The confusion among cardiac surgeons regarding whether or not to include the right atrial lesions of the Maze procedure for the treatment of atrial fibrillation (AF) is understandable. Our literature includes publications stating that right atrial lesions are essential to attaining optimal outcomes, a few suggest that right atrial lesions contribute nothing to the outcomes, , and still others argue that the right atrial lesions are helpful in some patients but not in others. , Many surgeons are convinced that inclusion of the right atrial lesions in a Maze procedure results in damage of the sinoatrial (SA) node and/or the atrioventricular (AV) node and that therefore right atrial lesions result in an increased need for new permanent pacemakers after surgery. Some also believe that including the right atrial lesions adds to cardiopulmonary bypass time and aortic cross-clamp time and that therefore, they add to perioperative morbidity and mortality.
One reason for this confusion is that we are accustomed to thinking of the right atrium (RA) and left atrium (LA) as separate anatomic structures because they provide different hemodynamic roles. However, atrial myocardium is one contiguous muscle mass, and the fact that it is partitioned into two separate chambers by the atrial septum is irrelevant to it being a single electrophysiological entity ( Fig. 13.1 ). The presence of multiple interatrial muscle bands with preferential routes of conduction further emphasizes the singular nature of electrical activity involving the myocardium of the two atria ( Fig. 13.2 ). Global electrophysiologic maps of the atria confirm that the two atria act as one mass of myocardium with no electrical distinction between the RA and LA during normal sinus rhythm or during atrial arrhythmias, including AF ( Fig. 13.3 ).
Early electrocardiography-gated magnetic resonance scans from 1986 that allowed three-dimensional reconstruction of the atria demonstrating that the myocardium of the two atria is one muscle mass. (A) Anterior view. (B) Left lateral view. (C) Superior view. LA, Left atrium; LAA, left atrial appendage; RA, right atrium; RAA, right atrial appendage; SVC, superior vena cava.
Interconnecting muscle fibers (dark red lines) in the vicinity of the pulmonary veins and between the left atrium and right atrium emphasizing their electrical continuity. IVC, Inferior vena cava; LAA, Left atrial appendage; LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; RAA, right atrial appendage; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein; SVC, superior vena cava.
A single frame from the video of a phase map of the atria during atrial fibrillation. Note that the electrical activity between the two atria is contiguous and uninterrupted. (A) Anterior view of the atria. (B) Posterior view of the atria. IVC, Inferior vena cava; LA, left atrium; LAA, left atrial appendage; LIPV, left inferior pulmonary vein; LSPV, left superior pulmonary vein; MV, mitral valve; RA, right atrium; RAA, right atrial appendage; RIPV, right inferior pulmonary vein; RSPV, right superior pulmonary vein; SVC, superior vena cava; TV, tricuspid valve.
This dilemma of when to include the right atrial lesions as part of a surgical AF ablation procedure is further confounded by the fact that the treatment goals in paroxysmal (PAF) and non-paroxysmal AF (non-PAF) are completely different as explained in Chapter 10 . The objective of surgery for PAF is to isolate as many of the atrial triggers as possible to decrease the likelihood of an AF episode being induced. The objective of surgery for non-PAF is to interrupt the macro-reentrant circuits (drivers) that sustain AF, thereby rendering the atria incapable of fibrillating.
Biatrial Lesions Versus Left Atrial–Only Lesions
The first meta-analysis comparing biatrial versus left atrial–only lesions was published in 2006 by Barnett and Ad. Their meta-analysis included 69 studies comprising 5885 patients who had undergone concomitant AF ablation at the time of coronary artery bypass graft or valve surgery. The study showed that biatrial lesions were more successful in ablating AF at 3 months, 1 year, 2 years, and 3 years ( Fig. 13.4 ). Their findings were confirmed by subsequent reports from Burnett et al. in 2014 and Gelsomino et al. in 2016. However, in 2013, Soni et al. reported that in 325 patients, a complete LA lesion set was better than pulmonary vein isolation (PVI) alone and that adding the right atrial lesions to a complete left atrial lesion set did not improve the success rate for AF ablation. Two years later, the cardiothoracic surgical trials network (CTSN) Clinical Trials Investigators reported the results of the first multicenter randomized controlled trial (RCT) showing that surgical AF ablation in patients with persistent forms of AF undergoing mitral valve surgery had better outcomes than patients in whom the AF was left untreated. They then did an inadequately powered comparison between PVI and what they called a “biatrial Maze” procedure and showed that at 1 year, 61% of the PVI group had no recurrent AF versus 66% of the “biatrial Maze” group with no recurrent AF. As discussed in the article, this suggested that PVI was as effective as a “biatrial Maze” procedure for the treatment of patients with persistent forms of AF associated with mitral valve disease. However, in 2019, Blackstone et al. published a follow-up study of the same cohort of patients using more sophisticated statistical analysis and showed that in fact, the “biatrial Maze” group had less recurrent AF than did the PVI group. In 2017, Ad et al. reported their personal experience in 800 patients undergoing AF ablation between 2005 and 2015, 110 of whom had LA-only ablation. Their conclusions were that LA-only ablation yielded “acceptable” success rates in patients with shorter AF duration and smaller left atria, but the success rate for LA-only ablation was not as effective as biatrial Maze procedures in patients with traditional predictors of failure. Finally, the group at Northwestern University published results suggesting that in their 836 patients with AF undergoing mitral valve surgery, adding the right atrial lesions did not affect the outcomes. Because everything that is known about the electrophysiology of AF indicates this to be impossible, there must be some alternate explanation.
