|
|
||||||||
Ann Thorac Surg 2003;75:1194-1199
© 2003 The Society of Thoracic Surgeons
a Cardiological and Cardiosurgical Department, Niguarda Hospital, Milan, Italy
b Cardiological and Cardiosurgical Department, Ospedali Riuniti, Bergamo, Italy
c Cardiological and Cardiosurgical Department, Ospedale G. Pasquinucci, Massa, Italy
Accepted for publication October 14, 2002.
* Address reprint requests to Dr Vignati, Via Ovada 43, 20142 Milan, Italy
e-mail: famvi{at}iol.it
| Abstract |
|---|
|
|
|---|
METHODS: We evaluated the results of intraoperative ablation in a group of 23 patients with chronic supraventricular tachyarrhythmias (mostly intraatrial reentry) that were unresponsive to conventional medical therapy. All procedures were performed consecutively between September 1999 and November 2001. Ablation was done during redo operations (Fontan conversion to total cavopulmonary connection in 16 patients) in 18 patients and during primary surgical correction in 5 patients. The mean age at operation was 25 ± 12 years (2 to 50 years). Cryoablation was done in 10 patients and radiofrequency ablation in 13 patients. Nineteen patients had ablation in one atrium and 4 had ablation in both atria. A generous atrial reduction was always performed at the end of the operation.
RESULTS: The operative mortality rate was 13% (3 patients) from causes unrelated to ablation. In 20 survivors, the ablation was effective immediately. Eight patients required a permanent pacemaker. During a mean follow-up of 22 ± 4 months, atrial arrhythmias recurred in 25% (5 patients) and were controlled with medical therapy, whereas 1 patient required pacemaker implantation.
CONCLUSIONS: Intraoperative treatment of unresponsive atrial tachyarrhythmias associated with operated or unoperated congenital heart disease is feasible and the midterm results are encouraging.
| Introduction |
|---|
|
|
|---|
The surgical treatment of atrial arrhythmias has usually been for the therapy of atrial fibrillation during open heart operations for valvular or ischemic heart disease [57] rather than for congenital heart defects [9]. More recently, intraoperative ablative techniques have been used, with promising results, for the treatment of atrial tachycardia during conversion of a failing Fontan operation to a total extracardiac cavopulmonary connection [1011].
In this study we report our preliminary experience with intraoperative ablation of atrial tachyarrhythmias in association with repairs for congenital heart disease.
| Patients and methods |
|---|
|
|
|---|
Eighteen patients had previous open heart surgery: 16 patients had a Fontan operation, 1 patient had correction of AV canal, and 1 patient had septation of common atrium. Eight patients were in New York Heart Association functional class III (7 patients with a Fontan operation and 1 patient with severe pulmonary stenosis), the remaining 15 patients were in New York Heart Association functional class II. Preoperative ejection fraction was normal in all but 3 patients with a previous Fontan operation (ejection fractions of 50%, 38%, and 27%).
Twenty-one of 23 patients had supraventricular arrhythmias. Supraventricular arrhythmias were classified according to surface electrocardiographic features. Intraatrial reentry tachycardia was defined as rapid atrial rhythm that was independent of AV conduction, with sudden onset and termination, and constant cycle length. The P wave had an abnormal axis or morphology during tachycardia. Common atrial flutter was defined as rapid atrial rhythm with constant cycle length and variable AV conduction. Typical atrial flutter was identified by the presence of characteristic sawtooth flutter waves in leads II, III, and AVF, with an isoelectric-positive pattern in V1. Atrial fibrillation was defined as irregular, disordered, and desynchronized atrial activation characterized by high-frequency f waves without a true isoelectric base line, with very irregular AV conduction. According to these definitions, 17 patients had intraatrial reentry tachycardia, 2 patients had atrial flutter, and 2 patients had atrial fibrillation. Table 1 summarizes the types of arrhythmia in relation to the type of congenital heart disease. Intraatrial reentry tachycardia was found almost exclusively in patients who had previous open heart operation, with the one exception being a patient with unoperated double inlet left ventricle, mitral incompetence, and pulmonary stenosis.
|
Operative technique
Primary surgical correction was performed in 5 patients (closure of an atrial septal defect, mitral valve repair, tricuspid valve repair, bidirectional Glenn anastomosis with relief of severe pulmonary stenosis, and bidirectional Glenn anastomosis with repair of a severely incompetent mitral valve). Eighteen patients had a redo operation (Fontan conversion to total extracardiac cavopulmonary connection in 16 patients, mitral valve replacement in 1 patient, and closure of a residual atrial septal defect in 1 patient). The mean time interval between first and second surgical procedures was 9 ± 4 years. All operations were done through a median sternotomy using total cardiopulmonary bypass, mild systemic hypothermia, and current cardioplegic techniques.
