment. Since the surface P wave may be partially obscured on some beats by the QRS complex or the preceding T wave, additional information was obtained by examining the endocardial activation sequence from at least eight widely separated sites during entrainment to show that it was not different from the sequence during spontaneous tachycardia.
Relationship Between AT and Ablation Success
On the basis of recent work in which entrainment mapping was used to classify ventricular tachycardia circuits in patients with coronary artery disease,11 12 we classified those sites at which the AT was <30% of the TCL as exit sites, those between 30% and 60% as central zones, and those >60% as entrance sites.11 We found no relationship between success of ablation and any particular region in the critical isthmus, whether exit, central, or entrance. In theory, successful ablation should be achieved with a lesion that completely divides the critical isthmus at any part of its course. In practice, this will be achieved most readily in the region at which the isthmus is narrowest or at the anatomic location most easily accessible with the ablation catheter. In view of the considerable anatomic variations that may exist among patients who have undergone prior reparative atrial surgery, it would be unlikely that this site would be always entrance, central, or exit and we have therefore not specifically targeted sites according to these criteria. Consistent with this, recent reports of ablation of both type 1 (typical) human atrial flutter6 9 15 and reentrant ventricular tachycardia complicating coronary artery disease11 16 have described success at regions with characteristics of entrance sites, central zones, or exit sites.
Identification of Anatomic Barriers
Once the presence of a critical isthmus of slow conduction has been demonstrated by entrainment mapping, then the challenge is to determine the anatomic location of the confining barriers. A schematic representation of potential reentrant circuits and the barriers involved in patients with reentrant atrial tachycardia after repair of an atrial septal defect is shown in Fig 4⇑. In our experience, the location of at least one of these barriers is usually readily apparent. Such barriers included the TA, the atrial septal patch, the IVC or SVC in patients with atrial septal defect repair, or in other groups the base of a Fontan conduit or an atrial septal baffle, the latter identified by the loss of an electrogram recording. More difficult to determine were the locations of barriers such as an atriotomy that could not be precisely localized with fluoroscopy. One technique used to identify these sites was the recording of discrete split potentials. Entrainment techniques were then used to verify that the split potentials did represent activation on either side of a barrier that was critical for maintenance of the tachycardia. The presence of split potentials has previously been demonstrated both in canine models of atrial flutter17 18 19 and in human atrial flutter14 and has been attributed to the presence of an area of block, either functional or fixed. Feld and Shahandeh-Rad17 18 recorded double potentials in a canine right atrial crush injury model of atrial flutter and attributed them to activation on either side of a line of block produced by the crush injury. We have recently recorded split potentials during human type 1 atrial flutter.20 These split potentials represented activation on either side of the crista terminalis (confirmed with intracardiac echocardiography) that acted as a barrier to conduction.
Whether conduction in the critical isthmus must necessarily be slow to maintain reentry in these patients is uncertain, and there is conflicting evidence from animal models.17 18 21 22 23 The presence of broad fractionated electrograms in all patients in this study at the sites of successful ablation suggests the presence of relative uncoupling of atrial fibers and/or distorted fiber orientation (nonuniform anisotropic conduction) and provides indirect evidence that slow conduction was present.24 25
In this study, acute success was achieved in 83% (15/18) of patients. In 72% (13/18) of patients, long-term clinical improvement was observed; and 50% (9/18) of patients were asymptomatic and did not require medical therapy at a mean follow-up of 17 months. Although these success rates are considerably lower than when ablation is performed in patients without structural heart disease,4 5 this is not unexpected in view of the considerable atrial pathology present in patients with repaired congenital heart disease. Clinical recurrences may potentially represent progression of the disease process with emergence of a new tachycardia or recurrence of the original arrhythmia. In a thickened and scarred atrium, the lesions created with radiofrequency energy may possibly be of insufficient depth.
Comparison With Previous Studies
Recently, Triedman et al8 described 10 patients undergoing successful radiofrequency ablation of intra-atrial reentrant tachycardia after surgical palliation of congenital heart disease. Patients had undergone a Fontan procedure (6), a Mustard or Senning repair (2), or repair of a ventricular septal defect with outflow obstruction (2). These authors targeted areas of slow conduction identified on the basis of electrogram timing and fractionation but were only able to demonstrate concealed entrainment in a small proportion of tachycardia circuits. Our study extends the results of this report by (1) positing the utility of concealed entrainment to identify a critical isthmus and (2) including patients with reentrant atrial tachycardia complicating atrial septal defect repair.
Cruz et al26 recently reported preliminary data of successful radiofrequency ablation in four patients with reentrant atrial tachycardia occurring late after atrial septal defect repair. In all four patients the successful site was located between the inferior atriotomy and the IVC at sites at which early and fractionated electrograms were recorded. Entrainment data were not presented.
We based our approach to ablation on previous data demonstrating that the PPI minus the TCL and the ST minus the AT were important discriminators of those sites critical to the circuit from those located in bystander areas.11 In patients with already extensively scarred atria and potential substrate for multiple reentrant arrhythmias, we did not consider it desirable to create additional lesions at sites that were considered to carry a low probability of success. As a result we cannot estimate the sensitivity and specificity of the finding of concealed entrainment with criteria consistent with a critical slow zone site compared with areas consistent with bystander regions or with outer loop sites that did not show concealed entrainment but were within the circuit. It is possible that a critical zone may be mistaken for a bystander area if decremental conduction occurs during pacing. However, to decrease the likelihood of this occurrence, stimulation was performed at the longest cycle length at which capture and entrainment could be demonstrated. In addition, we did not observe a transient increase in the TCL after pacing, suggesting that conduction velocity was not altered by pacing at rates used in this study.