Percent of patients free of atrial fibrillation (AF) at indicated time intervals after surgery for AF. The results for biatrial ablation procedures (bilateral) were superior to those for procedures that did not include the right atrial lesions (unilateral). These data are from a metaanalysis of 69 reports totaling 5885 patients. (Reproduced from Barnett SD, Ad N. Surgical ablation as treatment for the elimination of atrial fibrillation: a meta-analysis. J Thorac Cardiovasc Surg . 2006;131(5):1029–1035.)
One possibility that affected the Northwestern study is that there are a few cardiac surgeons who are recognized for their expertise in mitral valve surgery and who therefore receive more referrals for mitral surgery than do other surgeons. This higher volume of mitral valve patients is associated with earlier referrals as well. Earlier surgical intervention means patients undergo surgery at a less advanced stage of mitral valve disease with a greater likelihood of intervening before the RA has become actively involved in sustaining the AF. In such patients with early AF secondary to mitral valve disease, confining the Maze lesions to the LA would be expected to produce similar outcomes to those of a full biatrial Maze procedure.
Another possible explanation for surgeons getting similar outcomes after LA-only procedures and biatrial Maze procedures is that they perform the biatrial Maze procedure incorrectly so that an LA-only procedure becomes as successful as an incorrectly performed biatrial Maze. That was a problem in the CTSN RCT mentioned earlier in which the authors compared the results of the two different types of surgical procedures in the subgroup of surgically treated patients. They acknowledged that their study was not sufficiently powered to compare the LA-only procedure (PVI) with the biatrial Maze procedure, but they did so anyway. Despite the follow-up report by Blackstone showing the superiority of the biatrial procedure over the PVI, the fact remains that this oft-quoted RCT had one unfortunate consequence. Its unpowered comparison between PVI and a biatrial Maze procedure suggested that PVI was as effective as a biatrial Maze procedure (when imperfectly performed) for the treatment of patients with persistent AF. This misinformation subsequently led to an inadequate surgical procedure (PVI) being performed for concomitant persistent AF in patients all over the world.
In the biatrial Maze subgroup of the CTSN trial, the surgeons were apparently trying to perform a Maze-IV procedure, but not a single patient in the subgroup received a properly performed Maze procedure. That is why the 1-year success rate for a “biatrial Maze” procedure in the study was a dismal 66%, which, by definition, means that the procedure was being done incorrectly. Not only were no coronary sinus lesions performed in any of the patients (an obligatory 10–15% failure rate), but most of the left atrial lesions and all of the right atrial lesions were performed with unipolar ablation devices that are notoriously unreliable for creating complete (contiguous, uniformly transmural) atrial lesions. In addition, surgeons in the trial were allowed to perform all of the right atrial lesions of the Maze procedure even if a cavo-tricuspid isthmus (CTI) lesion had been performed previously by catheter ablation. As explained later, this combination of lesions in the RA results in complete isolation of the lower third of the RA where the sinus bradycardia region of the atrial pacemaker complex resides.
Another problem with the CTSN trial was that the requirements for new permanent pacemakers in the biatrial Maze arm of the study was more than 20%. New pacemakers were required at this surprisingly high rate primarily because of postoperative heart block , not postoperative SA node dysfunction . However, it is anatomically impossible to create heart block with a surgical lesion that is not in the atrial septum because the AV node–His bundle complex resides exclusively within the atrial septum (see Chapter 3 , Fig. 3.17 ). No patient in that study had a septal lesion performed as a part of their “biatrial Maze procedure,” so the procedure itself could not possibly have caused the heart block. The actual reasons why a higher postoperative pacemaker requirement is common after a biatrial Maze procedure are discussed in more detail in Chapter 21 .
