Atrial fibrillation (AF) can cause multiple adverse sequelae, including palpitations, decreased cardiac output, systemic thromboembolism, and tachycardia-induced ventricular cardiomyopathy. Typically, the first symptoms of de novo AF are palpitations and the consequences of decreased cardiac output such as fatigue, shortness of breath, and dyspnea on exertion. These symptoms are usually worse in patients with paroxysmal AF (PAF) than in patients who are in AF continuously. This paradox is likely because patients who have intermittent bouts of AF are simply more aware of their symptoms during AF because of the noticeable discrepancy between how they feel while in AF versus how they feel when they are not in AF. Although systemic thromboembolism occurs with equal frequency in patients with PAF and persistent AF, tachycardia-induced ventricular cardiomyopathy occurs almost exclusively in patients with long-standing persistent AF (LSpAF) and poor rate control because of the time required for the ventricular cardiomyopathy to develop.
The primary objectives of catheter ablation for AF and surgical ablation for AF are fundamentally different; therefore, they can have different impacts on postprocedure atrial function. Catheter ablation is designed to isolate the triggers located primarily in and around the pulmonary vein orifices that are responsible for inducing AF episodes. The success of catheter ablation in restoring left atrial function depends largely on the extent of preexisting atrial fibrosis and the degree of additional scarring induced by the ablation procedure. Surgical ablation is designed to ablate the macro-reentrant circuits that sustain AF by creating a series of strategically placed linear lesions in the atria. Although the surgical Maze procedure has a higher success rate in achieving sinus rhythm, it is inherently more invasive than catheter ablation, raising concerns about its potential impact on postoperative atrial function. The degree of left atrial dysfunction after the Maze procedure has been associated with multiple factors such as the extent of preoperative atrial scarring, the number of surgical lesions created, and other preoperative conditions involving the left atrium (LA). Like patients having catheter ablation, surgical patients with preserved preoperative atrial function and minimal fibrosis are more likely to retain better atrial function postoperatively. Additionally, sinus rhythm restoration after the Maze procedure plays a pivotal role in promoting the functional recovery of both atria.
The ability to assess atrial function in an accurate and reproducible manner has evolved significantly with advancements in echocardiographic techniques. Speckle-tracking echocardiography (STE) has become a key tool for quantifying atrial strain and identifying early functional impairment (see Chapter 7 ). Left atrial reservoir strain and pump strain have proven to be sensitive markers of atrial function, and they can be used as predictors of AF recurrence after both catheter ablation and the Maze procedure.
Experimental Evaluation of Left Atrial Function After the Maze Procedure
As mentioned in Chapter 1 , there was initial concern that the multiple transmural atrial lesions of the Maze procedure might devascularize the atrial myocardium. Therefore, postmortem evaluation of the atria after experimental Maze procedures included injecting the right coronary artery with blue dye and the left main coronary artery with red dye to document that the coronary blood supply to the atria remained intact postoperatively ( Fig. 22.1 ). In addition, histologic studies of both atria failed to show any ischemic injury in either atrium after the experimental Maze procedure. After the safety and efficacy of the Maze procedure was confirmed in animals, major efforts were expended to determine the effect of the procedure on both right and left atrial function.
Photo of the right and left atria after an experimental canine Maze procedure. The right coronary artery was injected with blue dye, and the left main coronary artery was injected with red dye to determine if the lesions of the Maze procedure interfered with the blood supply of the atria. The tests showed that the coronary artery blood supply of the atria was not affected by the Maze procedure. Histologic examination of random tissue samples from both atria also confirmed the absence of atrial myocardial ischemia after the Maze procedure.
In the first series of experiments, an electromagnetic flow probe was placed around the ascending aorta for continuous measurement of the cardiac output (minus coronary artery flow) in dogs that had undergone a Maze procedure. Atrioventricular (AV) sequential pacing was instituted at a constant rate and optimal A-V interval, and the cardiac output was allowed to stabilize for several minutes ( Fig. 22.2 ). The atrial pacing wire was then suddenly disconnected from the pacemaker to convert from AV sequential pacing to ventricular pacing alone at the identical rate. The cardiac output immediately decreased by approximately 16% with the loss of synchronous atrial activation. After the cardiac output had stabilized with ventricular pacing, the atrial pacing wire was reattached to the pacemaker to restore AV sequential pacing, and the cardiac output returned to its previous level. This series of experiments clearly demonstrated that the atria functioned well after a Maze procedure.
