Since its introduction in 1987 [1], the maze procedure

Transcription

1 The Maze III Procedure for Atrial Fibrillation: Should the Indications Be Expanded? Roger C. Millar, MD, Joseph M. Arcidi, Jr, MD, and Peter J. M. Alison, MD Division of Cardiovascular and Thoracic Surgery, LDS Hospital, Salt Lake City, Utah Background. We favor the maze III procedure over nonsurgical treatments for atrial fibrillation and have advocated addition of the maze in symptomatic patients with other compelling indications for cardiac surgery. Methods. Characteristics and perioperative outcomes of our 76 cumulative maze recipients between 1993 to 1998 were reviewed. The original maze III technique was employed without modification. Results. Isolated maze III was performed in 19 patients (25%) and combined with other procedures in 57 patients (75%), 49 of these involving one or more valves. Patients having combined procedures were taking fewer antiarrhythmics (p < ), but were older (p < 0.01), more often female (p < 0.05), and more often had chronic atrial fibrillation (p < 0.01) compared with isolated maze III recipients. The mean duration of aortic clamping and cardiopulmonary bypass for isolated maze was and minutes, and for combined valve procedures it was and minutes. There was no operative mortality. Complications occurred in 15 patients (19.7%). At 3 months atrial fibrillation was cured in 73 of 75 patients (97.3%). Sick sinus syndrome required pacemaker implantation in 3 patients (4.0%). Conclusions. The maze III can be performed alone or as a combined procedure with equivalent success, and technical modifications may be unnecessary. A lower threshold for its expanded use in symptomatic patients with atrial fibrillation who require isolated or combined operations is appropriate. (Ann Thorac Surg 2000;70:1580 6) 2000 by The Society of Thoracic Surgeons Since its introduction in 1987 [1], the maze procedure stands as the only intervention capable of treating the rate-related, hemodynamic, and thromboembolic sequellae of atrial fibrillation. Its complexity relative to noncurative nodal ablation procedures, however, has tended to restrict its use to patients with solitary atrial fibrillation recalcitrant to medical therapy. The majority of patients with atrial fibrillation are not sufficiently symptomatic to warrant an intervention aside from rate control and anticoagulation. Conversely, many patients have clearly symptomatic atrial fibrillation in conjunction with other compelling indications for cardiac surgery. Our experience with the maze procedure has focused on this latter group of patients, and we have favored addition of the maze to concurrent cardiac surgical procedures when feasible. Moreover, based on the attendant success of the maze III demonstrated by Sundt and associates [2], we have employed the original maze III without modification, even in combined procedures. This report outlines our cumulative perioperative experience with the maze procedure as an assessment of our less restrictive approach to its implementation. Material and Methods The records and early follow-up of consecutive patients who underwent the maze III procedure in our practice between 1993 and 1998 were examined. Accepted for publication April 8, Address reprint requests to Dr Millar, Salt Lake Cardiovascular and Thoracic Surgery, 324 Tenth Ave, Suite 160, Salt Lake City, UT Demographic information, the duration of chronic or paroxysmal atrial fibrillation, number of trialed medications, and the need for hospital admission or electrical cardioversion were reviewed. The New York Heart Association (NYHA) functional classification was determined based on congestive failure symptoms in the majority of patients or anginal symptoms in the remainder. Symptoms of atrial fibrillation included intolerance of uncontrolled or rapid ventricular response, fatigue due to absence of atrioventricular synchrony, or documented thromboembolic episodes. Operative indications were reviewed and the principal indication was categorized as follows: atrial fibrillation symptoms or complications, congestive failure from mitral valve disease, angina, or symptoms from aortic valvular disease. Perioperative complications specifically noted included reoperation for bleeding, stroke, ventilator support beyond 48 hours, intraaortic balloon pump insertion, low cardiac output requiring more than 3 days of catecholamine support, infection, and permanent pacemaker implantation. Rhythm was assessed at discharge and at postoperative visits within the first 3 months. Operative Technique Operations in all patients were supervised or performed by one surgeon (RCM). The maze III as originally described by Cox and associates [3] was utilized without modification. Cryoablation was limited to the coronary sinus, the posterior mitral annulus, and two sites on the tricuspid valve annulus using 70 C for 3 minutes at the coronary sinus and 2 minutes at the valve annuli. The 2000 by The Society of Thoracic Surgeons /00/$20.00 Published by Elsevier Science Inc PII S (00)

