Left atrial appendage occluders—new hope for stroke prevention in high-risk patients

Atrial fibrillation (AF) is the most common sustained arrhythmia. It affects ∼2% of the general population and the prevalence is increasing with age up to 10% among people over 70 years. One of the most important risks associated with AF is the risk of a cardioembolic event. According to the current ESC guidelines, risk stratification should be done for all AF patients using the CHA2DS2VASc score. This scoring system is especially useful for low-risk patients. CHA2DS2VASc score can depict patients with very low cardioembolic risk: patients with a CHA2DS2VASc = 0 (or 1 in females) have a very low risk of stroke/thromboembolism, and should not be treated by oral anticoagulation (OAC). The uncertainities about OAC indication in patients with CHA2DS2VASc = 1 (or possibly in females = 2) were described in a recent position paper, with recommendation to consider new OAC in these patients.1 Compared with control or placebo groups, vitamin K antagonist therapy (primarily warfarin) reduces stroke risk by 64% and mortality by 26%.2 Warfarin has been mostly replaced by non-vitamin K anticoagulants (NOAC) due to 19% risk reduction in significant strokes/systemic embolisms, and a 51% reduction in haemorrhagic strokes.3 However, there are two limitations to OAC treatment in AF patients;1 it increases the risk of bleeding, and2 it is insufficient to avoid strokes in a significant portion of high-risk patients. In the population of OAC-treated AF patients, ∼20% stop long-term OAC therapy, with bleeding being the most common reason for withdrawing OAC. Thus, bleeding complications and nonadherence present barriers to effective anticoagulation therapy. Left atrial appendage closure (LAAC) is a rationale alternative for patients at high bleeding risk or patients unsuitable for long-term anticoagulation (Figure 1). In observational studies, LAAC was associated with ∼ 60–70% reduction in cardioembolic events compared with expected rate according to the CHA2DS2VASc score, and low rate of late-postprocedural bleeding risk.4 Three randomized controlled trial (RCT) documented that LAAC compared with OAC is associated with a reduction in non-procedural bleeding.5,6 The reduction of non-procedural bleeding was present even in comparison with NOACs. Interestingly, the pooled analysis of these three RCTs has shown that compared with OAC, LAAC was also associated with a reduction in mortality.7 Current RCTs have documented the relevance of bleeding as a risk factor in the AF population and the importance of reducing the bleeding risk for better AF prognoses. Thus, for patients with high bleeding risk, or patients who are deemed unsuitable for long-term OAC for other reasons, LAAC presents a rationale alternative. Percutaneous implantation of LAAC. Left: fluoroscopic image of LAAC in place. Right contrast injection to the left atrium outlines the atrial cavity and does not fill the left atrial appendage, where the occlude is properly positioned. In the initial report on the utility of the CHA2DS2VASc score in AF patients without OAC, the annual risk for cardioembolic events was 0.7% for CHA2DS2VASc = 1, 2.3% for CHA2DS2VASc = 4, and 10.1% for CHA2DS2VASc = 7. Importantly, the risk of stroke remains high in patients with higher CHA2DS2VASc scores despite anticoagulation treatment. When OAC-treated patients were divided into low risk (CHA2DS2VASc = 0), intermediate risk (CHA2DS2VASc = 1), and high risk (CHA2DS2VASc ≥ 2), higher CHA2DS2VASc scores were associated with a stepwise increase in the rate of cardioembolic events with an hazard ratio (HR) of 3.75 [95% confidence interval (CI), 1.20–11.73], indicating a 275% increase in the likelihood of cardioembolic event for each increase in the risk stratum (i.e. moving from low to intermediate or intermediate to high).8 It means, the risk of cardioembolic risk continuously increases with increasing CHA2DS2VASc. However, our treatment recommendations are the same for all high-risk patients, which is NOAC, i.e. the same treatment irrespective of risk stratum. The currently recommended and used CHA2DS2VASc scoring system is useless for the guidance of treatment of high-risk patients, which is associated with other current weaknesses in the treatment of very high-risk patients. This represents a substantially different situation from what exists in other areas of cardiology. For instance, treatment is highly individualized in patients with hyperlipidaemia according to patient risk (e.g. statins with different target LDL levels, ezetimibe, PCSK9 inhibitors), and the same is true for the treatment of hypertension and many other conditions. Unfortunately, for AF-related stroke prevention, our therapeutic arsenal is very limited, and more potent treatments for highest-risk patients are currently unavailable. There is a clear, unmet need for better treatments in this area, and LAAC may offer a promising alternative once it has been thoroughly evaluated in RCTs. In 2019, our group published the 5-year follow-up of the PRAGUE-12 study.9 In this randomized prospective study, we enrolled 207 surgical candidates (valve or bypass surgery) with an AF history. Patients were randomized to concomitant surgical left atrial ablation plus LAAC or no ablation and no LAAC; OAC was used in both arms after surgery. The 5-year incidence of stroke was significantly reduced in the ablation + LAAC group (sHR 0.32, 95% CI 0.12–0.84) even after adjustment for other clinical variables. Recently, the huge landmark Left Atrial Appendage Occlusion Study (LAAOS) III trial demonstrated the usefulness of surgical left atrial appendage occlusion/closure in patients with AF indicated for other cardiac surgery.10 In this trial, 4811 surgical candidates (mean CHA2DS2VASc = 4.2 and scheduled for coronary artery bypass grafting or valve surgery) were randomized to undergo or not undergo left atrial appendage occlusion/closure at the time of cardiac surgery. Patients in both groups were to be kept on OAC according to current recommendations; however, for several reasons, anticoagulation had been withdrawn in 20.7% of patients at 1 year and in 23.2% of patients by 3 years after randomization (but similar in both groups). Importantly, during a 3.8-year follow-up, the LAA occlusion/closure group was associated with a 33% reduction in stroke/systemic embolism (HR 0.67, 95% CI 0.53–0.85). This study has several important implications. First, it documents the importance of the LAA in the pathophysiology of stroke in AF patients. Next, it emphasizes the importance of the LAAC during cardiac surgery in AF patients, i.e. LAAC should be done in all patients referred to cardiac surgery with a history of AF. The study also describes how to perform an effective surgical LAAC by evaluating various surgical techniques, each with a unique efficacy. Finally, the study has significant implications for cardiology and future research in AF stroke prevention. Of course, the results of the LAAOS III study cannot be generalized to all currently available percutaneous methods for LAA closure. No similar data exist for percutaneous LAAC, and percutaneous LAAC cannot be offered on top of anticoagulation as a non-surgical LAAC alternative. Since the combination of antithrombotic treatment (e.g. anticoagulation plus antiplatelet) increases bleeding risk, and as such, this treatment strategy does not move us in a rational direction. A combination of pharmacological and non-pharmacological treatment (Figure 2) could, on the other hand, represent a reasonable option for high-risk AF patients, and it opens new areas for future research in the field. Possible future strategy if additional studies will support the combination of NOACs with LAAC in patients with the highest risk of stroke. Conflict of interest: none declared.