The Cox-Maze Procedure: The Cleveland Clinic Experience

The Cox-Maze Procedure: The Cleveland Clinic Experience Patrick M. McCarthy) A. Marc Gillinov) Lon Castle) Mina Chung) and Delos Cosgrove III The Cox-Maze procedure was designed to address the consequences of atrial fibrillation, tachycardia, hemodynamic impairment, and thromboembolism. From 1991 until June 1999, 100 patients underwent the Maze operation at the Cleveland Clinic Foundation. The group included 72 men with a mean age of 58 ± 11 years (range, 23 to 78 years). Initially, the Maze-I procedure was performed primarily for patients with lone atrial fibrillation. However, since 1995, the Maze-III procedure has been performed exclusively, and it is typically combined with mitral valve repair. Twenty-three patients had only a Maze procedure, 60 patients had the Maze procedure/mitral valve repair, 10 patients had Maze procedure/coronary artery bypass, 6 had Maze procedure/mitral valve replacement, and 1 had Maze procedure/atrial septal defect repair. Chronic atrial fibrillation was present in 78% of patients for a mean of 8 ± 9 years. There was a 1% perioperative mortality and 5% late mortality rate. Median hospital stay was 9 ± 5 days. Six patients required new early permanent pacemaker insertion. With a mean follow-up of 3 years, 90.4% of patients are in sinus rhythm (or atrial pacing). Preoperative symptoms were reduced: 24% had preoperative syncope; none had syncope in follow-up; 14% of patients preoperatively had cerebral or systemic emboli; and there were no perioperative or late embolic events. The Maze procedure effectively addressed the major complications of atrial fibrillation and was associated with low peri operative and late morbidity rates. Copyright © 2000 by IN.B. Saunders Company Key words: Atrial fibrillation, Maze procedure.

OX's Maze operation was designed to eliminate the 3 complications of atrial fibrillation: symptomatic tachycardia, hemodynamic impairment secondary to loss of atrial contractility, and embolization of atrial thrombus. 1•4 Our experience with the operation began in 1991, when Dr. Cox was a visiting professor at the Cleveland Clinic Foundation. Our initial experience was with the Maze-I and Maze-II procedures, but since 1995 we have used the Maze-III procedure with minor modifications. 5. 7 This report describes our experience with the operation, our philosophy regarding indications for the procedure, and long-term results.

C

Patients and Methods From January 1991 until June 1999, 100 patients underwent the Maze operation. The patients (Table

From the Department of Thoracic and Cardiovascular Surgery and the Department of Cardiolog, The Cleveland Clinic Foundation, Cleveland, Ohio. Address reprint requests to Patrick M. McCartky, MD, Department of Thoracic and Cardiovascular Surgery. 9500 Euclid Avenue, F25, Cleveland, Ohio 44195. Copyright © 2000 by WE. Saunders Company 1043-0679/00/1201·0004$10.00/0

1) ranged from 23 to 78 years old (mean, 58 ± 11 years), and 72% were men. Our initial experience was with the Maze-I operation for patients with lone atrial fibrillation. From February 1994 until March 1995, 10 patients had the Maze-II procedure.s Since April 1995, only the Maze-III operation has been used. 9 In all, 23 patients had only a Maze procedure, 60 patients had the Maze operation combined with mitral valve repair (including tricuspid valve repair in 13 patients and coronary artery bypass in 6), 6 patients had the Maze procedure performed with mitral valve replacement using a porcine valve (including 3 patients who had tricuspid valve repair), 10 patients had the Maze procedure combined with coronary artery bypass surgery only, and 1 patient had the Maze procedure combined with atrial septal defect repair. Sixteen patients had preoperative permanent pacemaker insertion after atrioventricular node ablation. Twenty-four patients had syncope preoperatively, 9 patients experienced a stroke or transient ischemic attack preoperatively, and 5 patients had peripheral emboli preoperatively. Patients with chronic atrial fibrillation (78%) had this for 5 months to 42 years (mean 8 ± 9 years). Paroxysmal atrial fibrillation was present in 22%.

