Background
Following years of preclinical research, the Cox-Maze procedure was first performed clinically in 1987 to ablate all types of atrial fibrillation (AF) and reduce the risk of stroke. , The original procedure, which is now referred to as the Cox Maze-I procedure, went through several subsequent modifications to the Cox Maze-II procedure in 1990 and the Cox Maze-III procedure in 1992 (see Chapter 14 ). All three of these first iterations of the procedure were created using a “cut-and-sew” technique, and they all demonstrated early safety and great success as determined by freedom from AF (FFAF), reduction in stroke, and improvement in quality of life. The Maze pattern of the lesions divided all potential macro-reentrant circuits (drivers) of AF and dictated that the subsequent sinus-generated impulse would activate both atria and traverse the atrioventricular (AV) node to activate the ventricles in a normal fashion. In all three iterations of the cut-and-sew Cox Maze procedure, small cryolesions were placed at the end of all incisions that terminated at a valve annulus and to create conduction block across the coronary sinus. Cryoablation has been shown to be very effective for a variety of arrhythmias, including the Wolff-Parkinson-White syndrome, ventricular tachycardia, and AV node reentry. In 1997, the cut-and-sew Cox Maze-III procedure was replaced by the minimally invasive Cox CryoMaze-III procedure in which all of the lesions were created with linear cryoprobes rather than with a scalpel and scissors. ,
Each of the lesions of the Cox Maze-III procedure is created to prevent the development of the macro-reentrant drivers that sustain AF as follows
Left Atrial Lesions
-
1.
Box lesion
-
•
Isolates 70% of the triggers that induce concomitant paroxysmal AF (PAF; Fig. 15.1 )
Fig. 15.1 Box lesion for concomitant paroxysmal atrial fibrillation (PAF). The left atrial box lesion of the Cox Maze-III procedure isolates 70% of the potential atrial triggers (focal yellow starbursts) responsible for inducing PAF that is associated with structural disorders of the left heart requiring surgery (“Concomitant PAF”). Left panel, Box lesion (blue lines) encompassing all four pulmonary veins and the intervening posterior left atrial wall between the right and left pulmonary veins. Right panel, After the creation of a box lesion, the isolated atrial triggers are still present but cannot induce AF (gray starbursts).
-
•
Prevents 30% of the macro-reentrant drivers of non-PAF that use the posterior left atrial wall between the right and left pulmonary veins (PVs) and serves as an anchor for the other lesions of the Maze procedure ( Fig. 15.2 )
Fig. 15.2 Box lesion for concomitant non-paroxysmal atrial fibrillation (non-PAF). The red circles represent the potential macro-reentrant drivers of AF, and the blue lines represent surgical lesions. Left panel, A box lesion in patients with non-PAF interrupts approximately 30% of the macro-reentrant drivers using the posterior left atrial wall as a part of their circuit. Right panel, After the creation of a box lesion, multiple sites for potential macro-reentrant AF drivers remain in both atria.
-
•
-
2.
Left atrial appendage (LAA)–left superior PV lesion across the coumadin ridge ( Fig. 15.3 )
-
•
Prevents reentry around the base of the LAA
Fig. 15.3 Left atrial appendage (LAA)–left superior pulmonary vein (LSPV) lesion across the coumadin ridge. In the Cox Maze-III procedure, a lesion is created from the LSPV to the orifice of the LAA across the coumadin ridge to prevent macro-reentry around the base of the LAA, which often uses this ridge. Left panel, Location of the LAA–LSPV lesion in the Cox Maze-III procedure. Right panel, Elimination of macro-reentry around the base of the LAA using the coumadin ridge.
-
•
-
3.
LAA closure ( Fig. 15.4 )
-
•
Isolates triggers and reentry in musculature of the LAA
-
•
Protects against subsequent stroke
Fig. 15.4 Left atrial appendage (LAA) closure with an AtriClip. Application of an AtriClip to close the LAA eliminates any triggers and any macro-reentrant drivers that use the LAA as a part of their circuit.
-
•
-
4.
