Results of Maze Surgery for Lone Paroxysmal Atrial Fibrillation
Background—If drug refractoriness to paroxysmal atrial fibrillation (PAF) occurs, arrhythmia surgery that involves channelling and the exclusion of specific atrial areas can abolish atrial fibrillation. The purpose of this study was to establish the effectiveness and safety of maze III surgery to abolish PAF.
Methods and Results—Surgery was performed in 41 selected patients who had long-standing, symptomatic, drug-refractory, lone PAF. At discharge, 35 patients (85%) were arrhythmia free, and 6 patients (15%) showed PAF and paroxysmal atrial tachycardia. Death or stroke did not occur during a mean follow-up of 31±16 months. At the end of follow-up, 39 patients (95%) had no PAF; however, in 2 patients (5%), PAF persisted and eventually required His bundle ablation and pacing. Three months after surgery, nodal escape rhythm was observed in only 1 patient, whereas sick-sinus syndrome emerged late after surgery in 2 patients. Antiarrhythmic drugs were used in 20% of patients during follow-up. The quality of life improved markedly after surgery and remained unchanged afterward. Echocardiographic findings did not alter, but exercise capacity increased.
Conclusions—This pilot study demonstrates the effectiveness and safety of maze III surgery for lone PAF. In patients without sick-sinus syndrome, this intervention offers a sensible alternative to His bundle ablation and lifelong pacemaker dependency.
Antiarrhythmic drugs remain the first-line therapy for the prevention and rate control of paroxysmal atrial fibrillation (PAF). If drug refractoriness emerges, however, catheter His bundle ablation followed by cardiac pacing is often applied; this therapy shows satisfying results.1 2 3 4 However, pacemaker dependency and the requested continuation of drug therapy after His bundle ablation to maintain sinus rhythm are often disliked by patients. Arrhythmia surgery is an alternative, nonpharmacological therapy that avoids the drawbacks of catheter His bundle ablation.
Atrial fibrillation (AF) is caused by random reentry, and a critical number of circulating reentrant wavelets is necessary to perpetuate the arrhythmia,5 6 although sometimes rapidly firing atrial foci may be responsible for the onset of PAF.7 A reduction of the atrial mass and surface limits the number of wavelets. Therefore, AF cannot be initiated by atrial premature beats or it terminates easily. Surgical exclusion, fractionation, and channeling of atrial areas8 are tools to suppress AF resulting from random reentry and to preserve atrial contraction and filling and sinus node function. Various surgical concepts, including the corridor procedure,9 left atrial isolation,10 11 maze surgery,12 13 pulmonary button isolation,14 and the atrial compartment operation15 16 have been performed in selected patients. However, the long-term results of surgery for PAF in patients without structural cardiac disease are few13 17 18 because this surgery is usually done in conjunction with mitral valve surgery. This study reports our long-term experience with maze III surgery as a treatment for lone PAF.
Patients younger than 70 years of age who were suffering from symptomatic, drug-refractory PAF were offered surgery as an alternative treatment. PAF was defined as AF with spontaneous termination,19 although electrical cardioversion was needed at least once in most cases. Drug refractoriness was defined as the failure of ≥4 different antiarrhythmic drugs or withdrawal from use of the drugs because of side effects. Sick-sinus syndrome, permanent (“accepted”) AF, focal atrial tachycardias, or serious ventricular arrhythmias ruled out surgery. Patients with PAF who also showed structural cardiac disease or who had suspected tachycardiomyopathy were also excluded. Verbal consent for surgery was obtained after oral information about the surgery was given to the patient.
Preoperatively, the duration of PAF and the number of failed antiarrhythmic drugs were noted. Standard ECG, 2-channel Holter recordings, and bicycle stress testing with oxygen consumption were used to document PAF and to exclude sinus node disease and abnormal atrioventricular conduction. Two-dimensional and pulsed-wave Doppler echocardiographic studies were performed to measure cardiac chamber size and function and to exclude structural cardiac disease. Left and right heart catheterization and coronary angiography were performed to rule out cardiac abnormalities and to obtain information on the arterial supply of the sinus node area. Catheter mapping of AF for detecting a focal origin or classifying the arrhythmia was not done.
The technique of maze surgery has been reported in detail previously12 and can be summarized as follows. Initial dissection consists of extensive mobilization of both caval veins and the roof of the left atrium. After bicaval cannulation, total cardiopulmonary bypass is instituted, and the heart is arrested. The right and left atrium are incised (Figure 1⇓). After suturing all incisions, air is evacuated from the heart, and the patient is weaned from cardiopulmonary bypass. In addition to standard postoperative care, anticoagulant therapy with Coumadin is administered from the first postoperative day for up to 3 months after discharge. This is followed by treatment with aspirin.