We paced from the distal electrode pair of the catheter and recorded from the proximal pair, which may potentially introduce error because the recording site is not identical to the pacing site. Stevenson et al12 have recently suggested unipolar pacing from the distal electrode and recording from the distal bipole in an attempt to minimize this problem.
Although we have provided strong circumstantial evidence for the presence of barriers to conduction, direct endocardial visualization was not performed.
We have demonstrated the feasibility of terminating reentrant atrial tachycardia complicating surgery for congenital heart disease using a technique involving creation of an ablative lesion in a critical isthmus of conduction between two anatomic barriers. This isthmus may be identified by the presence of entrainment with concealed fusion. Confirmation that the site is critical to the tachycardia circuit is obtained by an analysis of the relationship between the PPI and the TCL and between the AT and the ST during entrainment of the tachycardia. Because the surgical repair sites are invariably critical to the development and maintenance of reentrant atrial tachycardia, we use the term incisional reentry to describe these arrhythmias. Patients in this group who have presented a considerable challenge to effective therapy may be candidates for potentially curative percutaneous ablation guided by electrophysiological mapping principles.
Selected Abbreviations and Acronyms
|IVC||=||inferior vena cava|
|SVC||=||superior vena cava|
|TCL||=||tachycardia cycle length|
This work was supported by National Institutes of Health grant HL-45664. Dr Kalman is funded as the Ralph Reader Overseas Research Fellow of the National Heart Foundation of Australia and is the recipient of a Telectronics traveling grant from the Royal Australasian College of Physicians. Dr Olgin is funded by the North American Society of Pacing and Electrophysiology and is the recipient of the Albert Hyman Post-Doctoral Fellowship.
- Received May 30, 1995.
- Revision received August 10, 1995.
- Accepted September 11, 1995.
- Copyright © 1996 by American Heart Association
Driscoll DJ, Offord KP, Feldt RH, Schaff HV, Puga FJ, Danielson GK. Five and fifteen year follow-up after Fontan operation. Circulation. 1992;85:469-496.
Lesh MD, Van Hare GF, Schamp DJ, Chien W, Lee MA, Griffin JC, Langberg JJ, Cohen TJ, Lurie KG, Scheinman MM. Curative percutaneous catheter ablation using radiofrequency energy for accessory pathways in all locations: results in 100 consecutive patients. J Am Coll Cardiol. 1992;19:1303-1309.
Lesh MD, Van Hare GF, Epstein LM, Fitzpatrick AP, Scheinman MM, Lee RJ, Kwasman MA, Grogin HR, Griffin JC. Radiofrequency catheter ablation of atrial arrhythmias: results and mechanisms. Circulation. 1994;89:1074-1089.
Triedman JK, Saul JP, Weindling SN, Walsh EP. Radiofrequency ablation of intra-atrial reentrant tachycardia after surgical palliation of congenital heart disease. Circulation. 1995;91:707-714.
Feld GK, Fleck RP, Chen PS, Boyce K, Bahnson TD, Stein JB, Calisi CM, Ibarra M. Radiofrequency catheter ablation for the treatment of human type 1 atrial flutter: identification of a critical zone in the reentrant circuit by endocardial mapping techniques. Circulation. 1992;86:1233-1240.
Stevenson WG, Khan H, Sager P, Saxon LA, Middlekauff HR, Natterson PD, Wiener I. Identification of reentry circuit sites during catheter mapping and radiofrequency ablation of ventricular tachycardia late after myocardial infarction. Circulation. 1993;88:1647-1670.
Waldo AL, Carlson MD, Henthorn RW. Atrial Flutter: Transient Entrainment and Related Phenomena. Philadelphia, Pa: WB Saunders Co; 1990:530-536.
Kirkorian G, Moncada E, Chevalier P, Canu G, Claudel JP, Bellon C, Lyon L, Touboul P. Radiofrequency ablation of atrial flutter: efficacy of an anatomically guided approach. Circulation. 1994;90:2804-2814.
Morady F, Harvey M, Kalbfleisch SJ, el-Atassi R, Calkins H, Langberg JJ. Radiofrequency catheter ablation of ventricular tachycardia in patients with coronary artery disease. Circulation. 1993;87:363-372.
Feld GK, Shahandeh-Rad F. Mechanism of double potentials recorded during sustained atrial flutter in the canine right atrial crush-injury model. Circulation. 1992;86:628-641.
Olgin JE, Kalman JM, Fitzpatrick AP, Lesh MD. The role of right atrial endocardial structures as barriers to conduction during human type 1 atrial flutter: activation and entrainment mapping guided by intracardiac echocardiography. Circulation. 1995;92:1839-1848.
Frame LH, Page RL, Boyden PA, Fenoglio JJ Jr, Hoffman BF. Circus movement in the canine atrium around the tricuspid ring during experimental atrial flutter and during reentry in vitro. Circulation. 1987;76:1155-1175.
Frame LH, Page RL, Hoffman BF. Atrial reentry around an anatomic barrier with a partially refractory excitable gap: a canine model of atrial flutter. Circ Res. 1986;58:495-511.
Ortiz J, Igarashi M, Gonzalez HX, Laurita K, Rudy Y, Waldo AL. Mechanism of spontaneous termination of stable atrial flutter in the canine sterile pericarditis model. Circulation. 1993;88:1866-1877.
Ellis WS, Auslander DM, Lesh MD. Effects of coupling heterogeneity on fractionated electrograms in a model of nonuniformly anisotropic ventricular myocardium. J Electrocardiol. 1994;27:171-177.
Cruz FES, Fagundes M, Boghossian S, Ribeiro JC, Vanheusden L, Maia IG. A comparison of radiofrequency ablation of common type and atriotomy scar related atrial flutter. PACE. 1995;18:860. Abstract.