2 Ann Thorac Surg MILLAR ET AL 2000;70: MAZE III PROCEDURE EXPANDED INDICATIONS 1581 patient was maintained at normothermia during performance of the right atriotomies. During creation of the left atriotomies, moderate systemic hypothermia and myocardial preservation with intermittently dosed 6 C to 8 C retrograde blood cardioplegia were used uniformly. Valvular procedures and coronary artery bypass were integrated into the maze operation. Tricuspid annuloplasty was accomplished after completion of the right atriotomies with the aorta unclamped. Distal saphenous vein coronary anastomoses were constructed shortly after aortic clamping. Mitral replacement or reconstruction was perfomed before left atrial closure. Aortic valve replacement or root reconstruction was undertaken after left atrial closure. Right atrial closure was completed after aortic clamp removal. Postoperative Management Patterns of postoperative recovery were identical to those for routine open heart surgical patients. Temporary pacing if needed was employed for at least 7 days before commitment to permanent pacer implantation for complete heart block or sick sinus syndrome. Negative fluid balance was maintained with intermittent diuretic administration, rather than a continuous infusion, until an optimal dry weight was attained. Antiarrhythmic medication was not routinely used postoperatively; recurrent atrial fibrillation was treated de novo or based on preoperative antiarrhythmic use. Based on Cox s early practice [5] warfarin anticoagulation was continued for 6 to 8 weeks postoperatively irrespective of the rhythm at discharge. Within the past year this has been omitted for selected patients who do not have specific indications for anticoagulation. Statistical Analysis The distributions of continuous variables were expressed as mean standard deviation, and comparisons were performed using unpaired t tests. Comparisons of categorical variables were performed using 2 tests. Results Patient Characteristics and Operative Indications The maze III procedure was performed in 76 consecutive patients, as an isolated procedure in 19 patients (25%), and in combination with other cardiac procedures in 57 patients (75%). The mean age was years (range 43 to 76), and 45 (59%) of patients were male. The mean NYHA functional class was and mean left ventricular ejection fraction was 60% 10%. Presenting features of atrial fibrillation are summarized in Table 1. Chronic atrial fibrillation predominated over paroxysmal and the duration of atrial fibrillation was most commonly 1 to 5 years or more than 10 years. Eighteen patients (24%) presented with complications from atrial fibrillation or its treatment. This included neurologic symptoms from thromboembolism in 13 patients and a history of warfarin intolerance or noncompliance in 4 patients. A DDD pacer had been implanted preoperatively in 1 patient Table 1. Presentation of Atrial Fibrillation in Maze III Patients (n 76) Atrial Fibrillation Characteristic n % Chronic Paroxysmal Neurologic symptoms 13 Contraindication to anticoagulation 4 Amiodarone pulmonary toxicity Duration 1 year years years years Number of trialed medications Amiodarone trial 7 10 Electrical cardioversion Previous ablation 3 4 for sick sinus syndrome. Three patients had received a total of five attempts at electrical ablation, but amiodarone was not commonly used. Compared with isolated maze III recipients, patients undergoing combined operations with the maze were older ( versus years; p 0.01), and more often female (28 of 57 [49%] versus 3 of 19 [16%]; p 0.05). They were more likely to have chronic atrial fibrillation (49 of 57 [86%] versus 11 of 19 [58%]; p 0.01), had received fewer trialed antiarrhythmics ( versus ; p ), and were less likely to have one or more cardioversion attempts (20 of 57 [35%] versus 14 of 19 [74%]; p 0.01). The primary indication for operation was symptoms or complications from atrial fibrillation alone or in combination with congestive failure symptoms from mitral valve disease in 52 patients (68%). Symptoms of atrial fibrillation were present in 24 other patients (32%) but atrial fibrillation was not itself the compelling indication for operation. Operative Procedures The operative experience by year is shown in Figure 1, and the distribution of procedures performed is outlined in Table 2, along with the mean durations of aortic Fig 1. Experience with the maze III procedure by year.