Pairs of temporary epicardial wires were attached to the left and right atrium and the right ventricle for the diagnosis of postoperative arrhythmias.18 Before discharge, we attempted to induce AF using programmed electrical stimulation.
All patients were followed in our outpatient department, initially at 3- to 6-month intervals, and then annually after the first year. Patients were asked in particular about palpitations and symptoms of cerebrovascular accidents; daily lifestyle and exercise capacity were also questioned. Holter recordings and bicycle stress tests were performed 6 months after discharge and annually thereafter to detect AF, sinus node dysfunction, and abnormal atrioventricular conduction. Echocardiographic studies were done simultaneously.
Quality of Life
Quality of life was assessed before surgery and at 3 and 12 months after surgery using a Dutch translation of the 36-item Short-Form Health Survey questionnaire.20 This tool was supplemented with a question about perceptions of changes in health and a self-designed questionnaire that assessed cardiac symptoms, sleep, cognitive functioning, mental health, and social functioning. Because the range of scores varied for all subscales, scores were normalized to a scale ranging from 0 to 100; lower scores represent a lower quality of life.
The primary end point of maze surgery was permanent absence of AF. A second end point was preserved sinus node function. Chronotropic incompetence of the sinus node was defined as symptomatic sinus bradycardia at rest of <60 bpm, a requirement for pacemaker therapy, or an insufficient increase of sinus rate at stress testing, which was arbitrarily chosen as <80% of the predicted value according to sex and age criteria.21
All values are expressed as mean±SD. Hypothesis testing was computed using paired tests. For proportions, the McNemar test was used; for normally distributed data, the paired t test was used; and for non-normally distributed data, the Wilcoxon test was applied. P<0.05 was considered significant.
From June 1993 to July 1998, 41 patients with lone PAF underwent maze surgery. Table 1⇓ shows the preoperative clinical characteristics of this consecutive series. The common patient profile was a 50-year-old man who had PAF for 5 years.
After surgery, neither major complications nor death occurred in any patient. One patient had an immediate recurrence of AF (Figure 2⇑) and showed severe heart failure 1 week after surgery, which required high dosages of diuretics. Before discharge, programmed electrical stimulation was performed in 32 of the 41 patients (78%). AF could not be induced in 29 of these 32 patients (92%); during follow-up, 1 of these 29 patients temporarily showed PAF, and 1 patient showed paroxysmal atrial tachycardia (Figure 3⇑). In 3 of the 32 stimulated patients (8%), AF could be induced: only 1 of these 3 patients suffered intractable PAF requiring His bundle ablation afterward; the other 2 patients remained arrhythmia-free. Postoperative stimulation was not done in 9 patients: 1 patient refused the test, AF was present in 3 patients, and a loss of atrial capture was present in 5 patients. At discharge (Figure 4⇑), 35 of the 41 patients were free from PAF (85%); 4 of these 35 patients showed nodal escape rhythm due to incompetent sinus rhythm. Six of the 41 patients (15%) had PAF (Figure 2⇑), flutter, or atrial tachycardia, which lasted from seconds to a maximum of 4 hours, with spontaneous conversion to normal sinus rhythm. A total of 37 of the 41 patients (90%) were discharged with no antiarrhythmic drug therapy.
All patients were alive after a mean follow-up of 31±16 months. Neither neurological nor cardiac complications were observed. Only 1 patient complained of incessant sinus tachycardia22 ; this patient required β-blockers some months after surgery. The proportion of patients with sinus rhythm without atrial arrhythmias increased from 76% immediately after discharge to 92% at the 12-month follow-up; thereafter, this diminished to 85% (Figure 4⇑). After resumed sinus rhythm, the number of patients with nodal escape rhythm without AF diminished to 2 asymptomatic patients at 3 months and to 1 patient after 3 months. Exercise testing of this patient 34 months after surgery initiated sinus tachycardia with a peak rate of 154 bpm (96%) at 89% of maximal exercise. In 2 other patients, infrequent asymptomatic sinus arrests of >4 s were recorded 26 and 27 months after surgery, respectively; 1 patient required atrial pacing. The number of patients with atrial arrhythmias diminished 3 months after surgery (Figure 4⇑). Two patients with recurrent drug-refractory PAF postoperatively underwent His bundle ablation and pacemaker implantation 13 and 29 months after surgery, respectively (Figure 2⇑). The percentage of patients who did not take antiarrhythmic drugs to prevent atrial arrhythmias remained stable (80%).