Seminars in Thoracic and Cardiovascular Surgery', Vol 12, No 1 (january), 2000: pp 25-29

25

26

McCarthy et at

Table 1. Demographics of Patients Who Underwent Maze Procedure Only and Maze Procedure With Mitral Valve Repair

Age ChronicAF DurationAF Cross-clamp time Cardiopulmonary bypass time

MaQ Procedure Onfy

Maq Procedure/ MVRepair

48 ± l2yr 45% 7.2 ± 7yr 56 ± 16 min

59 ± 10 yr 95% 7.6 ± 8.7 yr 72 ± 16 min

114 ± 28 min

119 ± 24 min

Abbreviations: AF, atrial fibrillation; MV, mitral valve.

Preoperatively, the left atrial diameter for the patients who had lone atrial fibrillation or who were undergoing coronary artery bypass surgery were within normal limits or only slightly enlarged (Table 2). For those who underwent only a Maze operation, the mean left atrial diameter was 48 ± 10 mm (range, 35 to 64 mm). For those who underwent a Maze operation with coronary artery bypass surgery, the mean left atrial diameter was 55 ± 18 mm (range, 30 to 80 mm). Patients were selected for a Maze procedure combined with mitral valve surgery in part because of enlarged left atria. Therefore, the left atrial diameter for the patients undergoing the Maze procedure combined with mitral valve surgery was larger: 66 ± 14 mm (range, 34 to 99 mm). For patients who underwent the Maze procedure only, the aortic cross clamp time was 56 ± 16 minutes (range, 29 to 92 minutes). The total cardiopulmonary bypass time was 114 ± 28 minutes (range, 63 to 171 minutes). For patients undergoing the Maze procedure combined with mitral valve repair, the aortic cross clamp time was 72 ± 16 minutes (range, 40 to 117 minutes). The total cardiopulmonary bypass time was 119 ± 24 minutes (range, 75 to 180 minutes). Table 2. Echocardiographic Data on Patients Undergoing the Maze Procedure

Maze procedure only Maze procedurelMV repair Maze procedure/CAB

BifOre Operation

A.fter Operation

48 ± lOmm

47 ± 8mm

66 ± 14mm 55 ± 18mm

53 ± 3 mm 42 ± 2mm

Abbreviations: CAB, coronary artery bypass; MV, mitral valve.

Results All patients were discharged from the hospital. However, 1 of our early patients was readmitted after discharge and died within 30 days of surgery, so the total operative mortality was 1% (Table 3). This death occurred in a patient with a variation of hypertrophic cardiomyopathy (Yamaguchi's disease) who had a history of 2 previous strokes and pulmonary fibrosis from amiodarone therapy.5 The patient had been discharged but was readmitted with acute renal failure and died from cardiac arrest precipitated by hyperkalemia. Atrial size was reduced as measured by intraoperative echocardiography (Table 2). The median hospital stay was 9 ± 5 days. The incidence of postoperative atrial fibrillation, 33%, was similar to that of most cardiac surgery operations. Initially, this was treated with flecainide. Recently, however, procainamide or oral amiodarone were used with success. Six patients required early postoperative permanent pacemaker insertion for sick sinus syndrome (n = 5) and complete heart block (n = 1). Patients with preoperative pacemakers had upgrades to DDD-R mode after surgery. The troublesome postoperative fluid retention and accumulation we experienced early in this series resolved with the addition of spironalactone (75 mg, three times per day) for 6 weeks postoperatively.5 Also, more recently, we avoided right atrial appendage excision and instead incised through the right atrial appendage without removing the appendage. IO This may blunt the atrial natriuretic factor depletion reported in these patients. 11 ,12 In follow-up (mean, 3 years; range, 6 months to 9 years), there were 5 late deaths. One death was caused by sudden cardiac arrest out of hospital. One patient who had undergone the Maze procedure/ coronary artery bypass died after myocardial infarction, one patient died from congestive heart failure, one patient died from malignant glioblastoma, and one died from an unknown cause. Five patients Table 3. Clinical Outcomes Perioperatively and in Late Follow-up Mter the Maze Procedure 30-day mortality Late mortality Normal sinus rhythm Perioperative AF Perioperative pacemaker Late AF ablation Abbreviation: AV, atrioventricular node.