3 1582 MILLAR ET AL Ann Thorac Surg MAZE III PROCEDURE EXPANDED INDICATIONS 2000;70: Table 2. Operative Procedures, Mean Duration of Aortic Clamping and Cardiopulmonary Bypass (n 76) Procedure n % AXC (min) CPB (min) Isolated maze III Maze mitral valve a,b mitral valve tricuspid valve a mitral homograft tricuspid valve c Maze aortic valve aortic valve mitral valve aortic valve mitral valve tricuspid valve Total maze valve d d Maze CABG or other procedures a CABG included. b Septal myomectomy included. c Atrial septal defect closure included. d All maze valve patients included. AXC aortic crossclamp; CPB cardiopulmonary bypass; CABG coronary artery bypass grafting. Table 3. Distribution of Valve Procedures Combined With the Maze III in 49 Patients Valve Location and Procedure Mitral valve (43 patients) Replacements Bioprosthetic 8 Mechanical 21 Homograft 3 Repairs 11 Reoperative mitral valve procedures 6 Aortic valve (13 patients) Replacements Freestyle stentless bioprosthesis 8 Stented bioprosthesis 1 Mechanical 3 Valve-sparing reconstruction with 1 ascending aneurysm repair Reoperative aortic valve procedures 3 Tricuspid valve (14 patients) Bioprosthetic replacements 2 Repairs DeVega annuloplasty 10 Carpentier annuloplasty 2 n cross-clamp application and cardiopulmonary bypass. The maze was combined with one or more valve procedures in 49 patients (64%). Coronary artery bypass grafting was included in 12 patients (16%) with a mean of distal anastomoses per patient constructed. Septal myomectomy for hypertrophic cardiomyopathy was performed in 1 patient, and 5 underwent atrial septal defect closure. The longest duration of aortic clamping and cardiopulmonary bypass occurred in 3 patients who underwent mitral homograft implantation. Mitral valve procedures were performed in 43 patients (57%) and were the most frequent operation combined with the maze III procedure. The range of valve procedures is listed in Table 3. Mitral repair was attempted in 15 patients and was successful in 11. A mechanical prosthesis was used in the majority of mitral replacements, whereas the Freestyle porcine stentless bioprosthesis (Medtronic Cardiac Surgery, Minneapolis, MN) was used in the majority of aortic valve replacements. Six of the mitral and 3 of the aortic valve procedures constituted reoperative valve procedures. DeVega annuloplasty was the most commonly performed tricuspid valve procedure. Among the 49 valve patients, rheumatic disease was the underlying pathology in 32 patients (65%). The NYHA Class in patients with rheumatic pathology was higher compared with the other patients in the series ( versus , p 0.001). Operation included replacement of a degenerated porcine prosthesis in 2 of these patients and repair of a paravalvular leak in another. Degenerative valve disease was present in 16 patients (33%) and was responsible for most of the aortic valve pathology. Systolic anterior motion of the mitral valve was present in 1 patient (2%) with hypertrophic obstructive cardiomyopathy. Perioperative Outcomes There was no operative mortality. Major complications occurred in 15 patients (19.7%) and are outlined in Table 4. Reoperation for bleeding occurred in 3 patients (3.9%). Intraaortic balloon pump insertion was performed in 1 patient, following a reexploration for bleeding. Prolonged inotropic or respiratory support was required in 8 patients, all of whom had combined procedures. Intraoperative cerebrovascular accidents occurred in 2 patients, 1 of whom had preoperative neurologic symptoms and underwent concomitant coronary bypass grafting and another with associated mitral pathology. Two patients who required permanent pacemaker implantation for complete heart block had undergone combined operations, which included a tricuspid valve replacement in the one and a reoperative mitral replacement in the other. One isolated maze patient with early postoperative ventricular tachycardia, who showed sinus rhythm at discharge and 1 month postoperatively, died suddenly 7 weeks after operation. She was not taking any antiar-

