Atrial fibrillation (AF) affects 1% of the general population and about 11% of patients undergoing cardiac surgery, and it is an important risk factor for stroke. Most ischemic strokes in patients with AF are cardio-embolic in nature, with the majority originating in the left atrial appendage (LAA). Anticoagulation is an effective approach to stroke prevention, but it requires lifelong adherence and its use is limited by bleeding. Surgical occlusion of the LAA has been shown to be an effective complement to antithrombotic therapy and is also effective as a stand-alone treatment.
Oral anticoagulation (OAC) reduces the risk of stroke in patients with AF, and therefore it is recommended for most patients with AF. Nonetheless, contemporary registries show that only about half of high-risk patients with AF receive OAC. For those started on anticoagulation, discontinuation rates exceed 50% at 3 years. Temporary and permanent anticoagulation discontinuation occurs for a variety of reasons, including surgery, illness, bleeding, and so on. The main shortcoming of anticoagulation is that maintaining anticoagulation treatment over long periods for a lifelong stroke risk is difficult and inconvenient.
Physical elimination or exclusion of the LAA is an alternative approach to stroke prevention in AF that does not suffer from the limitations of pharmacologic therapy, including noncompliance, discontinuation, and ongoing risk of bleeding. After it has been performed, the effects of the occlusion are permanent and do not require patient compliance. Despite first being described in 1948, randomized trials did not begin until 2001.
Percutaneous Left Atrial Appendage Occlusion Trials
The first randomized trials that informed clinical practice were completed in the transcatheter LAA occlusion (LAAO) field. We limit our discussion to the trials using the Watchman device because these trials have most advanced the evidence in that field. We have learned that LAAO is likely beneficial as an alternative to warfarin in patients with AF from randomized controlled trials (RCTs) of percutaneous device closure (PROTECT-AF and PREVAIL ). A meta-analysis of these trials shows noninferiority of the device compared with warfarin. However, much of the benefit of LAAO with a device compared with warfarin is due to the absence of intracranial bleeding, which is often seen with warfarin. The effect of LAAO on ischemic stroke remained to be determined and was suggested to be inferior to warfarin based on that meta-analysis. Important differences exist between device-based and surgical LAAO, which can be performed at the same time as cardiac surgery for another indication or as a stand-alone procedure. Device-based LAAO is costly and associated with procedural risk. As such, it is mostly used for the subset of patients who cannot receive anticoagulation because of very high bleeding risk. Furthermore, patients enrolled in device trials were very different from the surgical patients with AF who have higher CHA 2 DS 2 -VASc scores and who often have valvular heart disease. (These patients were excluded in the percutaneous trials, and they may have more frequent thrombi outside of the LAA. ) Finally, whereas surgical occlusion is an extravascular procedure, endovascular device occlusion carries the risk of device-related thrombus formation and embolism. Therefore, the results from percutaneous LAA closure cannot be extrapolated to surgical LAAO and vice versa. In summary, the device-based trials provided proof of concept that LAAO may be beneficial but did not address patients with AF who undergo cardiac surgery, did not inform on the combination of LAAO and medical prophylaxis, and did not evaluate a surgical intervention that can be offered to millions of patients as concomitant therapy with cardiac surgery at virtually no additional cost or risk.
Surgical Left Atrial Appendage Occlusion Trials
The RCT evidence that LAAO reduces the risk of ischemic stroke in AF comes from the surgical trial Left Atrial Appendage Occlusion Study (LAAOS) III. This seminal trial was part of a program established by researchers at McMaster University, Hamilton, Canada.
Left Atrial Appendage Occlusion Study I
The LAAOS research program started with LAAOS I, which was a two-center pilot trial to demonstrate the safety of surgical LAAO. The trial randomized 77 patients to LAAO at the time of cardiac surgery versus no occlusion and reported that surgical occlusion could be performed with a low rate of adverse events. LAAOS I included patients with and without AF at increased risk of stroke. This trial also established a commonly used definition of LAAO success based on transesophageal echocardiography (TEE) parameters: a residual stump smaller than 1 cm and no leak across the occlusion line. Based on this definition, LAAOS I suggested that complete removal of the appendage is preferred to oversewing.
