Author + information
- Received January 12, 2004
- Revision received April 13, 2004
- Accepted April 18, 2004
- Published online July 21, 2004.
- Atul Verma, MD*,
- Nassir F. Marrouche, MD*,* (, )
- Niranjan Seshadri, MD*,
- Robert A. Schweikert, MD*,
- Mandeep Bhargava, MD*,
- J.David Burkhardt, MD*,
- Fethi Kilicaslan, MD*,
- Jennifer Cummings, MD*,
- Walid Saliba, MD* and
- Andrea Natale, MD*
- ↵*Reprint requests and correspondence:
Dr. Nassir F. Marrouche, Section of Cardiac Electrophysiology, Department of Cardiology/Desk F15, Cleveland Clinic Foundation, 9500 Euclid Avenue, Cleveland, Ohio 44195, USA.
Objectives In patients with atrial flutter (AFL) and postoperative right atrial incisional scars, we sought to assess if the use of additional ablative lesions that targeted all potential re-entrant circuits, regardless of the presenting type of flutter, would prevent long-term recurrence.
Background Patients with AFL and incisional scars have a complex atrial substrate that may promote multiple mechanisms of intra-atrial re-entry.
Methods Twenty-nine patients with single right atrial incisional scars undergoing ablation for scar-dependent (n = 15) and cavotricuspid isthmus (CTI)-dependent (n = 14) flutter were studied.
Results In the scar-dependent group, 9 of 15 (60%) patients had inducible or spontaneous CTI-dependent flutter immediately after ablation. In the group with CTI flutter, 7 of 14 (50%) patients had scar-related flutter immediately after ablation. If a second type of flutter was found during the initial ablation, a second ablation was performed either along the isthmus (scar-dependent group) or from the scar to another anatomic boundary (isthmus-dependent group). Patients were followed for 24 ± 5 months and 18 ± 6 months in the scar- and CTI-dependent groups, respectively. In the scar-dependent group, five of six (83%) who underwent only a single flutter line had recurrence at 3 ± 1 months. In the isthmus-dependent group, three of seven (42%) patients who had only one flutter line performed had recurrence at 5 ± 3 months. There was no flutter recurrence in patients who initially received two different flutter lines or in patients who subsequently underwent a second flutter line at follow-up.
Conclusions In patients with postoperative right atrial incisional scar and flutter, multiple ablation lines that target both scar-related and classic isthmuses appear necessary to prevent long-term recurrence.
In patients after cardiac surgery, atrial flutter (AFL), also referred to as an intra-atrial re-entrant tachycardia, is a common occurrence and represents a significant cause for morbidity and even mortality (1,2). As understanding of the electrophysiologic mechanisms of postoperative AFL has evolved, radiofrequency (RF) ablation has become an effective therapy for AFL in this group of patients (3).
The re-entrant circuit of AFL in patients after cardiac surgery is well-characterized (1,4–7). The flutter circuit may be “atypical” in which the critical isthmus is located in a region bound by the scar (most commonly due to previous right lateral atriotomy) and another anatomic structure, such as the tricuspid annulus (1). The circuit may also be “typical” with a critical zone of conduction in the classic isthmus bound by the tricuspid valve, eustachian ridge, and inferior vena cava—the cavotricuspid isthmus (CTI) (5). The coexistence of both mechanisms in a single patient has also been reported, sometimes involving a figure-eight activation pattern of the two loops simultaneously (4,8).
When ablating either of these flutters, most reports have looked at employing a single ablation line across the appropriate critical zone of conduction (1,3,5,9–12). For typical CTI-dependent AFL, an ablation line is done along the tricuspid valve-eustachian ridge isthmus or the tricuspid valve-inferior vena cava isthmus (13). For right lateral atriotomy scar-related AFL, the ablation line is done from the scar to the tricuspid annulus or the inferior vena cava (1). When employing this strategy, however, there can be a significant rate of recurrence of postoperative AFL as high as 30% to 40% over 1.5 to 2 years (10).