4 Ann Thorac Surg MILLAR ET AL 2000;70: MAZE III PROCEDURE EXPANDED INDICATIONS 1583 Table 4. Perioperative Outcomes (n 76) Number Mortality 0 Major complications (15 patients) Reoperation for bleeding 3 Cerebrovascular accident 2 Respiratory support 48 hours 5 Inotropic support 72 hours 4 Intraaortic balloon pump insertion 1 Sternal infection 2 Pacemaker for complete heart block 2 Mean ICU stay days (median 2) Mean postoperative stay days (median 7) ICU intensive care unit. rhythmic medication but had frequent premature ventricular contractions. Postoperative Rhythm The evolution of postoperative rhythm is shown in Table 5, with 3-month data obtained in all 75 patients. At discharge 57 patients (75%) showed sinus rhythm. Thirty patients (39%) were discharged on antiarrhythmic medication because of early postoperative atrial dysrhythmias, including 19 of the patients with sinus rhythm. The 11 other patients discharged on antiarrhythmic medication included 7 with atrial fibrillation, 3 with junctional rhythm, and 1 paced for complete heart block. Among the 8 patients discharged in atrial fibrillation or flutter, 7 recovered sinus rhythm within 3 months, as did 6 of the 9 patients with junctional rhythm at discharge. Three patients (4%) had pacemakers implanted for sick sinus syndrome within the first 8 months. At 3 months 68 of 75 patients (90.6%), including 1 with a DDD pacer inserted preoperatively, showed sinus rhythm. Restoration of sinus rhythm occurred with equivalent frequency among patients undergoing combined procedures (51 of 57, 89.5%) or an isolated maze III operation (16 of 18, 88.9%). Cure of atrial fibrillation at 3 months was obtained in 73 of 75 patients (97.3%). Table 5. Evolution of Postoperative Rhythm Following the Maze III Procedure Rhythm At Discharge (n 76) No of Patients % At 3 Months (n 75) No of Patients % Sinus rhythm Junctional rhythm Atrial fibrillation Atrial flutter Pacer for complete heart block Pacer for sick sinus syndrome Comment The maze III procedure offers a solution for patients with symptomatic atrial fibrillation, but one which is complex and technically challenging. This concern is consistent in reports from American centers [4, 5] and particularly from Asian institutions [6 8], where substantial experience with the maze procedure has been acquired. Interruption of macroreentrant ciruits responsible for atrial fibrillation requires performance of 10 directed incisions in the right and left atria and the interatrial septum. Although injury to coronary arterial and venous structures has been infrequently reported [9], bleeding from less accessible suture lines has been a common concern with isolated maze procedures. This may become a significant consideration when the maze is combined with mitral valve replacement or other cardiac procedures [6, 7]. The complexity of patients undergoing the maze in our series resulted from a lack of isolated atrial fibrillation patient referrals in our practice. Our experience reflects a motivation for its performance in patients referred specifically for correction of valvular pathology rather than for correction of symptomatic atrial fibrillation. Consequently, our 57 patients (75%) who underwent combined procedures were fundamentally different from the 19 patients (25%) who received an isolated maze III. Recipients of combined procedures were more likely to have presented with chronic atrial fibrillation (p 0.01) and had received fewer trialed antiarrhythmics (p ) or cardioversion attempts (p 0.01) than isolated maze recipients. Although isolated maze III patients were frequently young males, patients having combined maze operations were older (p 0.01), and more often female (p 0.05). Consistent with this description, the most common combined operation was a mitral valve procedure, performed on 75% of these 57 patients. The complexity in our maze experience is distinct from all other American series (Table 6) with regard to the proportions of combined procedures, mitral valve procedures, and multiple valve procedures. Our geographic location in the Intermountain West region is associated with a continued incidence of rheumatic valvular disease, which was observed in 65% of combined valve and maze procedures, nearly twice that of degenerative valve disease. The NYHA class in patients with rheumatic disease was higher than in other patients (p 0.001) and 23 of 32 patients (72%) had NHYA class III or IV function. Each of the two multiple reoperative and 86% (6 of 7) of singly reoperative procedures, as well as 79% (15 of 19) of multiple valve and maze operations occurred in patients with underlying rheumatic valvular disease. Given this complexity our perioperative outcomes, which form the basis of this report, have been particularly encouraging. In addition to an absence of early mortality, complications related to bleeding have been rare in our recent experience. Patients have generally required only modest postoperative catecholamine support, and the single intraaortic balloon pump recipient underwent insertion after a return for bleeding. Neither

