‘Will the king ever be dethroned?’ The relationship and the future of oral anticoagulation therapy versus LAA closure devices

KEYWORDS:

• Editorial
• left atrial appendage occluders
• OAT
• OAT alternatives
• OAT contraindications

1. Introduction

Currently, an impressive number of patients have been diagnosed with atrial fibrillation (AF) receiving a reliable indication for oral anticoagulation therapy (OAT) either with vitamin K inhibitors (VKAs) or with direct oral anticoagulants (DOACs) [1]. However, there is a nonnegligible percentage of these patients who cannot receive anticoagulants due to major contraindications [2]. A significant part of the patients who receive OAT develops major bleeding events requiring strict monitoring and anticoagulation withdrawal, whilst the thrombo-embolic risk persists [3].

One of the few alternatives to OAT is left atrial appendage (LAA) occlusion with various devices, an idea taken from the LAA surgical exclusion from the ‘50s. Even though no randomized controlled trials prove the efficacy of LAA surgical exclusion in the prevention of thromboembolic events of AF patients, it seems that this exclusion led to a decrease in the rate of stroke or systemic embolism, including mortality from all causes [4].

Kleineke et al. have recently published a comprehensive review in the Expert Review of Cardiovascular Therapy [5] in which they synthetically systematized the indications, the development of the closing technique (with the two devices currently available on the market: Amplazer and Watchman), and the possible complications of LAA closure. Moreover, the authors also analyzed three randomized studies and two retrospective trials in which LAA occlusion (LAAO) was compared to various OAT schemes.

It seems that as promising as these devices display in theory, there has been a stalemate from which it might not be possible to progress too smoothly so that one can consider LAAO as a safe and trustworthy alternative to OAT. In the following Editorial, we will provide a critical perspective to the present and the possible future of the indications and techniques for percutaneously closing the LAA.

2. Comparing the current evidence

LAAO first appeared in cardiology guidelines in 2014, when it was stated that it might be considered in patients undergoing primary angioplasty with a high risk of stroke and a contraindication to antiplatelet and anticoagulant therapy association [6]. The recommendation is extended in 2016 also to patients with atrial fibrillation (AF) and contraindication to anticoagulant therapy. The very recent guidelines of the European Society of Cardiology confirm this decision [7]. However, in all three cases, the recommendation is (still) weak, i.e., IIb/B.

We think several reasons led to this decision. There is a lack of randomized controlled trials (RCTs) comparing LAAO to OAT in populations with contraindication to OAT. In fact, the only three existing RCTs comparing LAAO with OAT, both VKAs and DOACs, are all made in populations that had an indication but did not have a contraindication for OAT [8–10].

If the aim is to offer LAAO as an alternative to OAT (VKAs or DOACs), we think it is a losing fight.

Since most but not all thrombi form in the left atrial appendage [11,12], LAAO may not protect against thromboembolic stroke’s overall risk related to atrial cardiomyopathy. Another critical point is that in LAAO patients, the post-procedure therapy provides one or two antiplatelet agents, stopping OAT [13]. On the one hand, this therapeutic approach could undertreat the risk of left atrial cavity thrombosis and, on the other hand, not reduce the bleeding events in patients with very high hemorrhagic risk, especially in the case of dual antiplatelet therapy.

For these reasons, in our opinion, LAAO should be aimed only at those patients who truly cannot take OAT and who, if they do not undergo the procedure, would be doomed to have no protection against thromboembolic risk. Moreover, in patients with an exceptionally high risk of bleeding, tailored antithrombotic treatment using even reduced strategies such as single antiplatelet or no therapy after LAAO should be considered.

The significant problem is that in populations at very hemorrhagic high risk, no RCTs may be designed to compare OAT and LAAO. The only possible randomization could be the one between LAAO and no therapy. However, some doubts could be raised about this approach’s ethics, given that numerous real-life data show that LAAO reduces the risk of thromboembolism according to the CHA2DS2VASc score in AF populations at high bleeding risk [14,15]. Moreover, the number of periprocedural complications of some LAAO trials is still too high, and the importance of entrusting the procedure only to a team of very expert interventional cardiologists should be strongly emphasized.