1% 5% 90.4% 33% 6% 5%

Cox-Maze: The Cleveland Clinic Experience

underwent late ablation for recurrent postoperative atrial flutter or fibrillation. All of the flutter re-entry circuits were located in the right atrium, and all patients were successfully returned to normal sinus rhythm. Of the 94 surviving patients, 85 are in sinus rhythm (90.4%). In addition to the freedom from atrial fibrillation, there was a significant reduction in symptoms and embolic events. Whereas 24% had preoperative syncope, there were no patients with postoperative syncope. Although 14% of patients had either cerebral or systemic embolism preoperatively, there were no perioperative or late postoperative embolic events (Figure 1).

Discussion This late follow-up of our Maze experience is very encouraging in 3 respects. First, perioperative morbidity and mortality were low. Although we initially had several patients develop significant fluid accumulation, this decreased considerably with the addition of spironalactone.5 Perioperative atrial fibrillation continues to be a nuisance, with a similar incidence during the Maze procedure as during other cardiac surgery operations. 5,13 Second, late success of the Maze operation is excellent. Over 90% of patients were in sinus rhythm, despite a long duration of atrial fibrillation in the majority of patients and enlarged atria in the patients with mitral valve disease. Finally, as was recently reported by Cox, we found a striking reduction in late embolic events.14 The reduction in embolic events is likely multifactoral. First, the left atrial appendage, which is a source of thromboemboli in patients with chronic atrial fibrillation, is excised. Second, the resumption of sinus rhythm is associated with atrial systole, albeit less vigorous than in normal atria. 15-17 Identifying and quantifying atrial systole is difficult. We have studied

14%

Poe, Syncope

Figure 1. Clinical events before and in follow-up after the

Maze procedure.

27

patients who received magnetic resonance imaging confirmation of atrial systole on the same day that echocardiography indicated that there was no atrial systole. 5,16 The excluded pulmonary veins are a denervated area of atria that will not contribute to atrial transport. A third reason for reduced embolic events may be the reduction in the size of the left atrium. 1O,17,18 In patients with enlarged atria, who were usually undergoing mitral valve surgery, we excised portions of the dome of the left atrium, the atrial free wall surrounding the left atrial appendage, and the inferior portion of the left atrium. 6,10 Intraoperative echocardiography in outpatients showed a significant reduction in left atrium size in the Maze procedure/mitral valve repair group (Table 2). Not only does this decrease the risk for re-entrant circuits that can occur between the suture lines, but the smaller atria may also have better atrial systole. 16,18 The long-term benefits in reduction of embolic events can be achieved with a low perioperative risk. At introduction, the widespread use of the Maze procedure was questioned because the long-term benefits of the operation were not yet known, and the risks of perioperative morbidity were not yet clear. 19 Also, most patients with recent atrial fibrillation who undergo mitral valve surgery will return to sinus rhythm without the Maze procedure. 19 Our indications for adding the Maze procedure for patients undergoing mitral valve surgery were for those patients who had chronic atrial fibrillation for more than 1 year and enlarged left atria. Occasionally, we applied the Maze procedure for patients with highly symptomatic atrial fibrillation of less than I-year duration who were drug intolerant. With these criteria to add the Maze procedure to mitral valve surgery, we find that the Maze procedure is not often required. The practice in the United States is to refer patients early for mitral valve surgery because of the low risk of mitral valve repair. Most patients today are in sinus rhythm when referred for surgery. If patients have mitral valve disease and develop atrial fibrillation, that is another time when the patient will be referred for surgery. Therefore, most patients with mitral disease and atrial fibrillation have a brief duration of the arrhythmia. In general, we have not added the Maze procedure to mitral valve surgery for such patients. In our opinion, those who most urgently need the Maze procedure are patients with previous embolic episodes despite the use of warfarin. Fourteen percent of the patients in this series had previous

28

McCarthy et at

embolic events despite adequate anticoagulation with warfarin. From Cox's report and this series, it is evident that the Maze procedure is extremely effective in decreasing embolic events in long-term followUp.14

