Fig. 5.1
Normal heart anatomy with trabeculated left atrial appendage
A review of the literature elucidates that 70 % of all strokes in patients with AF are cardioembolic in nature [5]. In a landmark paper by Blackshear and Odell [6], the authors analyzed 23 studies that evaluated the left atrium and LAA in patients with AF. When comparing valvular AF vs. non-valvular AF, 254 of 446 patients (57 %) thrombi in patients with valvular AF were localized to the LAA, while in 201 of 222 patients (91 %) thrombi in non-valvular AF were isolated to, or originated in the LAA (p < 0.0001) [6]. In another study, researchers evaluated 233 patients with AF who did not undergo anticoagulation [7]. Transesophageal echocardiography showed that 34 patients (15 %) had a left atrial thrombus, and 33 of 34 (97 %) of these thrombi were originating in the LAA. Leung et al. examined a cohort of 272 non-valvular AF patients [8]. In their analysis, the authors found an 8 % incidence of having a left atrial clot, with 100 % of the clots isolated in the LAA [8].
Numerous studies have shown that the LAA is a chief source of thrombus in valvular and non-valvular AF [5]. It is no surprise then, that mechanical occlusion of this LAA, an old idea [9], has received renewed interest in decreasing the risk of stroke in patients with AF.
Excluding the Left Atrial Appendage from Systemic Circulation
In the 1940s [9], the feasibility of resection of the LAA had been studied in dogs, and the operation was suggested as a therapeutic approach to reduce the burden of arterial emboli in patients with AF and rheumatic valve disease. The first report of LAA occlusion in an animal model was by Hellerstein and associates in 1947 [9]. Hellerstein performed atrial appendectomy on 8 canine hearts. Post mortem studies on the canines revealed complete endothelialization at the atrial anastomosis, and the procedure was considered a viable option in patients with rheumatic mitral valve disease and recurrent emboli [9].
The first ever reported resection of the LAA in a human was by John Madden in 1949 [9]. Madden theorized that the “ideal prophylaxis for recurrent arterial emboli should be the removal of the thrombus together with its site of origin [9].” His colleague, Dr. William Dock in 1946 had suggested resection of the LAA in cases of rheumatic heart disease with AF. Madden’s first patient was a 38-year-old female, admitted with embolic occlusion of the left iliac artery and rheumatic mitral stenosis. In the days before the heart–lung bypass machine, the operation involved stopping the heart and immediate manual massage of the heart concomitant with artificial respiration. Under these conditions, the LAA was resected. However, postoperatively, the patient was found to have a left hemiparesis, thought to be secondary to a right cerebral embolus. Eight months post surgery, the patient continued to improve from the stroke symptoms, and Madden noted that there was “no recurrence of peripheral arterial emboli [9].” His second patient, a 52-year-old male with rheumatic mitral stenosis and AF, was admitted with an occlusion of the aorta at its bifurcation and underwent LAA resection. Postoperatively, the patient improved until the ninth day, when he suddenly died. Autopsy was not performed, and cause of death was not investigated [9]. In his discussion, Madden admitted at the operation was “not a standard one” and that a satisfactory technique needed to evolve [9]. In conclusion, Madden invited further study and clinical evaluation of patients in order to select those patients that would benefit most from the procedure. However, surgery for AF would first more intensely focus on rhythm control rather than the LAA alone.
Atrial Fibrillation Surgery
The burden of AF arises from both the thromboembolic risk, and the medications used for rate or rhythm control. The first report of corrective surgery for AF was published in 1980 by Dr. Williams in The Journal of Thoracic and Cardiovascular Surgery [10]. The study was designed to develop a technique to isolate the left atrium electrically from the rest of the heart. Ten dogs were subjected to cryoablation of the bundle of His, and initial results indicated that postoperatively all animals maintained normal sinus rhythm [10]. However, longer-term follow up showed that the LA could still fibrillate and continued to pose a thromboembolic threat.
In 1985, Guiraudon and colleagues reported a novel technique, named the “corridor” operation , for patients with disabling AF despite optimal medical therapy [11]. The open-heart procedure involved isolating a “corridor” of atrial tissue from the right and left atrium. The authors reported that the corridor consisted of isolating the left atrium electrically from a corridor containing the sinus node, AV node, and the connecting right atrial mass. They theorized that this corridor would re-route the electrical macro re-entrant circuits in AF and preserve the physiological driving mechanism of the ventricle. However, this technique failed to eliminate AF or reinstate atrial contractility [11]. As such, the risk of thromboembolism remained, and the procedure was soon abandoned.