Finally, global atrial mapping has documented that during episodes of sustained AF, up to 30% of the macro-reentrant circuits present in AF are located in the RA. , This means that failure to perform the right atrial lesions in patients with sustained AF dictates a potential failure rate of up to 30% unless there is some way to determine preoperatively if the RA is actively participating in the AF. Currently, short of a full preoperative electrophysiologic study or global intraoperative mapping, there is no reliable way to distinguish between passive right atrial involvement and active right atrial involvement (i.e., the RA is “driving” the AF). Therefore, it is impossible to be certain preoperatively whether the right atrial lesions are essential to surgical success. As a result, most experienced arrhythmia surgeons routinely perform both the left atrial lesions and the right atrial lesions in all patients with concomitant persistent or long-standing AF.
Evolution of the Right Atrial Lesion Pattern
Several different right atrial lesion sets have been used over the past 3 decades, depending on new knowledge gained with experience, individual surgeon preference, or untested surgical whims. Perhaps this explains why surgeons are consistently more perplexed by the right atrial lesions of the Maze procedure than by the left atrial lesions.
In the original Maze-I procedure, there was no intercaval lesion from the superior vena cava (SVC) to the inferior vena cava (IVC) and there was no “T” lesion from the intercaval line to the tricuspid annulus. Instead, there was a lesion that extended from the SVC down the posterior right atrial wall two-thirds of the way toward the IVC, where it was then curved across the lower RA and extended down to the tricuspid annulus ( Fig. 13.5 , left upper panel ). This lesion did not prevent potential macro-reentry around the orifice of the IVC; therefore, in the Maze-II and Maze-III procedures, the SVC lesion was extended to the IVC, and a separate “T” lesion was placed from the intercaval lesion to the tricuspid valve annulus ( Fig. 13.5 , right lower panel ). The right atrial lesions in the Maze-IV procedure are only slightly different from the Maze-II and Maze-III procedures and are described below.
Left upper panel, The original Maze-I procedure had no intercaval lesion from the superior vena cava (SVC) to the inferior vena cava (IVC) and therefore, offered no protection against postoperative reentry around the IVC orifice. Instead, a lesion was placed from the SVC down the posterior right atrial wall two-thirds of the way toward the IVC, where it was then curved across the lower right atrium and extended down to the tricuspid annulus.
Right lower panel, In the Maze-II procedure, an intercaval lesion from the SVC to the IVC was added, and a separate “T” lesion was placed from the intercaval lesion to the tricuspid valve annulus. This combination of intercaval and “T” lesions were retained in the Maze-III and Maze-IV procedures. See text for further discussion.
Currently Recommended Right Atrial Lesion Pattern
The potential locations of the right atrial macro-reentrant drivers of AF in a normal-sized atrium are illustrated in Fig. 13.6 . As described in Chapter 12 , three of these drivers (1, 2, or 3), when occurring alone, can drive classic atrial flutter (see Fig. 12.3 ). Importantly, all three of these “flutter waves” use the CTI in the RA between the orifice of the IVC and the tricuspid valve annulus. This explains why a CTI lesion, which is easily created with endocardial catheters, is so effective for the treatment of classic atrial flutter. Unfortunately, a CTI lesion cannot be performed surgically without opening the RA.
Potential sites of right atrial drivers of atrial fibrillation (AF). The red circles represent the locations of the macro-reentrant circuits that can potentially serve as the right atrial drivers of AF. When present with no other macro-reentrant drivers in either atrium, driver 1, 2, or 3 is responsible for classic atrial flutter. Note that all three of these potential “flutter waves” use the cavotricuspid isthmus (CTI) between the orifices of the inferior vena cava (IVC) and the tricuspid valve, explaining why a CTI lesion is so effective in treating atrial flutter. In addition, the circuit around the orifice of the IVC also uses the CTI. See text for further discussion.
There are three other sites in the RA where reentrant circuits have been documented to occur by endocardial mapping: (1) around the orifice of the SVC, (2) around the orifice of the IVC, and (3) around the base of the right atrial appendage (RAA). The right atrial lesions in the Maze procedure are designed to prevent the development of any of these documented macro-reentrant drivers of atrial flutter and/or AF. There are three right atrial lesions of the Maze procedure that accomplish that goal: the intercaval lesion, the “T” lesion, and the RAA lesion.