Hemodynamic evaluation of atrial function after a canine Maze procedure. An electromagnetic flow probe was placed around the ascending aorta for the continuous measurement of cardiac output (minus coronary blood flow). The cardiac output was approximately 16% better in the presence of synchronous atrial contraction during atrioventricular sequential pacing than it was without synchronous atrial contraction during ventricular pacing at the same rate.
Although these experiments documented that the Maze procedure had no significantly adverse effect on overall atrial function, they did not exclude the possibility that the LA was adversely affected by the Maze procedure. Lessons learned from the left atrial isolation procedure years before were extremely important in the accurate assessment of the effects of the Maze procedure on atrial function. For example, we had documented that after isolation of the LA, the loss of all left atrial function had no impact on forward cardiac output in the presence of normal right atrial function, normal right atrial-right ventricular synchrony, and a normal left ventricle (LV) ( Fig. 22.3 ). Thus we recognized that the experimental findings of preserved overall atrial function after the Maze procedure did not rule out the possibility that the procedure had injured the LA but not the RA. Therefore, we performed two-dimensional echocardiography and Doppler flow analysis after the Maze procedure specifically to evaluate postoperative function in the LA. Doppler flow across the mitral valve in the control group demonstrated excellent A-waves, indicating that the canine LA was functioning normally. Similar A-waves were consistently observed after the Maze procedure, indicating that the LA continued to function postoperatively ( Fig. 22.4 ).
Evaluation of the impact of left atrial contraction on multiple hemodynamic factors. The cardiac output (minus coronary blood flow) was continuously recorded with an electromagnetic flow probe around the ascending aorta. Systemic blood pressure (BP) was monitored continuously with an indwelling arterial cannula. Left ventricular end-diastolic pressure (LVEDP) was monitored continuously with a high-fidelity Millar catheter in the left ventricle. Continuous pulmonary artery (PA) pressure was monitored with a Swan-Ganz catheter located in the main PA.
Left panel, After surgical electrical isolation of the left atrium (LA) from the rest of the heart, normal sinus rhythm (NSR) was reproduced by pacing the right atrium and LA with a 40-ms delay in pacing of the LA. Activation of the ventricles occurred normally through the AV node–His bundle complex. This precisely reproduced the electrical activation sequence of the heart during normal sinus rhythm in someone who had not undergone a left atrial isolation procedure.
Right panel, The left atrial pacing wire was then suddenly disconnected, and the LA became electrically silent. Note that the complete loss of left atrial contraction did not affect the forward cardiac output, systemic BP, LVEDP, or PA pressure. (See text for further discussion.)
Echocardiographic evaluation of left atrial function after a canine Maze procedure. (A) Transmitral Doppler flow in a normal dog. (B–D) Transmitral Doppler flow in three separate dogs after undergoing the Maze procedure. Persistence of the A-wave after the Maze procedure confirms the presence of post-Maze left atrial function. ECG, Electrocardiogram.
Clinical Evaluation of Left Atrial Function After the Maze Procedure
Beginning with the first patient who underwent the Maze procedure, we performed certain intraoperative tests to evaluate atrial function soon after discontinuing cardiopulmonary bypass. Repeated Swan-Ganz thermodilution cardiac outputs were recorded during atrioventricular (AV) sequential pacing versus ventricular pacing, and transesophageal echocardiography (TEE) was used to record evidence of right atrial and left atrial contractile activity. Perhaps more important than either test was to simply look at both the RA and LA and watch them contract. The LA was sometimes slow to recover its capacitance capabilities, but in virtually every patient, the left atrial myocardium could be seen to contract briefly just before ventricular systole. I made a habit of bringing my electrophysiology colleagues, who had been involved in the pre-Maze mapping process, back into the operating room specifically to confirm visually that the atria were contracting. All patients were in normal sinus rhythm or being A-paced or AV sequentially paced during these assessments, and all showed evidence of almost immediate return of atrial function. In view of the well-known delay in atrial function after successful electrical cardioversion of AF to sinus rhythm, we were quite surprised to see atrial function return so soon after being surgically converted from AF to sinus or A-paced rhythm. We still have no explanation for this unexpected phenomenon except that perhaps unlike surgical conversion, electrical cardioversion might “stun” the atrial myocardium, which then takes more time to recover.