Cox Maze
In 1987, Cox et al. described a new surgical procedure for the treatment of AF [12]. Cox hypothesized that AF was based on “migrating wavelets of macroreentry” [13] and his Maze operation combated this by compartmentalizing both atria. The Cox Maze procedure has undergone several iterations and has become the cornerstone of AF surgical treatment. In a Cox Maze procedure, the electrical impulse is guided along a surgically determined course, and in theory halts AF. This is done by both incising and sewing in the atria in a “maze like pattern,” or through biatrial “lesion sets” via radiofrequency ablation lines (Maze IV). Lastly, the atrial appendages are surgically removed or obliterated. The Cox Maze procedure has been shown to be successful in preventing re-entry AF, and restoration of sinus rhythm [14]. In 1999, Cox published his 11.5-year results of performing the Maze procedure on 306 patients [15]. They reported only 2 perioperative strokes (0.7 %), and in the 265 patients followed for up to 11.5 years, there was only 1 late minor stroke [15]. Cox concluded that the low stroke rate observed in his cohort was secondary to the restoration of sinus rhythm and removal of the LAA, “where most thrombi develop” [15].
However, several criticisms of Cox’s results have been presented since publication. The 2012 Heart and Rhythm Society (HRS) recognized that patients in the study did not undergo follow up by present standards [16]. Few patients in the study had any regular monitoring of rhythm and most of the rhythms were documented only via telephone or mailed questionnaire. This may have led to underestimating the number of patients in paroxysmal or persistent AF. A recent systematic review examining the surgical treatment of AF [17] via “cut and sew” Maze or alternative energy sources found the procedures to be successful only 78.3 and 84.9 % ( p = 0.03) of the time, respectively. The HRS examined the evidence for surgical ablation of AF and concluded that more prospective multicentre trials are needed to better define the safety and efficacy of the surgical techniques [16]. Additionally, there is no robust data that demonstrates that AF ablation decreases stroke risk; the HRS guidelines, thus, do not recommend discontinuation of anticoagulation therapy after ablation. Currently, the guidelines recommend surgical AF ablation only for symptomatic AF patients, who are refractory to antiarrhythmic medications, and undergoing surgery for other indications (Class IIa, Level of Evidence C) [16]. Stand-alone surgical ablation of AF is even less warranted, and should only be considered in patients who have failed one or more attempts at catheter ablation (Class IIb, Level of Evidence C). The committee, however, did recommend LAA occlusion during surgical AF ablation whenever feasible, as this was thought to decrease stroke risk. Further studies into long-term outcomes and stroke risk were encouraged, but the topic of LAA occlusion was brought to the forefront.
Surgical Techniques for Left Atrial Appendage Occlusion
Over the years, surgeons have proposed several techniques to close the LAA. The principle behind LAA occlusion involves either excluding the LAA from circulation, or excising the LAA altogether. The techniques include (1) Simple neck ligation, (2) Purse-string technique, (3) Endocardial suturing, (4) Envagination and double suture technique, (5) Surgical amputation and closure, (6) Surgical Stapler, and (7) Novel occlusion devices (Table 5.1). Although many of these techniques have been practiced for decades, there is little data comparing the approaches. The only paper to whose objective was to directly compare techniques was published by Kanderian in 2008 [18]. In their landmark paper, the authors studied 137 patients who underwent surgical LAA closure. A key finding of the study was that only 40 % of the patients (n = 55) had successful LAA closure. Unsuccessful closure was defined by the presence of a patent LAA (persistent communication between LAA and left atrium) or LAA stump >1 cm. Out of these patients (n = 55), successful LAA closure was seen in 38 patients with excision of the LAA (73 %) vs. 17 patients with suture exclusion (20 %) and 0 patients with stapler exclusion (0 %) (p < 0.001). In analyzing the unsuccessful LAA closures, Kanderian et al. [18] found that of all 12 patients with stapler exclusion, 58 % of them had a persistent LAA stump >1 cm. The authors noted that the presence of a residual LAA stump greater than 1 cm has been classified as unsuccessful closure in literature [18], and this can theoretically pose a risk for harboring thrombus. Out of 73 patients who had suture exclusion of the LAA, 61 % of patients had persistent flow into the appendage via the stump. The implication is that persistent flow into the LAA after exclusion indicates persistent communication [18], and thus thrombi can still develop over time. In these patients the risk of stroke may be even higher post surgery because the LAA is more prone to thrombose when partially closed as blood would remain more stagnant [18]. In their series, the prevalence of LAA thrombus in appendages with persistent flow was high (46 % suture exclusion, 67 % stapler exclusion). The study authors therefore concluded that excision of the appendage is the most reliable method of LAA closure, and recommended that anticoagulation not be discontinued without echo investigation post surgery. The common surgical techniques for occlusion are described below.