The mechanisms underlying atrial fibrillation have recently become better defined. Under the traditional multiple wavelength hypothesis, atrial fibrillation is believed to be sustained by the propagation of multiple re-entrant circuits. 20,21 The continuously changing, wandering pathways are determined by local refractoriness, excitability, and conduction properties of the atrial tissue. This hypothesis was supported by electrophysiological mapping studies performed by Cox et al l -4 in humans that showed the presence of multiple re-entrant wavelets. The perpetuation of atrial fibrillation can depend on increasing atrial size. Structural enlargement can predispose the atria to the persistence of atrial fibrillation by allowing more re-entrant circuits to be sustained in the atria. These concepts formed the basis of Cox's Maze procedure and atrial reduction surgery, which limit re-entry by forming lines of block that limit re-entrant circuits and by reducing atrial size, which in turn limits the number of circuits that can be sustained to perpetuate atrial fibrillation. More recently, not only the substrate that predisposes the atria to the sustaining of fibrillation but the initiating triggers of atrial fibrillation have been highlighted as important in the pathogenesis of atrial fibrillation. Focal sources that initiate atrial fibrillation have been identified, particularly in patients with lone atrial fibrillation in the absence of structural heart disease. 22 These foci most commonly arise from the ostia of the pulmonary veins and have been the potential target of curative ablation. This mechanism may underlie the success of the Maze procedure in some patients with lone atrial fibrillation, because pulmonary vein isolation is a critical part of this operation. Indeed, Sueda et al23 ,24 have reported that refractory periods, as estimated by fibrillatory cycle lengths, are shorter in the left posterior atrial wall, particularly at the base of the left atrial appendage and the orifice of the left pulmonary veins. These results suggest that this is a critical area that acts as a driver for maintaining atrial fibrillation. This group also reported regular and repetitive activation in the left atria of 7 of 11 patients with chronic atrial fibrillation and mitral valve disease. These findings are consistent with catheter mapping and ablation studies showing focal discharges most commonly seen from the ostium of the left superior pulmonary

vein and Marshall's ligament. 22 ,25 Sueda described a posterior left atrial ablation (left atrial Maze) procedure that was performed with mitral valve operations in patients with chronic atrial fibrillation and mitral valve disease. 23 The procedure was successful in reducing or eliminating chronic atrial fibrillation in 31 of 36 patients (86%) by discharge, compared with 4 of 15 patients (27%) who underwent mitral valve surgery alone. A range of nonsurgical approaches to atrial fibrillation has been developed including atrioventricularnode ablation with insertion of permanent pacemaker to control symptoms (but not embolic events), catheter-based Maze approaches, pulmonary vein isolation and ablation, and implantable atrial de fibrillators. 26 ,27 Each of these therapies has had some success. In general, however, they are not as effective as the Maze procedure. As more reports indicate the low risk of performing the Maze procedure and the long-term benefits, more widespread application of the procedure is warranted. With the limitations of other therapies, the Maze procedure should be used more often for patients at chronic risk of thromboembolism.

Acknowledgments The authors would like to acknowledge the time and effort in data management and follow-up by Barb Elias, RN, and Kathy Hoercher, RN, and statistical analysis by Jennifer White,MS.

References 1. Cox JL, Schuessler RE, Cain ME, et al: Surgery for atrial fibrillation. Semin Thorac Cardiovasc Surg 1:67-73, 1989 2. CoxJL, Schuessler RE, BoineauJP: The surgical treatment of atrial fibrillation. I. Summary of the current concepts of the mechanisms of atrial flutter and atrial fibrillation. J Thorac Cardiovasc Surg 101:402-405, 1991 3. COX JL, Canavan TE, Schuessler RE, et al: The surgical treatment of atrial fibrillation. II. Intraoperative electrophysiologic mapping and description of the electrophysiologic basis of atrial flutter and atrial fibrillation. J Thorac Cardiovasc Surg 101:406-426,1991 4. COX JL: The surgical treatment of atrial fibrillation. IV. Surgical technique. J Thorac Cardiovasc Surg 101:584-592, 1991 5. McCarthy PM, Castle LW, Maloney]D, et al: Initial experience with the maze procedure for atrial fibrillation. J Thorac Cardiovasc Surg 105:1077-1087, 1993 6. McCarthy PM, Cosgrove DM, Castle LW, et al: Combined treatment of mitral regurgitation and atrial fibrillation with valvuloplasty and the Maze procedure. Am J Cardiol 71 :483486, 1993