Mitral line and coronary sinus lesions ( Fig. 15.5 )
-
•
Prevent postoperative peri-mitral flutter
Fig. 15.5 Mitral line and coronary sinus lesions. A serious complication of performing a box lesion is peri-mitral atrial flutter caused by a large macro-reentrant circuit that uses the anterior, posterior, lateral, and medial (septal) areas of the left atrium. To prevent this iatrogenic postoperative arrhythmia, the Cox Maze-III procedure includes the creation of a “mitral line” across the left atrial isthmus between the box lesion and the posterior mitral valve annulus and a coronary sinus cryolesion in the same plane as the mitral line (see Chapter 48 ).
-
•
Right Atrial Lesions
-
1.
Superior vena cava (SVC)–inferior vena cava (IVC) lesion ( Fig. 15.6 )
-
•
Prevents reentry around the SVC orifice and around the IVC orifice
-
•
Prevents on atrial flutter-wave that circles posterior to SVC orifice
-
•
Serves as an anchor for “T” lesion to tricuspid valve annulus
Fig. 15.6 Superior vena cava (SVC)–inferior vena cava (IVC) lesion. The right atrial lesions of a Cox Maze-III procedure include an intercaval lesion between the SVC and the IVC placed immediately anterior to the crista terminalis in the thin portion of the posterior right atrium. This lesion prevents macro-reentry around the orifices of both the SVC and IVC and prevents one of the three potential reentrant circuits of atrial flutter (i.e., the one that circles behind the SVC).
-
•
-
2.
“T” lesion ( Fig. 15.7 )
-
•
Prevents atrial flutter-wave that circles anterior to the SVC orifice
-
•
Prevents atrial flutter wave around the tricuspid valve annulus
Fig. 15.7 “T” lesion. The “T” lesion of the Cox Maze-III procedure extends from the intercaval lesion across the lower third of the right atrium down to the tricuspid valve annulus at what is traditionally called the 2 o’clock position. The “T” lesion prevents reentry around the tricuspid annulus and also another reentrant circuit that can also cause atrial flutter that circles anterior to the superior vena.
-
•
-
3.
Right atrial appendage (RAA) lesion ( Fig. 15.8 )
-
•
Prevents reentry around base of RAA
Fig. 15.8 Right atrial appendage (RAA) lesion. The final right atrial lesion in the Cox Maze-III procedure is a single lesion anchored on the “T” lesion that is created along the anterior edge of the right atrium from the “T” lesion to the tip of the RAA. This lesion prevents reentry around the base of the RAA.
-
•
Not many surgeons mastered the lesion set used for cut-and-sew operations. There were concerns by most surgeons, and there still are, about cross-clamp and cardiopulmonary bypass (CPB) time when adding a Maze procedure to other cardiac operations. The Maze procedure was a natural fit with mitral valve (MV) operations because the left atriotomy was one of the Cox Maze-III incisions, and patients with MV disease have the highest incidence of AF before mitral surgery compared with other heart operations. , Combined MV repair and Maze surgery was first reported in 1993 from the Cleveland Clinic. Still today, MV surgery is the most common concomitant operation with a Maze procedure. , , The specific manner in which we modified the Cox CryoMaze-III procedure to create all of the cryolesions of the Cox CryoMaze-III procedure described above in less time is described in Chapter 27 .
In the “Maze-IV” procedure, bipolar radiofrequency (RF) and cryoablation were used to replace the incisions of the cut-and-sew Cox Maze-III and the cryolesions of the CryoMaze-III procedures. , Although the lesions were created in a different manner, the ultimate lesion pattern was electrophysiologically identical to the Maze III procedure (see Chapter 17 ). After US Food and Drug Administration approval of the Synergy bipolar RF clamp (AtriCure Inc.) in November 2012, it became easier to teach surgeons to perform the Cox Maze-IV lesion set than it had been trying to teach them the Cox Maze-III procedure. This resulted in an increase in concomitant ablation and more standardization of the lesion set. Other technologies (microwave, laser, high-intensity-focused ultrasonography) appeared in practice and faded away because they did not consistently provide transmural lesions and effectiveness was not acceptable (see Chapter 20 ). However, the operation was still complex, so many violations of the basic principles of the Maze procedure followed in an effort to simplify the procedure. Besides disagreement about the lesion set and which technology was best, there were also unanswered questions and strong opinions about which patients were candidates for AF surgery. The excessive limitation of ablation by excluding patients with a long AF duration, very large atria, and combination procedures such as multiple valves, valve and coronary bypass, and valve and aneurysm resulted in higher success rates but left the majority of patients untreated and therefore, deprived of the long-term benefits of the surgery. In addition, the Ablate-AF trial and similar studies that indicated an increased need for new pacemakers and prolonged hospitalization following AF ablation appeared just as “length of stay” was becoming a publicly reported quality metric. , , , Many surgeons hesitated. They needed more assurance that there were clinically important benefits to adding a Maze procedure to routine coronary artery bypass graft (CABG) procedures and valve surgery and that risks could be minimized, efficiency improved, and the lesion set made easier to understand.