Table 5.1
Techniques for surgical left atrial appendage occlusion
Surgical technique | Important aspects |
|---|---|
• Simple neck ligation | – Most basic of all techniques |
– Challenging when large or irregular LAA base | |
• Purse-string technique | – May lead to puckered suture line and subsequent thrombus formation endocardially |
– High rate of LAA recanalization | |
• Endocardial suturing | – Technique for closing the LAA from within the open LA |
– High rate of unsuccessful closure | |
• Envagination and double suture technique | – Double suture technique requires more time |
– Good postoperative results | |
• Surgical amputation and closure (cut-and-sew method) | – Important to de-air left heart prior to release to aortic clamp with this method |
– Good postoperative results | |
• Surgical Stapler | – Stapler more costly than sutures |
– Studies have shown fewer leaks across closure line when using a stapler over ligation technique | |
• Atriclip surgical device | – Device more costly than sutures |
– Easy to use | |
– Can be applied thoracoscopically | |
• TigerPaw surgical device | – Device more costly than sutures |
– Easy to use | |
– Can be applied through small thoracotomy | |
• Endoloop snare | – Can be applied thoracoscopically |
– Small study (n = 8) showed increased risk of adverse | |
– events (serious bleeding) | |
• LigaSure TM vessel sealing system | – Bipolar device to weld the LAA shut |
– Small study (n = 12) showed no adverse events but authors advocated for large-scale studies |
Simple Neck Ligation
This exclusion technique is perhaps the most basic of all the techniques. Typically the surgeon places a vascular clamp around the neck of the LAA. Following this, a 1-0 or 0 silk tie is placed around the clamp and the suture tied down. Once a knot has been placed, the vascular clamp is removed, and the remaining knots tied down to secure the suture in place. The aim of this technique is to tie off the LAA at the base from the outside of the atrium. Difficulty arises when a large LAA, a wide neck, or an irregular LAA morphology is encountered.
Purse-String Technique
This exclusion technique involves closing the LAA via purse-string from the outside using 4-0 or 3-0 prolene. The suture is placed across the base of the LAA, and usually requires 4 or 5 passes to encircle the base. Surgeons may use Teflon pledgets to prevent tearing and distribute the tension evenly, as the LAA tissues can be fragile. Although no strong evidence exists, purse-string sutures may lead to a “puckered” suture line [19]. This has been theorized to lead to thrombus formation on the endocardial surface, thus leading to a higher stroke risk [19].
Small cases series have been published using this technique. One such study was by Lynch et al. [20], which reported a high incidence of recanalization of the LAA after purse-string closure of the LAA. On follow up echo, all patients (n = 6) demonstrated communication between the LAA and the atrium proper, mainly at the superior-lateral border of the appendage. The authors hypothesized that flow into an incompletely ligated LAA could lead to tension along the suture lines, and further breakdown of the LAA inlet, with complete reopening of the closed orifice. The authors recommended that techniques other than purse-string suture be adopted for LAA closure.
Endocardial Suturing
This exclusion technique involves closing the LAA from within the open left atrium. Schneider et al. [21] reported a series of 6 patients who underwent surgical LAA closure at the time of mitral and/or aortic valve surgery. The authors described a technique of endocardial LAA closure. To accomplish this, a double row of 3-0 prolene sutures are placing in a running fashion, sewn from within the left atrium (Fig. 5.2). The authors investigated LAA closure via echo post surgery, and reported that 83 % of patients had unsuccessful closure. No major complication or mortality was reported, but the authors acknowledged that surgical LAA occlusion using endocardial methods may lead to incomplete closures.