Cox-MaJJ?: The Cleveland Clinic Experience

7. McCarthy PM, Castle LW, Trohman RG, et al: The Maze procedure: Surgical therapy for refractory atrial fibrillation. Cleve ClinJMed 60:161-165,1993 8. CoxJL, BoineauJP, Scheussler RB, et al: Five-year experience with the maze procedure for atrial fibrillation. Ann Thorac Surg 56:814-824,1994 9. CoxJL,Jaquiss RI, Schuessler RB, et al: Modification of the maze procedure for atrial flutter and atrial fibrillation. II. Surgical technique of the maze ill procedure. J Thorac Cardiovasc Surg 110:485-495, 1995 10. McCarthy PM. Cox-Maze ill procedure with mitral valve repair. Operative Techniques in Thoracic and Cardiovascular Surgery (in press) I!. Yoshihara F, Nishikimi T, Kosakai Y, et al: Atrial natriuretic peptide secretion and body fluid balance after bilateral atrial appendectomy by the maze procedure. J Thorac Cardiovasc Surg 116:213-219, 1998 12. Kim KB, Lee CH, Kim CH, et al: Effect of the Cox maze procedure on the secretion of atrial natriuretic peptide. J Thorac Cardiovasc Surg 115:139-147, 1998 13. Sundt TM 3rd, Camillo CJ, CoxJL: The maze procedure for cure of atrial fibrillation. Cardiol Clin 15:739-748, 1997 14. CoxJL, Ad N, Palazzo T: Impact of the maze procedure on the stroke rate in patients with atrial fibrillation.J Thorac Cardiovasc Surg 118:833-840, 1999 15. Feinberg MS, Waggoner D, Kater KM, et al: Restoration of atrial function after the maze procedure for patients with atrial fibrillation: Assessment by Doppler echocardiography. Circulation 90:II285-292, 1994 16. Albirini A, Scalia GM, Murray RD, et al: Left and right atrial transport function after the Maze procedure for atrial fibrillation: An echocardiographic Doppler follow-up study.J Am Soc Echocardiogr 10:937-945, 1997 17. Yashima N, Nasu M, Kawazoe K, et al: Serial evaluation of atrial function by Doppler echocardiography after the maze

18.

19.

20.

21. 22.

23.

24.

25.

26.

27.

29

procedure for chronic atrial fibrillation. Eur Heart J 18:496502, 1997 Yuda S, Nakatani S, Isobe F, et al: Comparative efficacy of the maze procedure for restoration of atrial contraction in patients with and without giant left atrium associated with mitral valve disease.J Am Coli Cardiol 31:1097-1 102, 1998 Chua YL, SchaffHV, Orszulak TA, et al: Outcome of mitral valve repair in patients with preoperative atrial fibrillation: Should the maze procedure be combined with mitral valvuloplasty? J Thorac Cardiovasc Surg 107:408-415, 1994 Moe G, Abildskov J: Atrial fibrillation as a self sustaining arrhythmia independent of focal discharge. Am Heart J 58:59-70, 1959 Moe G: On the multiple wavelet hypothesis of atrial fibrillation. Arch Int Pharmacodyn Ther 140:83-188, 1962 Haissaguerre M,Jais P, Shah DC, et al: Spontaneous initiation of atrial fibrillation by ectopic beats originating in the pulmonary veins. N EnglJ Med 339:659-666, 1998 Sueda T, Nagata H, Orihashi K, et al: Efficacy of a simple left atrial procedure for chronic atrial fibrillation in mitral valve operations. Ann Thorac Surg 63: 1070-1075, 1997 Sueda T, Nagata H, Shikata H, et al: Simple left atrial procedure for chronic atrial fibrillation associated with mitral valve disease. Ann Thorac Surg 62: 1796-1800, 1996 Hwang C, Karagueuzian H, Chen P: Idiopathic paroxysmal atrial fibrillation induced by a focal discharge mechanism in the left superior pulmonary vein: Possible roles of the ligament of Marshall. J Cardiovasc Electrophysiol 10:636-648, 1999 Baker BM, Smith JM, Cain ME: Nonpharmacologic approaches to the treatment of atrial fibrillation and atrial flutter.J Cardiovasc ElectrophysioI6:972-978, 1995 Luderitz B, Pfeiffer D, TebbenjohannsJ, et al: Nonpharmacologic strategies for treating atrial fibrillation. Am J Cardiol 77:45A-52A,1996