Another sensitive topic resides in the functional and endocrine properties of the LAA. Since it is 2.6 times as compliant as the left atrial body, LAA contributes significantly to left atrial reservoir function and seems essential for the adaption to pressure and volume overload [16]. Besides, the LAA is an essential structure for the release of atrial natriuretic peptide (ANP) and brain natriuretic peptide (BNP), and a significant decrease of ANP- and BNP-serum levels after interventional LAA closure has been described [17]. The question of whether or not LAAO favors the onset of heart failure remains an open question [18]. Unfortunately, the development and aggravation of heart failure was not consistently assessed in any trial on LAA occlusion and is not planned to be registered as an endpoint or adverse event in ongoing studies.

Especially at an early stage (e.g., the results of PROTECT-AF at two years [19]: 7.4 per 100 patient-years, 95%CI 5.5–9.7 vs. 4.4 per 100 patient-years, 95% CI 2.5–6.7; RR 1.69, 1.01–3.19), periprocedural complications were high, as are those of PRAGUE-17 (major LAAO-related complications occurred in 9 (4.5%) patients [10]). However, as the experience of various centers increased, complications declined (e.g., the results of PROTECT-AF at four years [8]: the rate of composite primary safety events was similar in the two groups – 3.6 events per 100 patient-years in the device group vs. 3.1 in the warfarin group (HR, 1.21; 95% CI, 0.78–1.94, P = 0.41). It is also important to note that, even in such a fragile population as that of Genovesi et al. (two major peri-procedural complications were recorded in 92 patients), the complication rate is certainly acceptable, even if not negligible [20].

3. LAAO – a veritable solution for selected patients

Patients undergoing dialysis are examples of a population where LAAO can genuinely be a viable alternative to OAT. Indeed, in this population, where the prevalence of AF is very high, VKAs are associated with a high incidence of bleeding, and DOACs are contraindicated [7]. Initial encouraging data suggest that LAAO occlusion in dialysis patients has a good efficacy and safety profile [20]. A Spanish study is currently underway in ESRD patients (WATCH-HD, NCT03446794), including two arms, the Watchman device vs. no-therapy. The study is still under recruitment.

Considering that in AF patients, a high percentage of thrombus is formed in the left atrial appendage, in this group of AF patients who do not take OAT, the procedure has the meaning of protecting them at least from the events associated with thrombosis of the left atrial appendage.

Other categories of patients in which LAAO could be very helpful are patients with previous intracranial hemorrhage(s), patients with major gastrointestinal bleeding due to non-removable causes (esophageal varices, colon angiodysplasia, hereditary hemorrhagic telangiectasia), patients with cerebral amyloid angiopathy, patients with recurrences of ischemic stroke despite taking OAT and frail elderly patients with high risk of bleeding and falling. Again, factual data on the efficacy and safety of LAAO in these populations are currently lacking.

4. Future devices, future alternatives

There are two main approaches to catheter-based LAA occlusion, the endocardial and the epicardial approach, with several devices available in different sizes to suit the atria’s various anatomy. Currently, the European approved devices are shown in Table 1.

Table 1. Current and future percutaneous LAA closure devices

Name	Manufacturer	CE mark	Available sizes	Design	Location
Watchman/Watchman FLX	Boston Scientific	2015/2019	20, 24, 27, 31, 35	Single (lobe)	Endocardial
Amplatzer Cardiac Plug/Amplatzer Amulet	Abbot Vascular	2008/2013	16, 18, 20, 22, 25, 28, 31, 34	Double (lobe and disc)	Endocardial
Occlutech	Occlutech International AB	2016	15, 18, 21, 24, 27, 30, 33, 36, 39	Single (lobe)	Endocardial
LAmbre	Lifetech Scientific	2016	Umbrella: 16–26 Disc: 22–38	Double (umbrella and cover)	Endocardial
WaveCrest	Biosense Webster	2013	22, 27, 32	Single (lobe)	Endocardial
Ultraseal	Cardia	2016	16, 18, 20, 22, 24, 26, 28, 30, 32	Double (bulb and sail)	Endocardial
Lariat	SentreHEART	2015	max 40 mm (max 45 mm Lariat+)	Endoepicardial
Sierra	Aegis Medical Innovations Inc.	Undergoing clinical evaluation (LASSO-AF)	Single size	Epicardial

When it comes to endocardial devices, one can distinguish between two systems in particular:

1. Plug-LAA closure systems, consisting of a self-expanding umbrella/body that anchors itself in the LAA’s neck area and isolates it from the circulation (e.g., Watchman – Boston Scientific and WaveCrest – Biosense Webster).