Fig. 5.2
Excluded left atrium via endocardial suturing (internal view). Note the suture line
Envagination and Double Suture Technique
This exclusion technique was described by Hernandez-Estefania et al. [22]. The authors describe that the LAA is completely invaginated into the LA with forceps. Then a 4-0 prolene purse-string suture is carefully sutured on to the base of the appendage. The author points out that it is important to encircle the oval shape of the inlet orifice. The LAA is then gently pulled outward with forceps, whilst pulling on the two ends of the prolene suture. After this layer is tied up, a second running suture is then placed along the long axis of the LAA. The authors report that in their experience, LAA invagination followed by double suture technique facilitates the exposure of the real LAA rim with subsequent closure and complete occlusion. All patients in their series (n = 8 patients) had echo post surgery that demonstrated no flow inside the LAA or residual stump. No major complication or mortality was reported.
Surgical Amputation and Closure (Cut and Sew)
This excision technique involves amputating the LAA once the heart is arrested. Following this the LAA is completely excised until no trabeculated portion remains, and the neck of the LAA is sewn closed (Fig. 5.3). In theory, this would not only remove the trabeculated portions of the LAA where clots may form, but also remove any ability for the LAA to recanalize. Another term for this technique is left atrial appendectomy . It is important to de-air the left cardiac chambers prior to release of the aortic cross-clamp with this approach.
Fig. 5.3
Epicardial suture line and closure of LAA (external view)
The largest series of appendectomy in published literature was from Johnson et al. [23], who reported 391 patients who underwent the procedure. The authors did not report any major complication, or any mortality owing to appendectomy.
Surgical Stapler
This approach can be an excision or exclusion technique of the LAA. Initial experience with surgical stapling of the LAA demonstrated issues with bleeding from the staple line. To avoid this, Gillinov and associates [24] from the Cleveland Clinic have called for a pericardial buttressing of the LAA staple line.
Most commonly after cardioplegic arrest, an Endo GIA II stapler with 4.8 mm staples is used to exclude the LAA [24]. The positioning of the staple is described as being parallel to the base of the LAA, approximately 5 mm from the circumflex coronary artery (Fig. 5.4). On the device itself, the staple line is buttressed with a bovine pericardial strip. In their series of 222 patients, Gillinov et al. [24] report 10 % of patients requiring additional pledget-supported sutures for tears caused by the stapler. Reoperation for bleeding was seen in 7 patients (3 %) but none of these were attributable to the LAA staple line. No mortalities or any other major complications were reported.
Fig. 5.4
Surgical stapler placed over base of LAA (left) and subsequent removal of appendage (right). Note the staple line (right)
Healey et al. [25] examined factors for successful LAA closure in their LAAOS trial. The authors noted that there was a trend for higher occlusion rates when using a stapling device over ligation technique. Use of a stapling device was associated with complete occlusion in 72 % of patients vs. 45 % of patients using ligation alone (p = 0.14). Additionally, with the use of a stapler device, all incomplete occlusions were due to residual cul de sac. That is, there was no persistent flow across the staple line from the LA body to the LAA when compared to the ligation technique (p = 0.0001).
Novel Techniques for LAA Occlusion
In recent years, industry has had increased interest in the field of LAA occlusion and new methods and devices have been developed. These new techniques aim to make surgical LAA occlusion easier, and decrease the risk of incomplete closure.
Atriclip
This novel device recently received FDA approval for occlusion of the LAA under direct visualization, and in conjunction with other open-heart procedures (510(k) K093679). The early results of a multicentre trial (7 centers) have been published [26]. Ailawadi et al. studied a total of 70 patients who underwent elective cardiac surgery and received a concomitant AtriClip (Atricure Inc, Westchester, Ohio) [26]. Efficacy of LAA exclusion was assessed intraoperatively, and at 3 months of follow up. No patients were reported to have any adverse events related to the device, and no perioperative mortality was reported. Out of the 62 patients who had 3-month follow up, 60 patients (98.4 %) had successful LAA exclusion via Atriclip. The authors concluded that safe and atraumatic exclusion of the LAA could be achieved via the Atriclip, but urged for further studies. Additionally, the authors also commented that the device could be used in conjunction with minimally invasive procedures. The device comes in three sizes (35, 40, and 50 mm), can be applied at any point either thoracoscopically or after sternotomy on or off bypass.
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