The right atrium was always opened along the crista terminalis, extending the incisions into both venae cavae (Fig 1). Intraoperative ablation was done before correction of anomalies at the level of the atria or the AV valves. Two patients without arrhythmias before operation received ablation because of important dilatation of the atrial chambers. One had a failing Fontan conversion, and the other had severe mitral regurgitation from a dysplastic valve.
|
The ablation lines are shown in Figures 13. Lines of ablation were chosen in accordance with the type of arrhythmia. In common atrial flutter, a linear lesion connecting the inferior vena cava and coronary sinus os to the tricuspid valve was performed (Fig 1). For intraatrial reentry tachycardia, a compartmentalization of the right atrium was performed (Fig 2). Atrial fibrillation was treated with isolation of the four pulmonary veins and/or compartmentalization of the right atrium (Figs 2 and 3). In 19 patients, the lesions were placed in only one atrium (right atrium, 18 patients; left atrium, 1 patient). Four patients had a biatrial approach (Table 1). The decision to treat one or both atria depended on the dimension of the atria and the type of arrhythmia.
|
|
Follow-up
Clinical status was assessed on the basis of the New York Heart Association functional class. Assessment of arrhythmia included review of clinical history, electrocardiograms, and 24-hour Holter monitoring at 3-, 6-, and 12-month intervals after operation and then every 6 months. The first 10 patients, all of whom had Fontan conversion, were evaluated with tranesophageal atrial stimulation 6 months postoperatively. All patients were given antiarrhythmic medication postoperatively. Amiodarone or propafenone were started immediately after the operation and were continued until the third or sixth postoperative month.
| Results |
|---|
|
|
|---|
Effect of ablation on rhythm
All 23 patients left the operating room with sinus rhythm or atrial pacing. Eight patients had implantation of a permanent pacemaker before discharge (7 patients for sinus node dysfunction and 1 patient for complete AV block). All patients with sinus node dysfunction had undergone a Fontan conversion, whereas the patient with complete AV block had a mitral valve replacement. At discharge, 16 of 20 patients were in sinus rhythm, and 4 had a stable pacemaker rhythm.
The mean duration of follow up in the 20 survivors was 22 ± 4 months (range 6 to 32 months). All patients were in New York Heart Association functional class I or II. Five patients (25%) had recurrence of atrial tachycardia after a mean interval of 6 months. Four of these 5 patients had a Fontan conversion, whereas the fifth patient had closure of a residual atrial septal defect. Three of the five patients had radiofrequency ablation and two had cryoablation. The arrhythmias responded to conventional medical treatment in all 5 patients. After institution of antiarrhythmic treatment, there were no recurrences. All these patients responded to single-drug therapy. Before operation, various combinations of drugs had been ineffective. At the latest follow-up, only the 5 patients with recurrences were taking antiarrhythmic drugs.
One patient with a previous Fontan conversion required pacemaker implantation 2 years postoperatively because of sinus node dysfunction. Ten patients were evaluated by using transesophageal atrial stimulation a mean of 6 months after Fontan conversion. It was not possible to induce atrial arrhythmias in any of these patients. Three patients had a prolonged corrected sinus node recovery time (mean value, 950 milliseconds; range 520 to 1,800 milliseconds).
| Comment |
|---|
|
|
|---|
Conventional medical treatment has limited efficacy in controlling atrial arrhythmias [14, 1618]. This is confirmed by our experience, as 91% of our patients had preoperative atrial arrhythmias that were poorly controlled by medical therapy. These observations make it evident that treatments other than pharmacologic treatment and percutaneous ablation should be explored.
Surgical treatment of atrial arrhythmias appears to be a promising technique. The efficacy of the maze procedure and its modifications is well known, not only for treatment of atrial fibrillation and flutter associated with valvular and ischemic heart disesase [25, 8], but also for arrhythmias associated with atrial septal defects, as reported by Kobayashi and colleagues [9]. Mavroudis and associates [10, 11] reported the efficacy of cyoablation for the treatment of arrhythmias unresponsive to medical therapy associated with Fontan conversion to total cavopulmonary connection.