The mean values of echocardiographic parameters did not clearly change after surgery, except for right atrial volume (Table 2⇓). The average preoperative left atrial volume was markedly larger in the 6 patients with PAF at discharge compared with that of the arrhythmia-free patients (62±17 and 49±11 mL, respectively, [P=0.03]), but other atrial measurements did not differ. Mitral and tricuspid valve damage was not detected after surgery. The mean left ventricular end-diastolic diameter remained unchanged.
No patient showed PAF during exercise testing after surgery. Exercise capacity, maximal workload, and peak oxygen consumption (V̇o2) significantly increased (Table 2⇑) after surgery.
Quality of Life
This parameter was assessed in 18 of 41 patients (34%). It significantly improved 3 months after surgery and stabilized thereafter (Table 3⇓).
This prospective study demonstrated very satisfying immediate and midterm results of maze surgery for lone PAF. Twelve months after surgery, PAF was abolished in >90% of patients, and sinus node function remained undisturbed in 97%. In addition, quality of life and exercise tolerance significantly improved, whereas atrial echocardiographic measurements did not show deleterious effects of surgery.
Atrial Arrhythmias After Surgery
It has been reported that PAF, atrial tachycardia, and flutter frequently emerge soon after maze surgery and continue for some months14 because of atrial edema, pericarditis, surgical trauma, and elevated levels of catecholamines, which inhibit the recovery of preoperatively abnormal atrial electrophysiological properties.14 22 Recovery from the surgical trauma takes 2 to 3 months, as evidenced by a spontaneous reduction of atrial arrhythmias, which was also observed in our study (Figure 4⇑). Left atrial size seems more strongly associated with a cure of AF in patients undergoing maze surgery in conjunction with other cardiac surgery than other factors, such as age, sex, and duration of AF.23 The left atrial volume was markedly larger in our patients with early recurrent PAF than in those without postoperative arrhythmias. Finally, scarring from the atrial incisions can result in de novo atrial tachycardia (Figure 3⇑), although unexcluded atrial foci should be considered as well.
Sinus Node Function After Surgery
In 4 of our 41 patients, a nodal escape rhythm was present at discharge; it persisted in only 1 patient. Temporary denervation or degeneration of sinus node cells due to atrial surgery is the supposed mechanism of this dysfunction,24 although an impairment by PAF cannot be excluded.25 The surgical trauma sometimes also causes a temporary reduction of the circadian variations of autonomic nerve activity24 26 or incessant sinus node tachycardia24 ; only 1 of our 41 patients transiently experienced incessant sinus tachycardia. The heart rate response to exercise can be reduced soon after maze surgery, but it usually normalizes within 1 year.24 27 An unchanged average maximal heart rate at exercise 12 months after surgery was observed in our patients (Table 2⇑). Although sick-sinus syndrome was ruled out before surgery, in 2 patients, longstanding sinus arrests emerged >2 years later. Our findings confirm the previously reported progression of atrial disease in AF as evidenced by sick-sinus syndrome.28 29
Atrial Function After Surgery
Preoperative left and right atrial contraction could only be observed in half of our patients (Table 2⇑), probably because of PAF.30 Twelve months after surgery, the number of patients with atrial contraction had not clearly increased. This finding suggests that surgery is detrimental to atrial contraction due to the multiple atrial incisions and sutures and the subsequent scarring, particularly of the left atrium,31 because contraction of the right atrium was more frequently seen than contraction of the left atrium.31 32 A second reason for this difference could be a diminished synchronization between the left atrium and ventricle due to the surgically created prolonged route of conduction from the right to left atrium.31 Maze surgery did not markedly reduce mean atrial volumes and diameters (Table 2⇑) or adversely affect mitral or tricuspid valve or ventricular function. Longitudinal studies of atrial function are needed to establish the risks of maze surgery on atrial contractility.31
In view of the large physical impact of surgery for PAF, long-term patient improvement is the crucial objective, but it was not an end point of this study. Although long-term arrhythmia monitoring was limited to regular Holter recordings and ECG at exercise, clinical experience shows that these patients with longstanding symptomatic PAF before surgery became very sensitive to palpitations. Interruption of their normal heart rhythm after surgery would be readily noticed, and they would ask for new examinations. The arrhythmia history of these patients can, therefore, provide reasonable insight into the long-term surgical result. At the end of follow-up, 95% of the patients had no PAF and 80% were drug-free; thus, it is not surprising that the quality of life, as documented in the patients who were operated on most recently (40%), dramatically improved after surgery and remained unchanged afterward. This was also true for the improvement in exercise tolerance.