5 1584 MILLAR ET AL Ann Thorac Surg MAZE III PROCEDURE EXPANDED INDICATIONS 2000;70: Table 6. Complexity and Outcomes in Maze Procedure Series n Combined Procedures Mitral Valve Procedures Multiple Valve Procedures Operative Mortality New Pacers for Sick Sinus Syndrome Cure of Atrial Fibrillation Cox (maze III)[2, 5] (40%) 25 (17%) 0 3 (2.1%) 20 (14%) 136 (97%) a McCarthy [4] 14 4 (29%) 2 (14%) 0 1 (7.1%) 2 (14%) 13 (100%) Kawaguchi [22] (90%) 162 (84%) 110 (57%) 4 (2.1%) 2 (1%) 162 (86%) Kamata [23] (100%) 65 (63%) 0 4 (3.8%) 6 (6%) 79 (79%) Millar (75%) 43 (57%) 20 (26%) 0 3 (4%) 73 (97%) a 100% success in curing recurrent atrial fibrillation with medication. the duration of catecholamine support nor mechanical ventilation has been prolonged, as reflected by a mean intensive care stay of 2.8 days (median 2). Fluid retention, consistent with diminished atrial naturetic peptide levels [11 13], has responded to combination diuretic therapy and has not required continuous diuretic infusions. We have observed an 8.5 day mean postoperative stay (median 7), equivalent to that expected for a multiple valve operation. Our outcomes reflect a consideration for ventricular function in patient selection; although the complexity of our patients has increased over time (Fig 1), the mean left ventricular ejection fraction in our series was 60% 10%. We have continued to evaluate patients with compromised left ventricular function individually, particularly those candidates for the maze III combined with more than one valve procedure. The restoration of sinus rhythm has validated our motivation toward performance of the maze procedure in this complex group of patients. In the early postoperative period, atrial fibrillation recurred in 30 patients (39%), consistent with a temporarily shortened atrial refractory period as outlined by Cox [5]. Our approach to perioperative atrial fibrillation since 1996 has been to individualize treatment and to avoid routine antiarrhythmic prophylaxis. Utilizing this strategy, 75% of all patients were able to be discharged in sinus rhythm, with 33% taking antiarrhythmic medication. By 3 months, the majority of patients discharged in junctional rhythm, atrial fibrillation, or atrial flutter converted to sinus rhythm, increasing the recovery of sinus rhythm to 90.6%. According to Cox s definition [5], the cure rate of atrial fibrillation has been 97.3%, and there has been equivalent success among combined and isolated maze III recipients. The complexity of the maze III procedure has been the principal incentive for the development of technical modifications, predominantly from Japanese centers. Although several maze variations [14, 15] share only modest resemblence to any iterations of the maze incisions, most modifications [16, 17], including that of Kamata and associates [18], involve substitution of linear cryolesions for full thickness atriotomies in order to decrease the duration of the procedure and the risk of suture line bleeding. Kosakai s modification [19] additionally aims to minimize injury to the arterial supply of the sinoatrial node. The recently introduced radial procedure for atrial fibrillation [20, 21] seeks additionally to optimize left atrial transport function. The complexity in our series is similar to mature experience reported by Kosakai s group [22] and Kamata and associates [23] (Table 6); however, our rate of cure of atrial fibrillation exceeded that obtained with modifications and was accomplished with favorable aortic cross-clamp and bypass times. Assessment of left atrial transport in the current series warrants further investigation, but our outcomes favor continued use of