2. The devices that use the ‘pacifier’ principle, equipped with an abluminal disc and an anchoring unit (e.g., Amplatzer Cardiac Plug, Amplatzer Amulet – Abbott Vascular, LAmbre – Lifetech Scientific).

In the last twenty years, LAA closure has suffered a rapid evolution, as increasing clinical experience translated into significant improvements in both implant success and procedural safety. As long term follow-up revealed encouraging data on safety events between LAA closure and OAC strategies, the future objective would be to tip the balance in favor of the percutaneous approach for carefully selected patients by diminishing the periprocedural complications and increasing safety and efficacy (mainly by addressing peri-device leak and device thrombosis). The rate of thrombus formation on the device during long-term follow-up, extending 12 months, is mostly unknown. Most antithrombotic regimens are usually based on the protocols used after patent foramen ovale (PFO) closure procedures.

However, it should be mentioned that new peri-device leaks have been described after an initially completely closed LAA [21]. Since no systematic investigations regarding leaks >12 months postintervention have been performed, the incidence and consequence of such late peri-device leaks remain to be determined.

Already newly designed devices assert lower device thrombosis rates by reshaping the device exposed surface and significantly less peri-device leakage due to better flexibility and device attachment to various LAA anatomies.

The new generation of Watchman devices, Watchman FLX, has recently received approval (21 July 2020) from the FDA. For other devices currently in use, studies are in progress to confirm their safety and efficacy by comparing to already established Watchman device (e.g., AMULET-IDE trial – NCT02879448 and WAVECREST-2 – NCT03302494).

5. Conclusions

This technique has a more significant potential than the current studies and guidelines claim nowadays. We believe that the results of ongoing studies will help introduce a stronger indication in future guidelines and select categories of patients who will have clear benefits from LAAO. The years to come will bring with them even more experience in the very practice of implanting upgraded and less thrombogenic devices.

6. Expert opinion

There is a lack of RCTs comparing LAAO to OAT in populations with contraindication to OAT, and in fact the only three existing RCTs are made in populations that had an indication but did not have a contraindication for OAT. If the aim is to offer LAAO as an alternative to OAT (VKAs or DOACs), we think it is a losing fight. The significant problem is that in populations at very hemorrhagic high risk, no RCTs may be designed to compare OAT and LAAO. 

In our opinion, LAAO should be aimed only at those patients who truly cannot take OAT and who, if they do not undergo the procedure, would be doomed to have no protection against thromboembolic risk. LAAO can genuinely be a viable solution for selected patients, like patients undergoing dialysis, patients with previous intracranial hemorrhage(s), patients with major gastrointestinal bleeding due to non-removable causes, patients with cerebral amyloid angiopathy, patients with recurrences of ischemic stroke despite taking OAT and frail elderly patients with a high risk of bleeding and falling. It should be mentioned that factual data on the efficacy and safety of LAAO in these populations are currently lacking.

As long term follow-up revealed encouraging data on safety events between LAA closure and OAC strategies, the future objective would be to tip the balance in favor of the percutaneous approach for carefully selected patients by diminishing the periprocedural complications and increasing safety and efficacy. Already newly designed devices assert lower device thrombosis rates by reshaping the device exposed surface and significantly less peri-device leakage due to better flexibility and device attachment to various LAA anatomies. In patients with an exceptionally high risk of bleeding, tailored antithrombotic treatment using even reduced strategies such as single antiplatelet or no therapy after LAAO should be considered.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose. 

Additional information

Funding

This paper was not funded.