The first 69 patients undergoing the Maze-I ( n = 32), Maze-II ( n = 14), or Maze-III ( n = 23) procedure had the following tests and evaluations performed 6 months after surgery:
-
1.
Complete endocardial catheter electrophysiology study that included the identical programmed electrical stimulation protocol and burst-pacing protocol used preoperatively in an effort to induce AF. These stimulation protocols were performed both at rest and during isoproterenol infusion.
-
2.
Swan-Ganz thermodilution cardiac output determinations during AV sequential pacing versus ventricular pacing. Cardiac output was recorded as the average of three separate thermodilution outputs performed during AV sequential pacing and during ventricular pacing
-
3.
Standard 12-lead electrocardiography (ECG)
-
4.
24-hour Holter monitor
-
5.
Dynamic magnetic resonance imaging (MRI) scans of the heart
-
6.
Transthoracic echocardiography (TTE)
-
7.
TEE if the TTE showed no apparent left atrial function
AF could not be induced in any of the 69 patients 6 months after having a Maze procedure. In a few patients, self-terminating atrial flutter was induced briefly during the infusion of isoproterenol, but it did not occur clinically either before or after the 6-month electrophysiology study. As observed in the experimental studies, cardiac output was consistently higher during AV sequential pacing than during ventricular pacing at the same rate, indicating the presence of atrial function. No AF was seen on either the 12-lead ECGs or on the 24-hour Holter monitors. Our electrophysiologists, Drs. Bruce Lindsay and Michael Cain, stopped performing the electrophysiology studies after the first 69 patients because they believed that the extensive efforts to induce AF in patients after a Maze procedure were “an unnecessary inconvenience to the patients and waste of our time and resources because the operation clearly works.”
The results of the thermodilution cardiac outputs, dynamic MRI scans, and TTEs that were performed to evaluate postoperative atrial function did not always coincide. As a result, we took the position that if any one of these tests showed positive atrial function, then the results would be considered positive. Using this criterion, right atrial function was documented in 98% of patients, and left atrial function was documented in 93% of patients. Nine patients showed no apparent left atrial function in any of the tests, including the TTE. We proceeded with a TEE in these 9 patients, and definitive left atrial function was documented in 8 of them. This suggested that TEE was more sensitive than TTE in the assessment of left atrial function.
Over my (JLC) objection, our group published an article in 1994 describing only the Doppler echo results in 46 of the Maze patients and showed that using Doppler echocardiography only, right atrial function was present in 83%, and left atrial function was present in only 61%, even though many of these same patients were shown to have biatrial function by one of the other tests, particularly the dynamic MRI scans. Thus this article resulted in a profound misconception that suggested that the LA “didn’t work” in more than one-third of patients after the Maze procedure, when in fact, left atrial function was documented by at least one test in 93% of them. Another 1997 study in 31 Maze patients, which was also confined to the evaluation of atrial function by Doppler echocardiography only, showed right atrial function in 81% and left atrial function in 71%. This study noted that the mitral A-wave was “lower” after the Maze procedure, suggesting to the authors that the restored atrial function after a Maze procedure was “less than” that in control patients despite a significant decrease in both left atrial and right atrial sizes with no left atrial fractional area change postoperatively, though it remained below normal in both atria.
In 2008, Damiano and colleagues reported a superb experimental study in pigs in which a group undergoing a Maze procedure via a median sternotomy and pericardiotomy was compared with a sham group undergoing only a median sternotomy and pericardiotomy. Postoperative left atrial function, including left atrial reservoir, conduit, and contraction parameters, was assessed by load-sensitive MRI and load-independent conductance catheter techniques. Their findings showed that “many of the previously-observed changes in LA function following the Cox Maze procedure are due to the median sternotomy and cardiotomy.”
Several studies indicate that the restoration of atrial pump function after a Maze procedure is affected by comorbidities and the duration of AF. In patients undergoing a Maze only or in conjunction with coronary artery bypass graft surgery, atrial function typically recovers quickly. However, patients with long-standing degenerative mitral regurgitation can have both global and regional atrial dysfunction that are exacerbated by preexisting conditions such as scar tissue, which can result in irreversible remodeling of the LA, resulting in an elevated risk of recurrent AF and stroke.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