the original maze III procedure and suggest that technical modifications may not be necessary. The aortic cross-clamp and cardiopulmonary bypass times (Table 3) among categories of operations parallel our experience in combining valve operations with the maze III procedure. Comparison of the minute aortic clamp and minute bypass duration for isolated maze III to and minutes, respectively, for mitral valve procedures demonstrates that mitral valve repair or replacement added roughly 40 minutes of additional aortic clamp and bypass time to the maze. Tricuspid valve procedures, when added to mitral valve procedures, required no additional aortic clamp time and 20 minutes of bypass time. The exposure for these atrioventricular valve procedures is provided in the course of performance of the maze III and no additional incision is necessary. Aortic valve replacement, however, is a separate procedure from the maze; the additional 65 minutes include the time necessary to create and close an aortotomy. Combination of aortic and mitral valve operations with the maze III was not undertaken until recently; the shorter time increment reflects a level of experience that was not present while learning to combine single valve operations with the maze III procedure. If the goal of the maze III procedure is to eliminate all of the symptoms or complications associated with atrial fibrillation, then a limitation of our series might be the utilization of mechanical prostheses, particularly in the mitral position. Given that these patients will require life-long warfarin anticoagulation with its attendant risks of hemorrhage and thrombosis, addition of the maze procedure might appear to carry excess risk in order to provide the benefit of atrioventricular synchrony as well as rate control. Although implantation of a thrombogenic prosthesis might seem an incomplete solution for combined valvular heart disease and atrial fibrillation, it attests to the pronounced incidence of rheumatic valvular disease in our practice. Our approach to rheumatic mitral valve

6 Ann Thorac Surg MILLAR ET AL 2000;70: MAZE III PROCEDURE EXPANDED INDICATIONS 1585 disease has included an attempt at mitral repair in patients with less extensive fusion of the subvalvar apparatus. Our attempts at mitral repair were successful in only 1 of 3 patients with rheumatic disease compared with 9 of 10 patients with degenerative disease. Published results for repair of rheumatic disease [24 26] have been considerably less favorable than for degenerative disease [27, 28]. Homograft replacement of the mitral valve combined with the maze III procedure was performed in 3 patients in this series and has been separately reported by our group [29]. In addition to avoiding anticoagulation, this combination may provide durable restoration of mitral anatomy and physiology. It should, however, be reserved for patients with rheumatic disease limited to the mitral valve, given the long ischemic time and cardiopulmonary bypass time required for this demanding procedure. As our experience with homograft mitral replacement matures, we anticipate its more frequent use in combination with the maze procedure. Our encouraging early outcomes with the maze III performed alone or as a combined operation raise implications regarding patient selection for the maze procedure. Our surgical recommendation for adding the maze III to recipients of combined procedures was based on the coexistence of moderately symptomatic atrial fibrillation; traditional criteria involving failure of an exhaustive medical regimen were applicable only to patients undergoing an isolated maze procedure. This less restrictive approach was most evident in 32% of our patients who underwent the maze III, but did not, retrospectively, appear to have symptoms or complications of atrial fibrillation as a compelling indication for operation. This observation demonstrates that atrial fibrillation has not been the only symptom of concern when a combined procedure was undertaken. We have felt, however, that the maze III carries substantial benefit, and we have extended it to patients taking fewer medications or having fewer cardioversion attempts than cardiologists generally administer. Although longer follow-up of our patients is