Left Atrial Appendage Occlusion Study II
LAAOS II was a pilot study at four Canadian centers designed to establish the feasibility of a trial of LAAO versus OAC in patients undergoing cardiac surgery with a preexisting history of AF. Specifically, it compared LAAO plus aspirin with guideline-directed oral anticoagulant therapy in 51 patients. LAAOS II struggled with recruitment outside of the lead center in Hamilton, Ontario, Canada. Furthermore, the use of OAC in both arms converged over time, suggesting a lack of buy-in of physicians regarding the ethics of withholding OAC despite LAAO. The study confirmed the effectiveness and safety of amputation and closure for occlusion with a 100% success rate and no instances of left atrial tears deemed to be life-threatening. The addition of amputation and closure to a cardiac surgical procedure for another indication added approximately 10 minutes of surgical time. Last, a cross-sectional study performed during LAAOS II better defined the prevalence of preexisting AF and related risk factors in the cardiac surgical population, further informing the final design of LAAOS III.
Left Atrial Appendage Occlusion Study III
The rationale and design of LAAOS III have been published. In summary, it was a multicenter, prospective, blinded, randomized trial of concomitant surgical LAAO in patients with AF and CHA 2 DS 2 -VASc scores of 2 or less undergoing cardiac surgery for other reasons. In the trial, LAAO could be performed using the following techniques: (1) amputation and closure (promoted as the preferred technique, (2) stapler closure, (3) double-layer linear closure from within the atrium (for mini-thoracotomy cases if successful occlusion could be confirmed by TEE), or (4) use of a surgical LAAO device that was approved by applicable regulatory authorities (i.e., AtriClip, AtriCure Inc.). The Trial Operations Committee approved other techniques on a case-by-case basis. Percutaneous closure was not permitted, nor was surgical purse-string suture closure. Intraoperative TEE was recommended to demonstrate successful closure of the appendage defined as the absence of flow across the suture line and a residual stump smaller than 1 cm. If the initial closure was not successful, the surgeons were expected to perform additional maneuvers to rectify. In patients with preoperative LAA thrombus, the LAA was opened to remove the thrombus before LAAO. In the LAAOS III, the patients and health care workers outside of the operating room were blinded. All patients were expected to receive best care for stroke prevention. The LAAOS III investigators actively promoted ongoing adherence to guidelines for antithrombotic therapy for stroke prevention during follow-up. The medical team or family physician, without knowledge of whether the patient had had occlusion of the LAA, oversaw antithrombotic management. The operating surgeon was recused from managing antithrombotic therapy after surgery. Patients in the control arm did not have surgical LAAO but otherwise were treated identically.
The primary question of LAAOS III was whether or not concomitant LAAO at the time of other cardiac surgery would reduce ischemic stroke in patients with AF. In the context of cardiac surgery for other indications, LAAO was tested in addition to usual stroke prevention care. The alternate approach of testing LAAO as an alternative to OAC was not only impractical as learned in LAAOS II, but it did not align with the idea that anticoagulation and LAAO should work well together. OAC is remarkably effective in those with AF but has limitations inherent in many long-term prophylactic medications (e.g., bleeding risk, failure to prescribe, failure to adhere and necessary discontinuations). It was hypothesized that the addition of LAAO to anticoagulation would mitigate the risk from those limitations.
LAAOS III enrolled 4811 patients from 105 centers in 27 countries. The surgical LAAO techniques used were as follows: amputation and closure, 56%; surgical LAAO device, 15%; linear closure from within atrium, 14%; stapler, 11%; and other approved techniques, 4%. After the second formal interim analysis of efficacy, the data safety monitoring board recommended termination of follow-up and reporting of the trial; the mean duration of follow-up was 3.8 years. The results showed clearly that LAAO reduces ischemic stroke by about 33% ( Fig. 46.1 ). The hazard ratio in favor of LAAO was 0.67 (95% confidence interval, 0.53–0.85; P =.001). Thus LAAOS III was the largest RCT to demonstrate the benefit of LAAO on ischemic events. Importantly, most patients (>75%) in LAAOS III received OAC.
Cumulative incidence of stroke or systemic arterial embolism. The participants in the occlusion group underwent left atrial appendage occlusion (LAAO) at the time of cardiac surgery for another indication, and those in the no-occlusion group did not undergo LAAO at the time of cardiac surgery; all participants were expected to receive usual care. The inset shows the same data on an enlarged y -axis. CI, Confidence interval.
(Reproduced from Whitlock RP, Belley-Cote EP, Paparella D, et al. Left atrial appendage occlusion during cardiac surgery to prevent stroke. N Eng J Med . 385(11):1054–1055.)
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