In patients with postcardiac surgery AFL, it is unknown whether a strategy of targeting every potential macro–re-entrant circuit would reduce long-term flutter recurrence. We identified a unique population of patients with postcardiac surgery AFL who had a single right atrial incisional scar secondary to retrograde cardioplegia infusion via the coronary sinus (CS) (14). In these patients, we sought to assess if ablation of the clinical AFL only versus ablation of every possible re-entrant circuit would achieve a higher long-term success rate.
This study was designed as a prospective cohort study.
Between December 2000 and January 2002, 204 patients presented for mapping and ablation of AFL to the electrophysiology laboratory of the Cleveland Clinic Foundation. Of this group, a prospective cohort was identified consisting of patients who fulfilled the following criteria at the time of electrophysiologic study: 1) a history of coronary artery bypass grafting or single valve surgery; 2) a single right atrial free wall incisional scar as defined by voltage <0.5 mV identified by the CARTO mapping system (Biosense-Webster, Diamond Bar, California); 3) symptomatic CTI-dependent AFL and/or AFL rotating around the defined right atrial free wall scar; and 4) flutter onset >6 months after cardiac surgery. Patients with more than one uniform scar, and/or microchannels within the uniform scar were excluded from study. A total of 29 of 204 patients (14%) were identified. These patients were unique compared with most postcardiac surgery patients because they underwent cardiopulmonary bypass (CPB) using a well-described technique of retrograde infusion of cardioplegic solution via the CS (14). This technique involves clamping the lower, lateral right atrium where a mattress suture is deployed and a small incision (up to 3 cm) is created for passage of a 14-F cannula into the CS. An increased risk of AFL has been reported in this type of patient previously (15). The scar is likely a result of the combination of incision and pressure from clamping. All patients gave informed consent for the procedures. The procedures followed were in accordance with institutional guidelines.
Mapping and ablation
All antiarrhythmic agents were discontinued for ≥72 h before performing electrophysiologic study and ablation.
All patients underwent standard electrophysiologic testing with single-plane, mobile fluoroscopy. Electrode catheters were placed in the CS, right ventricular apex, and His bundle positions. A duodecapolar catheter was also positioned in the right atrium encircling the tricuspid annulus anterior to the crista terminalis with the distal end at the CS os. Intracardiac electrograms as well as simultaneous electrocardiogram leads were recorded using a commercial amplifier system (CardioLab, Pruka Engineering Inc., Milwaukee, Wisconsin). Three-dimensional electroanatomic mapping was also performed in all patients using the specialized catheter-based CARTO electromagnetic mapping system (Biosense Webster Inc.).
If the patient presented in normal sinus rhythm, atrial programmed and burst pacing was performed to induce tachycardia. Programmed pacing included up to three atrial extrastimuli after a paced drive train of eight beats at two different cycle lengths. Burst pacing in the atrium was performed to a minimum cycle length of 250 ms or to the point of atrial refractoriness, whichever occurred first. Infusions of isoproterenol were not required for induction in any of the patients.
Entrainment mapping was performed, as previously described (1,16). First, we tried to confirm or exclude the CTI as the protected zone of slow conduction by trying to demonstrate concealed entrainment from the isthmus (see the Definitions section). If the CTI was identified as the critical zone, RF energy was delivered between the tricuspid annulus and the Eustachian ridge and continued until there was at least 90% reduction of electrogram amplitude along the ablation line. Once this end point was reached, bidirectional block was demonstrated during both proximal CS pacing and low-lateral right atrial pacing using the duodecapolar catheter. If block was not definitively demonstrated using this technique, isthmus conduction block was proven using the CARTO mapping system (Biosense Webster Inc.). If necessary, further ablation was performed to achieve these end points.