necessary, it will not be possible to rigorously evaluate the qualityof-life improvement in the current series of patients, owing to a lack of formal preoperative assessment. Quality-of-life assessment has been incorporated into trials [30, 31] of therapy for atrial fibrillation as a measurement for success of nonsurgical approaches, and limited enrollment in the maze treatment arm of the multicenter Atrial Fibrillation Follow-up Investigation of Rhythm Management (AFFIRM) trial [31] can be anticipated. Confirmation of the superiority of the maze procedure over nonsurgical treatments will require demonstration of a quality-of-life advantage. Our early success supports using a lower threshold for addition of the maze III to other cardiac procedures. It seems appropriate, therefore, to recommend a lower threshold for referral of patients requiring an isolated maze III as well. References 1. Cox JL, Schuessler RB, D Agostino HJ Jr, et al. The surgical treatment of atrial fibrillation. III. Development of a definitive surgical procedure. J Thorac Cardiovasc Surg 1991;101: Sundt TM III, Camillo CJ, Cox JL. The maze procedure for cure of atrial fibrillation. Cardiol Clin 1997;15: Cox JL, Jaquiss RD, Schuessler RB, Boineau JP. Modification of the maze procedure for atrial flutter and atrial fibrillation. II. Surgical technique of the maze III procedure. J Thorac Cardiovasc Surg 1995;110: McCarthy PM, Castle LW, Maloney JD, et al. Initial experience with the maze procedure for atrial fibrillation. J Thorac Cardiovasc Surg 1993;105: Cox JL, Boineau JP, Scheussler RB, Kater KM, Lappas DG. Five-year experience with the maze procedure for atrial fibrillation. Ann Thorac Surg 1993;56: Kobayashi J, Kosakai Y, Isobe F, et al. Rationale of the Cox maze procedure for atrial fibrillation during redo mitral valve operations. J Thorac Cardiovasc Surg 1996;112: Izumoto H, Kawazoe K, Kitahara H, Kamata J. Operative results after the Cox/maze procedure combined with a mitral valve operation. Ann Thorac Surg 1998;66: Isobe F, Kawashima Y. The outcome and indications of the Cox maze III procedure for chronic atrial fibrillation with mitral valve disease. J Thorac Cardiovasc Surg 1998;116: Sueda T, Shikata H, Mitsui N, Nagata H, Matsuura Y. Myocardial infarction after a maze procedure for idiopathic atrial fibrillation. J Thorac Cardiovasc Surg 1996;112: Kosakai Y, Kawaguchi AT, Fumitaka I, et al. Modified maze procedure for patients with atrial fibrillation undergoing simultaneous open heart surgery. Circulation 1995;92(suppl 2): Nakamura M, Niinuma H, Chiba M, et al. Effect of the maze procedure for atrial fibrillation on atrial and brain natriuretic peptide. Am J Cardiol 1997;79: Kim K, Lee C, Kim C, Cha Y. Effect of the Cox maze procedure on the secretion of atrial natriuretic peptide. J Thorac Cardiovasc Surg 1998;115: 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 1998;116: Hioki M, Ikeshita M, Iedokoro Y, et al. Successful combined operation for mitral stenosis and atrial fibrillation. Ann Thorac Surg 1993;55: Fukada J, Morishita K, Komatsu K, et al. Is atrial fibrillation resulting from rheumatic disease a proper indication for the maze procedure? Ann Thorac Surg 1998;65: Sandoval N, Velasco V, Orjuela H, et al. Concomitant mitral valve or atrial septal defect surgery and the modified Coxmaze procedure. Am J Cardiol 1996;7: Melo JQ, Neves J, Adragao P, et al. When and how to report results of surgery on atrial fibrillation. Eur J Cardiothorac Surg 1997;12: Kamata J, Kawazoe K, Izumoto H, et al. Predictors of sinus rhythm restoration after Cox maze procedure concomitant with other cardiac operations. Ann Thorac Surg 1997;64: Kosaikai Y, Kawaguchi AT, Isobe F, et al. Cox maze procedure for chronic atrial fibrillation associated with mitral valve disease. J Thorac Cardiovasc Surg 1994;108: Nitta T, Lee R, Schuessler RB, Boineau JP, Cox JL. Radial approach: a new concept in surgical treatment for atrial fibrillation. I. Concept, anatomic and physiologic bases and development of a procedure. Ann Thorac Surg 1999;67: Nitta T, Lee R, Watanabe H, et al. Radial approach: a new concept in surgical treatment for atrial fibrillation. II. Electrophysiologic effects and atrial contribution to ventricular filling. Ann Thorac Surg 1999;67:36 50.