Our experience shows the feasibility of intraoperative ablation in the setting of different congenital heart diseases, with promising results. We had better results in patients who had primary surgical correction of their heart disease, whereas after redo operations the midterm efficacy decreased to 67%. These different results are probably related to the type of arrhythmia treated. In the group with primary correction, atrial fibrillation or common atrial flutter were the most common arrhythmias, whereas patients who had redo operations mostly had intraatrial reentry tachycardia.
Ablative procedures for common atrial flutter and atrial fibrillation are now relatively well standardized [1923] because the mechanism of the origin of these arrhythmias is well known. Intraatrial reentry tachycardia is a complex arrhythmia that occurs frequently in patients who have had prior operation for congenital heart disease [1418]. The anatomic substrates that might support intraatrial tachycardia are different. Scars, patches, zones of fibrosis related to atrial dilatation, and the natural barriers, such as AV valves and the orifices of the venae cavae and the pulmonary veins limit the area of atrial wall in which reentrant circuits can develop [1719]. These data allow us to speculate that successful ablation of a dominant circuit in these arrhythmia-prone atria might allow subsequent emergence of new arrhythmic circuits that are critically dependent on other areas of atrial myocardium [1618]. This is the reason why successful ablation of intraatrial reentry tachycardia presents a considerable challenge, even if a compartmentalization of the atria is performed. This is supported by our experience. We observed recurrences in 25% of patients treated for intraatrial reentry tachycardia. Four of 5 of those patients had a Fontan conversion. None had atrial fibrillation before the second operation, and the recurrences were atrial tachycardia.
This relatively high rate of recurrence in patients who had redo Fontan conversion is related to several factors. First, these patients have a powerful arrhythmogenic substrate. Second, we limited the ablation to the right atrium. Some authors recently advised treatment of both atria [24]. Because 3 of 4 of our patients had tricuspid atresia, it is possible that the ablation of the cavotricuspid isthmus might have been incomplete. In tricuspid atresia, the right atrium has a muscular floor at the atretic AV junction, without a true orifice of the tricuspid valve. Accordingly, a reentry circuit is unlikely to be ablated until the maneuver is extended to the point of the dimple where atrial musculature is lacking [24]. It might be the case that in some patients we did not extend the ablation completely to the dimple of atretic valve, to avoid damaging the AV node. The location of the AV node is not always clear in the setting of tricuspid atresia, particularly when the right atrium is very enlarged. The fifth patient with recurrence had a common atrium, and the ablation lines were placed on both sides of the atrial chamber. The particular anatomic arrangement in this situation might explain the difficulty encountered in performing a complete compartmentalization of the atrial cavity. The importance of anatomic substrate in explaining recurrences was stressed recently by Mavroudis and colleagues [11]. Third, recurrences might be related to the presence of vital tissue among the lesions produced with ablation. This condition, according to previous reports [725], is more common with radiofrequency ablation, which produces less homogeneous lesions, than it is with cryoablation. Three of our 5 patients with recurrence had radiofrequency ablation. However, the ablative procedure in these patients with recurrences should not be considered completely ineffective, because the arrhythmias of these patients were completely controlled by single-drug treatment. This is in contrast to the preoperative status, in which none of the conventional treatments was effective.
We have adopted three different types of ablation according to the type of arrhythmia present. For common atrial flutter we used a less aggressive approach, because it is well known that this arrhythmia is related to a macroreentry circuit constrained in the right atrium between two natural barriers [1921]. These barriers are the annulus of the tricuspid valve and the orifice of the inferior vena cava. It is well documented that ablation at the muscular cavotricuspid isthmus abolishes the reentrant circuit of atrial flutter [2, 1920]. In the presence of intraatrial reentry tachycardia, we performed a compartmentalization of the right atrium according to the suggestions of Mavroudis and colleagues [10, 11]. As we have stated previously, this arrhythmia has complex features [14, 16, 17], and the ablative treatment should be as extensive as possible. Because intraatrial tachycardia could be the consequence of either a macroreentry circuit or a microreentry circuit [16, 18], an extensive resection of right atrial wall seems most likely to reduce the number of microreentry circuits.
