A collective European experience with left atrial appendage suture ligation using the LARIAT⁺ device

Abstract

Aims

We report the collective European experience of percutaneous left atrial appendage (LAA) suture ligation using the recent generation LARIAT⁺ suture delivery device.

Methods and results

A total of 141 patients with non-valvular atrial fibrillation and contraindication to oral anticoagulation (OAC), thrombo-embolic events despite OAC or electrical LAA isolation were enrolled at seven European hospitals to undergo LAA ligation. Patients were followed up by clinical visits and transoesophageal echocardiography (TOE) following LAA closure. Left atrial appendage ligation was completed in 138/141 patients (97.8%). Three patients did not undergo attempted deployment of the LARIAT device due to pericardial adhesion after previous epicardial ventricular tachycardia ablation (n = 1), a pericardial access-related complication (n = 1), and multiple posterior LAA lobes (n = 1). Serious 30-day procedural adverse events occurred in 4/141 patients (2.8%). There were two device-related LAA perforations (1.4%) not resulting in any corrective intervention as the LAA was completely sealed with the LARIAT. Minor adverse events occurred in 19 patients (13.5%), including two pericardial effusions due to procedure-related pericarditis requiring pericardiocentesis. Transoesophageal echocardiography was performed after LAA ligation in 103/138 patients (74.6%) after a mean of 181 ± 72 days. Complete LAA closure was documented in 100 patients (97.1%). Two patients (1.8% of patients with follow-up) experienced a transient ischaemic attack at 4 and 7 months follow-up, although there was no leak observed with TOE. There were two deaths during long-term follow-up which were both not device related.

Conclusion

Initial experience with the LARIAT⁺ device demonstrates feasibility of LAA exclusion. Further larger prospective studies with longer follow-up are warranted.

Atrial fibrillation, Cardioembolic stroke, LAA ligation, LARIAT, LAA isolation

What’s new?

• This is the largest published European cohort of patients undergoing epicardial left atrial appendage (LAA) closure with the second-generation LARIAT device.

• This study reports on a large collective of patients with previous electrical isolation of the LAA undergoing epicardial closure.

• This study found that LAA closure is feasible with acceptably low access complications and peri-procedural serious adverse events.

• Echocardiographic follow-up demonstrated complete LAA closure in the majority of patients.

• The status of epicardial LAA closure in stroke prevention requires a randomized trial.

Introduction

Atrial fibrillation (AF) is associated with significant morbidity and mortality. Atrial fibrillation is a major risk factor for thrombo-embolic events.¹ Several trials demonstrated that vitamin K antagonists (VKAs) reduce the risk of thrombo-embolic events and mortality.² However, VKAs have several significant limitations such as inter- and intra-patient variability, multiple drug and food interactions, and a significant bleeding risk. Novel oral anticoagulants overcome not all limitations.³ Factors such as poor patients compliance and bleeding complications result in insufficient oral anticoagulation (OAC) therapy in a significant proportion of AF patients.⁴

The left atrial appendage (LAA) has been demonstrated to be the major source of stroke. Thus, LAA occlusion devices have been developed to overcome limitations of OAC therapy. Endocardial devices have been used since 2001 and a plethora of data, including randomized controlled trials proved the safety and efficacy of these devices. On the other hand, the epicardially deployed LAA exclusion device (LARIAT, Sentreheart, Redwood City, CA, USA) is increasingly used in Europe and the USA.⁵ In observational studies, the LARIAT device effectively closed the LAA and had an acceptable procedural risk when a micropuncture needle was used for epicardial access.⁶ In addition, recent studies have shown that epicardial LAA ligation results in LAA electrical isolation and a reduction of the AF burden.^6–8

The LARIAT⁺ is the second generation of the device and has improved features as compared to the first generation device consisting of: (i) snare expansion from 40 mm to 45 mm, (ii) addition of a platinum–iridium ‘L’ Marker that allows identification of the device orientation under fluoroscopy, and (iii) a stainless steel wire braid on catheter shaft that provides improved ‘torque-ability’ of the catheter. We report on the collective European experience with the LARIAT⁺ device.