Because atrial function and dimensions did not change (Table 2⇑), one can conclude that suppression of atrial arrhythmias is the crucial contribution to patient improvement.
Comparison With Other Surgical Methods
A technical obstacle of corridor surgery was the sometimes irreducible conduction between the left atrium and corridor compartment through the muscle tissue of the coronary sinus, which resulted in postoperative recurrences of PAF.17 18 Sometimes, this technical failure is also observed when applying left atrial isolation.10 11 In maze surgery, the isolation of the coronary sinus is done more laterally to the coronary os, where the amount of accompanying muscular tissue to be destroyed diminishes strongly and where conduction is more easily interrupted. Our maze surgery results with lone PAF clearly surpass our previous results with corridor surgery in 36 comparable patients.18
At 3.5 years of follow-up in the patients who had corridor surgery, the actuarial freedom of atrial arrhythmia without drugs was 72±9%, actuarial freedom of sinus node dysfunction was 81±7%, and chronic pacing was needed in 16%.17 18 Comparisons with the outcomes of surgery for left atrial isolation,10 11 the atrial compartment procedure,15 16 and pulmonary button isolation14 in patients with lone PAF cannot be made because these interventions were mostly done in conjunction with mitral valve surgery. In the accumulated maze surgery series (n=178) of Cox et al,13 32 58% of patients had lone AF and 66% underwent arrhythmia surgery only. The authors reported a combined recurrence of atrial flutter and AF of 2% in the 118 patients who had maze III surgery; they also reported blunted sinus node chronotropy in 6% of patients, and iatrogenic sinus node injury was absent in the 82 patients without preoperative sick-sinus node disease.
As discussed earlier, the true incidence of PAF recurrence could not be determined because Holter recordings were done at intervals; symptoms during follow-up were always a reason for further arrhythmia analysis. Event recorders can undoubtedly provide better information, but the devices must be activated by the patient and asymptomatic arrhythmia will escape recording. This study was designed to validate maze III surgery for PAF; therefore, these results cannot provide a recommendation for selection between focal catheter ablation, His bundle ablation, or surgery in patients with problematic PAF. A larger patient population with more variables (such as atrial size)33 34 35 and a longer follow-up time are needed to assess the contribution of this surgery to the suppression of AF and to prove its superiority over other methods.28 Finally, because PAF was only diagnosed by ECG, recommendations for surgery for specific types of PAF cannot be given.
Successful surgery for lone PAF involves abolishment of the arrhythmia, undisturbed sinus node function, no mortality, and a low morbidity risk. In addition, the intervention must be widely applicable, not cause a deterioration in hemodynamic parameters and, finally, diminish the thromboembolic risk of AF. The consequence is an improved quality of life after surgery. This pilot study showed that these goals can be achieved with maze III surgery provided that surgical experience with atrial arrhythmia is available and sick-sinus syndrome can be ruled out before surgery. Longitudinal evaluation of atrial contraction and filling and of sinus node function is needed to define the position of maze surgery in the management of lone PAF. Finally, more information on the mechanism and pattern of PAF is needed to optimize the selection of surgery for drug-refractory PAF.
The authors thank Gerda J.L. van der Kuijl for manuscript editing.
- Received August 6, 1999.
- Revision received October 18, 1999.
- Accepted November 5, 1999.
- Copyright © 2000 by American Heart Association
Rodriquez LM, Smeets KLRM, Baiyan X, de Chillou C, Cheriex E, Pieters F, Metzger J, den Dulk K, Wellens HJJ. Improvement in left ventricular function by ablation of atrioventricular nodal conduction in selected patients with lone atrial fibrillation. Am J Cardiol. 1993;72:1137–1141.
Buys EM, van Hemel NM, Kelder JC, Ascoop CAPL, Dessel PFHM, Bakema H, Kingma JH. Exercise capacity after His bundle ablation and rate responsive ventricular pacing for drug refractory chronic atrial fibrillation. Heart. 1997;77:238–241.
Lee S-H, Chen S-A, Tai C-T, Chiang C-E, Wen Z-C, Cheng J-J, Ding Y-A, Chang M-S. Comparisons of quality of life and cardiac performance after complete atrioventricular junction modification in patients with medically refractory atrial fibrillation. J Am Coll Cardiol. 1998;31:637–644.