In the patients where the CTI was not confirmed as the protected zone of slow conduction, incisional AFL was confirmed using the CARTO electroanatomic mapping system (Biosense Webster Inc.) and entrainment mapping as previously described (17). Radiofrequency energy was delivered to the narrowest and/or most accessible part of the circuit, which was typically between the scar and the tricuspid annulus or inferior vena cava. Ablation was continued until there was at least 90% reduction of electrogram amplitude along the ablation line. Block across the line was demonstrated using the CARTO mapping system (Biosense Webster Inc.).
After ablation, inducibility of AFL was assessed using extrastimulus testing and burst pacing as described in the preceding text. In patients with both CTI and scar circuits, inducibility was assessed after ablation of each circuit.
Concealed entrainment from a critical isthmus was considered if pacing from that specific site at 10 to 20 ms faster than the tachycardia cycle length and at double the pacing threshold resulted in: 1) concealed fusion with no change in surface flutter wave morphology and endocardial activation sequence; and 2) a postpacing interval equal to the tachycardia cycle length. Slow conduction was defined as complex and fractionated activity of >50 ms duration. Zones of block were defined by double potentials separated by an isoelectric interval of ≥100 ms.
Patients were followed in the outpatient clinic at 2, 6, and 12 months after ablation. At each visit, they were asked about symptoms of recurrent AFL and were encouraged to notify the clinic in case of recurrence at any time during the follow-up period; 48-h Holter recording was done routinely in all patients immediately after the procedure and at the 2- and 6-month follow-up visit. In patients with symptoms suggestive of arrhythmia, continuous loop recorder monitoring was performed. Patients with recurrent AFL were offered a repeat ablation.
All results are expressed as mean ± SD.
The baseline characteristics and the flutter characteristics of the CTI-dependent and scar-dependent flutter groups are detailed in Table 1.
Isthmus-dependent AFL ablation
Fourteen patients underwent a typical isthmus line ablation for CTI-dependent AFL. The AFL was terminated successfully in seven (50%) of these patients. In the remaining seven (50%) patients, there was either an immediate shift to an atypical scar-dependent AFL immediately after ablation (n = 2), or there was inducible AFL around the scar line at the end of the procedure (n = 5) (Fig. 1). These seven patients then underwent a second ablation connecting the scar to the tricuspid annulus (n = 3) or the inferior vena cava (n = 4) with no further immediate recurrence at the time of the procedure.
Incisional scar-dependent AFL ablation
Fifteen patients underwent ablation for incisional scar-dependent flutter. In this group, 9 of 15 (60%) patients had either an immediate shift to a CTI-AFL immediately after ablation (n = 7) or had inducible typical AFL at the end of the procedure (n = 2). These nine patients then underwent a second isthmus line ablation with no other inducible arrhythmias at the time of the procedure (Fig. 2).
Patients were followed-up for 18 ± 6 months and 24 ± 5 months in the CTI- and incisional scar-dependent groups, respectively.
In the CTI-dependent AFL group, three of seven patients (42%) who initially received only one ablation line developed new scar-dependent AFL at an average of 5 ± 3 months after the initial ablation. These three patients subsequently underwent a second procedure to perform a flutter line ablation connecting the scar to the tricuspid annulus or the inferior vena cava. The seven patients who initially received two ablation lines at the initial procedure had no AFL recurrence at follow-up.
In the incisional scar-dependent group, five of six patients (83%) who initially received only one ablation line developed new typical CTI flutter at an average of 3 ± 1 months after the initial ablation. These patients subsequently underwent a second isthmus line ablation. In the nine patients who received two ablation lines at the initial procedure, only one had recurrence of typical CTI flutter, which was successfully re-ablated.
In both the CTI-dependent and scar-dependent patient groups, the patients who subsequently underwent a second ablation procedure during follow-up did not have any further recurrence.
In the patients who did well initially with only a single ablation line (n = 5), analysis of the electroanatomic map showed that, before the ablation, the lateral scar extended down to the inferior vena cava.