In the past several years, various treatments of atrial fibrillation have been proposed [15, 7, 26]. Evolution in the treatment of this arrhythmia has resulted from new insights into its pathogenesis. It is well known that atrial fibrillation is based on multiple reentrant wavelets occurring in a random order in the atrium. With the maze procedure, several linear incisions are made in both atria so that multiple reentrant wavelets are no longer possible [25]. More recently, it has been shown that atrial fibrillation has a rapidly firing focal source, usually in the pulmonary veins [23, 25]. Based on this observation, surgical isolation of the pulmonary veins or percutaneous radiofrequency ablation of pulmonary foci has been performed with encouraging results [7, 23, 25]. We treated atrial fibrillation with pulmonary vein isolation, as suggested by Gaita and associates [7], when there was left atrial dilatation. We performed compartmentalization of right atrium when this chamber was enlarged [2, 9]. We adopted a conservative approach for all patients, limiting the ablative treatment to the dilated atrium, when possible. We adopted this policy to try to maintain the contribution of atrial contraction to the hemodynamic performance of the ventricles.
Ablative procedures were planned without preoperative invasive electrophysiologic evaluation. The increased understanding of atrial arrhythmias obtained with percutaneous ablative procedures has improved our understanding of the correlations between the electrocardiographic patterns and the anatomic substrate of arrhythmias. This has reduced the number of preoperative diagnostic electrophysiologic studies [19, 20]. Preoperative electrophysiologic study may be helpful for evaluation of intraatrial reeentry tachycardia, because these patients can have multiple sites of reeentry [1618]. The precise identification of the circuits responsible for intraatrial tachycardia is required only in patients who have percutaneous ablative procedures. When a surgical ablative procedure is planned, every effort should be made to ablate all possible reentrant circuits, so a preoperative electrophysiologic evaluation is not so essential.
In the present study we used two different techniques of ablationcryoablation at the beginning of our experience and radiofrequency ablation when it became available to us. Some studies have noted the superiority of cyoablation over radiofrequency ablation [7, 25]. Cryoablation has the advantage of leaving the endocardium intact, thus reducing thromboembolic risk. It also has the advantage of producing lesions that are transmural and well circumscribed [25]. We did not observe significant differences between the two techniques. However, radiofrequency was easier and faster than cryoablation. These characteristics are important when considering the length of the operation.
The mortality rate in this study is on the high side for this type of surgical treatment [11, 24]. Death occurred only in patients who had Fontan conversions. Two of the 3 patients died of low cardiac output syndrome, and both had a low preoperative ejection fraction. This raises the question of whether Fontan conversion is the best choice for patients with poor ventricular function. Probably, in this particular subset of patients, cardiac trasplantation is the correct choice.
The most important complication of intraoperative ablation was sinus node dysfunction. This complication was limited to patients in whom a redo operation was performed. The causes of sinus node dysfunction were probably related, in part, to the extensive atrial wall reduction and, in part, to ablation lines placed near the sinus node. Sinus node dysfunction can present immediately after discontinuation of bypass or can occur later, as it did in 2 of our patients in whom sick sinus syndrome developed 1 and 24 months postoperatively. According to our results, an atrial wire should be inserted in every patient who has a Fontan conversion even if sinus rhythm is present at the end of operation. After cavopulmonary connection, it will not be possible to reach the heart through the venous system, and pacing can be performed only with an epicardial approach, which requires reoperation.
In conclusion, intraoperative treatment of arrhythmias associated with a primary surgical correction or with a redo operation for congenital heart disease is feasible, and the midterm results are encouraging. In our opinion, radiofrequency ablation is easier and faster than cryoablation. Ablation should be as extensive as possible and should address all known critical isthmi, especially in patients with Fontan conversion. Atrial wall reduction should be an integral part of surgical treatment of arrhythmias.
| Acknowledgments |
|---|
|
|
|---|
| References |
|---|
|
|
|---|
-year clinical experience with surgery for atrial fibrillation. Ann Thorac Surg 1996;224:267-273.
This article has been cited by other articles:
![]() |
M. Koh, T. Yagihara, H. Uemura, K. Kagisaki, I. Hagino, T. Ishizaka, and S. Kitamura Optimal timing of the Fontan conversion: Change in the P-wave characteristics precedes the onset of atrial tachyarrhythmias in patients with atriopulmonary connection J. Thorac. Cardiovasc. Surg., May 1, 2007; 133(5): 1295 - 1302. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| ANN THORAC SURG | ASIAN CARDIOVASC THORAC ANN | EUR J CARDIOTHORAC SURG |
| J THORAC CARDIOVASC SURG | ICVTS | ALL CTSNet JOURNALS |