Moe GK. On the multiple wavelet hypothesis of atrial fibrillation. Arch Int Pharmacodyn Ther. 1962;140:183–188.
Allessie MA, Lammers WJEP, Bonke FIM, Hollen J. Experimental evaluation of Moe’s multiple wavelets hypothesis of atrial fibrillation. In: Zipes DP, Jalife J, eds. Cardiac Electrophysiology and Arrhythmias. New York, NY: Grune and Stratton; 1985:265–75.
Jaïs P, Haïssaguerre M, Shah DC, Chouairi S, Gencel L, Hocini M, Clémenty J. A focal source of atrial fibrillation treated by discrete radiofrequency ablation. Circulation. 1997;95:572–576.
Guiraudon GM, Guiraudon CM, Klein GJ, Yee R. Atrial fibrillation in the future: the surgeon’s perspective. In: Murgatroyd FD, Camm AJ, eds. Nonpharmamacological Management of Atrial Fibrillation. Armonk, NY: Futura Publishing Co; 1997:215–237.
Guiraudon GM, Klein GJ, Guiraudon CM, Yee R. Treatment of atrial fibrillation: preservation of sinoventricular impulse conduction: the Corridor operation. In: Olsson SB, Allessie MA, Campbell RWF, eds. Atrial Fibrillation: Mechanisms and Therapeutic Strategies. Armonk, NY: Futura Publishing Co; 1994:349–371.
Vigano M, Graffigna A, Ressia L, Minzioni G, Pagani F, Aiello M, Gazzoli F. Surgery for atrial fibrillation. Eur J Cardiothorac Surg. 1996;10:490–497.
Melo JQ, Neves J, Adrago P, Ribeiras R, Ferreira MM, Bruges L. When and how to report results of surgery on atrial fibrillation. Eur J Cardiothorac Surg. 1997;12:739–745.
Lo HM, Lin FY, Lin JL, Tseng CD, Hsu KL, Chiang FT, Tseng YZ. Electrophysiological properties in patients undergoing atrial compartment operation for chronic atrial fibrillation with mitral valve disease. Eur Heart J. 1997;18:1805–1815.
van Hemel NM, Defauw JJAM, Kingma JH, Jaarsma W, Vermeulen FEE, de Bakker JMT, Guiraudon GM. Long-term results of the corridor operation for atrial fibrillation. Br Heart J. 1994;71:170–176.
Levy S, Breithardt G, Campbell RWF, Camm J, Daubert J-C, Allessie M, Aliot E, Capucci A, Cosio F, Crijns H, Jordaens L, Hauer RNW, Lombardi F, Luderitz B, on behalf of the Working Group on Arrhythmias of the European Society of Cardiology. Atrial fibrillation: current knowledge and recommendations for management. Eur Heart J. 1998;19:1294–1320.
Recommendations and standard guidelines for exercise testing. Eur Heart J. 1988;9(suppl K):3–37.
Tieleman RG, van Gelder IC, Crijns HJGM, de Kam P, van den Berg MP, Haaksma J, van der Woude HJ, Allessie MA. Early recurrences of atrial fibrillation after electrical cardioversion: a result of fibrillation-induced electrical remodeling of the atria? J Am Coll Cardiol. 1998;31:167–173.
Kamata J, Nakai K, Chiba N, Hosokawa S, Sato Y, Nasu M, Sasaki T, Kitahara H, Izumoto H, Yagi Y, Itoh C, Hiramori K, Kawazoe K. Electrocardiographic nature of restored sinus rhythm after Cox maze procedure in patients with chronic atrial fibrillation who also had other cardiac surgery. Heart. 1997;77:50–55.
Elvin A, Wylie K, Zipes DP. Pacing-induced chronic atrial fibrillation in pairs sinus node function in dogs. Circulation. 1996;94:953–960.
Tamai J, Kosakai Y, Yoshioka T, Ohnishi E, Takaki H, Okano Y, Kawashima Y. Delayed improvement in exercise capacity with restoration of sinoatrial node response in patients after combined treatment with surgical repair for organic heart disease and the maze procedure for atrial fibrillation. Circulation. 1995;91:2392–2399.
Feinberg MS, Waggoner AD, Kater KM, Cox JL, Lindsay BD, Pérez JE. Restoration of atrial function after the maze procedure for patients with atrial fibrillation: assessment by Doppler echocardiography. Circulation. 1994;90(part 2):II-285–II-92.