The mean cycle lengths of the recurrent flutters were not significantly different from the cycle lengths of the flutters described in Table 1.
The main finding of this study is that patients who have a single right lateral incisional scar and present with either scar-related or typical CTI-dependent AFL are at risk of developing either type of AFL regardless of their presenting arrhythmia. The presence of viable myocardium between the scar and the inferior vena cava appeared to predict induction and clinical occurrence of multiple AFL circuits. In some cases, both types of AFL were documented during the ablation procedure, and two ablation lines were performed with no recurrences over follow-up. In those patients that only had ablation of the presenting AFL (scar or isthmus) at the initial procedure, recurrence of AFL was common and appeared to involve the other circuit not initially ablated. Only after ablating the other critical isthmus did the patients avoid AFL recurrence over the follow-up period. Thus, optimal ablative treatment of this specific population of AFL patients may require initial treatment of all potential AFL circuits, regardless of the type of presenting AFL. This is the first study to compare outcomes in patients who present with classic CTI-dependent versus scar-dependent AFL in a postoperative setting.
Mechanisms of postincisional AFL
The mechanisms of re-entry circuits involved in postatriotomy AFL have been well-characterized. Early reports (1,18)identified the existence of protected zones of conduction bounded by the incisional scar and other natural barriers in the right atrium (such as the tricuspid annulus or inferior vena cava) that were critical portions of reentry circuits that allowed for AFL perpetuation. However, it was subsequently recognized that patients with incisional atrial scars were also predisposed to developing classic CTI-dependent AFL, which could be proven by means of entrainment mapping (5). In a cohort of postatriotomy patients reported by Akar et al. (4), 37% of patients had AFL due to incisional re-entry alone, while 19% had AFL that was CTI-dependent. Interestingly, the same study demonstrated that the majority of patients (44%) actually had coexistence of both circuits of AFL. This was in contrast with earlier reports that suggested that the coexistence of two AFL circuits was relatively rare (11). In our study, 9 of 29 patients (31%) had AFL due to both mechanisms at the time of initial electrophysiologic study, which supports the concept that coexistence is common. Although these circuits may be related in a figure-eight pattern of activation, it is known that elimination of one circuit will not necessarily interrupt conduction along the other, even if an ablation line is created along the common isthmus between the two circuits in the figure-eight (8). Thus, in at least some patients, conditions exist that are able to maintain different types of AFL.
One of the unique findings of our study, however, is that several patients may have coexistence of both AFL circuits, which is not demonstrable at the time of initial electrophysiologic assessment; 8 of our 29 patients (27%) only had one type of AFL inducible at the initial study, but subsequently had recurrence of AFL of the other type over the course of follow-up. This suggests that with conventional electrophysiologic assessment, it may be very difficult to predict future AFL recurrence and the culprit circuit. Other clinical and surface electrocardiographic data also have shown a limited ability to predict which AFL circuit will cause future arrhythmia (4,19).
Targeting ablation sites for postincisional AFL
Most of the literature to date has reported on ablation of a specific single site identified by entrainment mapping to eliminate incisional AFL (1,5). In our study, however, 8 of 13 patients (61%) who underwent only one ablation line at the initial study had AFL recurrence. This is higher than, but not inconsistent with, recurrence rates of 27% to 52% previously reported over intermediate and long-term follow-ups (3,20). In contrast, we have seen a negligible recurrence in patients who initially underwent two ablation lines. Given the difficulties discussed earlier in predicting patients who will develop future AFL involving different sites in the right atrium, it would seem that a strategy geared to empirically ablating both the classic isthmus and other incisional-related critical sites would be preferred. Such an approach has been hypothesized by other authors (6,9,21)but has never been assessed prospectively.