Rationale for the Modification of the Cox CryoMaze-III Procedure
At Northwestern University, we revamped the cardiac surgery program in 2004 into the Bluhm Cardiovascular Institute, which launched on Valentine’s Day 2005. An important aspect of the change was to create a database to track the outcomes of our cardiac surgery operations, not for just 30 days as required by the Society of Thoracic Surgeons (STS) but annually for years after surgery. Surgery for AF was a particular focus of the database creation, and many “custom” fields with details about the type and duration of AF, ablation techniques, and detailed follow-up were added to the basic STS information. We also created a team of nurses and research staff to maintain high-quality tracking. Early in our experience, we used the “cut-and-sew” Cox Maze-III procedure for patients with a dilated left atrium (LA; >6 cm), for all stand-alone AF surgery, and in young patients in whom we wanted to maximize the chances of success. Results were similar to our published series from the Cleveland Clinic. We had published an overview of the bipolar RF technology, and some of the operations included cryoablation using reusable probes. The reusable cryoprobes were stiff and had limited surface areas for ablation ( Fig. 15.9 ), which was satisfactory for short applications but not efficient in replacing linear ablation lines. As a result, we began using the new disposable cryoprobes in patients undergoing reoperative MV surgery. We started with reoperations because the epicardial adhesions essentially eliminated the ability to use a bipolar RF clamp. The extensive dissection around the left PVs, the left atrial dome, and posterior left atrial wall required for placement of a bipolar clamp was not an essential part of the redo MV operation. In addition, the optimal number of bipolar RF applications necessary to create a reliable transmural lesion was an ongoing debate at the time. It took years for the current routine of multiple “doublets,” clamping without release with two ablations, was developed, and even though this technique showed marked improvement experimentally, the clinical results remained suboptimal.
Examples of some of the reusable cryoprobes that were available in the 1990s to create cryolesions in the heart. They were inflexible, and most were developed to perform focal cryolesions at specific sites in the atrium or ventricle. The cryoprobe on the left was the first linear cryoprobe ever developed (c. 1998), but it was quite cumbersome to use to create linear cryolesions. (Reproduced from Lee AM, Clark K, Bailey MS, Aziz A, Schuessler RB, Damiano Jr. RJ. A minimally invasive Cox Maze procedure: operative technique and results. Innovations . 2010;5(4):281–286.)
Instead, we began treating reoperative patients with endocardial cryoablation only using the new disposable commercially available devices ( Fig. 15.10 ). These longer flexible probes allowed the surgeon to create long linear ablation lines with each application, which saved CPB time and aortic cross-clamp time. In addition, the disposable cryoprobes were malleable and therefore allowed for the creation of curved linear lesions, which was a major advantage when a lesion was placed across the dome of the LA, around the mitral annulus, and in other irregular areas (see Chapter 27 ). At that time, we were publishing our Northwestern Maze results regarding FFAF, and the number of patients undergoing reoperations who were treated with cryoablation was significantly less, We had expected the results in the cryoablation group to be worse because cryosurgery was used only in reoperations, so we were surprised to find that in midterm follow-up, there was no difference in FFAF or stroke rate between the RF and cryosurgical groups. Therefore, we developed a standardized cryoablation lesion set that followed the Cox Maze-III pattern, which we knew was highly effective. In addition, the new cryoprobes allowed an efficient operation that minimized cross-clamp and bypass times to reduce renal dysfunction from prolonged bypass. In addition, proper placement of the cryoablation lesions minimized the risk of pacemakers.