7 1586 MILLAR ET AL Ann Thorac Surg MAZE III PROCEDURE EXPANDED INDICATIONS 2000;70: Kawaguchi AT, Kosakai Y, Isobe F, et al. Factors affecting rhythm after the maze procedure for atrial fibrillation. Circulation 1996;94(Suppl 2): Kamata J, Nakai K, Chiba N, et al. 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: Gillinov AM, Cosgrove DM, Lytle BW, et al. Reoperation for failure of mitral repair. J Thorac Cardiovasc Surg 1997;113: Bernal JM, Fernandez-Vals M, Rabasa JM, Gutierrez-Garcia F, Morales C, Regvuelta JM. Repair of nonsevere rheumatic aortic valve disease during other valvular procedures: is it safe? J Thorac Cardiovasc Surg 1998;115: Duran CM, Gometza B, Saad E. Valve repair in rheumatic mitral disease: an unsolved problem. J Cardiac Surg 1994; 9(Suppl): David TE, Armstrong S, Sun Z, Daniel L. Late results of mitral valve repair for mitral regurgitation due to degenerative disease. Ann Thorac Surg 1993;56: Gillinov AM, Cosgrove DM, Blackstone EH, et al. Durability of mitral valve repair for degenerative disease. J Thorac Cardiovasc Surg 1998;116: Doty DB, Dilip KA, Millar RC. Mitral valve replacement with homograft and maze III procedure. Ann Thorac Surg 2000; 69: Kay GN, Ellenbogen KA, Giudici M, et al. The Ablate and Pace trial: a prospective study of catheter ablation of the AV conduction system and permanent pacemaker implantation for treatment of atrial fibrillation. J Intervent Cardiac Electrophys 1998;2: Wyse DG, Anderson JL, Antman EM, et al. Atrial fibrillation follow-up investigation of rhythm management the AF- FIRM study design. Am J Cardiol 1997;79: The Thoracic Surgery Foundation for Research and Education Datascope Makes Major Gift The Thoracic Surgery Foundation is most pleased to announce that Datascope Corporation recently pledged $1 million towards the $3.75 million in matching funds needed for The Foundation to secure a $5 million grant from the NHLBI. These funds will be used to support the jointly sponsored TSFRE and NHLBI Mentored Clinical Scientist Development Award Program. The ten-year program is designed to develop outstanding cardiothoracic surgeon-scientists. Datascope s gift to The Thoracic Surgery Foundation is a very important step forward in TSF s efforts to secure the NHLBI funds. For in 1999, the von Liebig Foundation also agreed to support The Foundation in this initiative if The Foundation could find other partners, such as Datascope, to commit to the matching funds. The Foundation is extremely grateful to Datascope and to Mr. Lawrence Saper, President and CEO of Datascope, for this very generous support Research Award Applications Information and applications for Thoracic Surgery Foundation s Research Grants, Research Fellowships and Career Development Awards are available now. To request an application, please contact Lainie Castle at the Foundation office at telephone: (312) , extension Applications can also be downloaded from The Foundation s home page, Kennedy School Sabbatical The Thoracic Surgery Foundation for Research and Education announces that applications for a yearlong sabbatical to do graduate studies in health care policy at Harvard University s Kennedy School of Government are now available. Candidates will be accepted early on a rolling admission basis by the graduate school, and scholarships will be awarded by The Foundation s Education Committee. The Kennedy School will review completed applications for the Program starting January 3, 2001, and is prepared to make decisions on applications within two weeks of their receipt. The Foundation will make a decision on an application for Alley-Sheridan Scholarship within two weeks of its receipt to The Foundation office. If an interested surgeon has applied to The Kennedy School very early in January, it is possible to have the admissions and scholarship decisions completed by February 1st. Please call Lainie Castle at The Foundation office for more information at (312) , extension by The Society of Thoracic Surgeons Ann Thorac Surg 2000;70: /00/$20.00 Published by Elsevier Science Inc