Our patient population was very selective, because patients had previous bypass or single-valve surgery and possessed a single right atrial incisional scar. However, an empiric approach of ablating all potential AFL circuits, regardless of the presenting AFL, could be extended to other groups of patients with “incisional” AFL, as has been suggested by others (7). In patients after congenital heart surgery, for example, large right atriotomy scars and other smaller scars in both the right and left atrium have been described (9,21). Depending on the number and position of scars in an individual patient, an attempt to identify and ablate as many potential isthmuses as possible during the initial procedure could maximize long-term success.
The population studied in this cohort is very unique and, therefore, limits the scope of our conclusions. Specifically, though scar-related AFL has been reported in some patients after bypass or valve surgery, it is not common (9). That is because right atrial wall incisions are not routinely performed with most techniques of CPB cannulation. Furthermore, the right atrial incisions for CPB are much smaller than atriotomies performed for congenital heart surgery; thus, it is likely that the incidence for incisional AFL will be lower in these particular postoperative patients. Therefore, our findings may have less relevance to most routine postcardiac surgery cardiac patients. Furthermore, it is unclear if an empiric approach using multiple ablation lines for AFL would be equally successful in patients who have multiple regions of scar, such as those after congenital heart disease repair. The number of lines required to effectively isolate all scarred regions in patients with multiple scars could be numerous and cumbersome. Although our conclusions raise some intriguing observations about the effectiveness of an empiric ablative approach to postatriotomy AFL, a randomized controlled trial would be required to directly compare the efficacy of the empiric versus targeted strategy. Selection bias of patients is a limitation in any cohort study, but this was hopefully limited by the fact that we studied consecutive patients in a prospective manner. Finally, the sample size used in this study was small. A larger sample size would help to define more precisely the outcomes in this patient population.
Patients with a single right atrial incisional scar are predisposed to both scar-related and CTI-dependent AFL. Identification of either a CTI-dependent or scar-related AFL and creation of one ablation line still resulted in recurrence of AFL from the other site. Placement of both a CTI linear lesion and a lesion connecting the scar to the tricuspid annulus or the inferior vena cava may minimize AFL recurrence versus a single ablation line approach.
☆ Dr. Verma is supported by a fellowship award from the Heart and Stroke Foundation of Canada.
- atrial flutter
- cardiopulmonary bypass
- coronary sinus
- cavotricuspid isthmus
- Received January 12, 2004.
- Revision received April 13, 2004.
- Accepted April 18, 2004.
- American College of Cardiology Foundation
- Kalman J.M.,
- VanHare G.F.,
- Olgin J.E.,
- Saxon L.A.,
- Stark S.I.,
- Lesh M.D.
- Garson A. Jr..,
- Bink-Boelkens M.,
- Hesslein P.S.,
- et al.
- Triedman J.K.,
- Saul J.P.,
- Weindling S.N.,
- Walsh E.P.
- Akar J.G.,
- Kok L.C.,
- Haines D.E.,
- DiMarco J.P.,
- Mounsey J.P.
- Chan D.P.,
- Van Hare G.F.,
- Mackall J.A.,
- Carlson M.M.,
- Waldo A.L.
- Nakagawa H.,
- Shah N.,
- Matsudaira K.,
- et al.
- Shah D.,
- Jais P.,
- Takahashi A.,
- et al.
- Anne W.,
- van Rensburg H.,
- Adams J.,
- Ector H.,
- Van de Werf F.,
- Heidbuchel H.
- Betts T.R.,
- Roberts P.R.,
- Allen S.A.,
- et al.
- Van Hare G.F.,
- Lesh M.D.,
- Stanger P.
- Kalman J.M.,
- Olgin J.E.,
- Saxon L.A.,
- Lee R.J.,
- Scheinman M.M.,
- Lesh M.D.
- Triedman J.K.,
- Alexander M.E.,
- Berul C.I.,
- Bevilacqua L.M.,
- Walsh E.P.
- Love B.A.,
- Collins K.K.,
- Walsh E.P.,
- Triedman J.K.