Methods

Study design

This was a multi-centre observational study in patients with non-valvular AF who underwent LAA ligation with the LARIAT⁺ device. Consecutive patients undergoing LAA ligation with the LARIAT⁺ device in Europe were included into the study. Seven clinical sites from four European countries included Asklepios Klinik St. Georg (Hamburg, Germany), CardioVascular Center Frankfurt CVC (Frankfurt, Germany), John Paul II Hospital (Krakow, Poland), Royal Brompton (London, United Kingdom), Cardiovascular Center (Bad Neustadt, Germany), University Herzzentrum Lübeck (Lübeck, Germany), Herzzentrum Dresden (Dresden, Germany), and the University Hospital Zürich (Zürich, Switzerland). All participants provided written informed consent to the procedure. All patient information was anonymized, and data were evaluated retrospectively. The study was approved by the local ethic review board. Study objectives were as follows: (i) to determine effectiveness of LAA closure with the LARIAT⁺ device; (ii) to assess procedural and 30-day peri-procedural safety; and (iii) to obtain initial clinical follow-up.

Pre-specified 30-day major peri-procedural adverse events based on the Munich LAA consensus document and aMAZE trial included the following (ix, x): (i) serious injury to cardiac-related structure requiring surgical intervention; (ii) bleeding defined as two or more units of packed red blood cells (PRBC) administered in the post-operative period; (iii) pericarditis requiring surgical treatment; (iv) haemothorax requiring surgery; (v) pneumothorax requiring surgery; (vi) vascular injury requiring surgery, hospital admission, or PRBC transfusion; (vii) pericardial effusions requiring surgery, blood transfusion, or aspiration of >500 mL of blood; (viii) injury to abdominal organs requiring surgery; (ix) stroke; and (x) death.

Adverse events and clinical endpoints were self-reported and adjudicated by the principal investigator of each institution, then reviewed with a principal investigator of another institution. Four sites were new institutions with no previous experience with the LARIAT procedure. The other institutions had performed >10 LARIAT cases before enrolling their first LARIAT⁺ patient.

Patient population

Patients were identified and enrolled between January 2014 and August 2016 for LAA ligation. Inclusion criteria were: (i) age 18 years or older; (ii) non-valvular AF; (iii) at least one risk factor of embolic stroke (CHADS₂ score of at least 1 or CHA₂DS₂-VASc score of at least 2; (iv) ineligible for OAC therapy (labile international normalized ratio level, non-compliant, contraindicated), or OAC failure (thromboembolism on OAC therapy), or as part of a strategy for LAA exclusion with pulmonary vein isolation (PVI) for symptomatic, antiarrhythmic drug-refractory AF; and (v) a life expectancy >1 year.

Patients were excluded from the study if they met any of the aMAZE trial exclusion criteria:⁹ (i) known history of pericarditis, thoracic radiation, or cardiac surgery; (ii) pectus excavatum; (iii) recent myocardial infarction within 3 months; (iv) prior embolic event within the last 30 days; (v) NYHA functional class IV heart failure symptoms; and (vi) left ventricular ejection fraction ≤30%.

Patients meeting the criteria for the study enrolment underwent a screening contrast cardiac computed tomography (CT) scan. Additional exclusion criteria based on LAA anatomy included: (i) LAA width >45 mm; (ii) a superiorly oriented LAA with the LAA apex directed behind the pulmonary trunk; (iii) bi-lobed LAA or multilobed LAA in which lobes were oriented in different planes exceeding 45 mm; and (iv) a posteriorly rotated heart.

Percutaneous suture ligation with the LARIAT⁺ device

Left atrial appendage ligation using the LARIAT⁺ suture delivery device was described previously.¹⁰ Pericardial access using a 21-gauge micropuncture needle with or without an 18-gauge telescopic needle approach was performed as previously described.¹⁰ If the telescopic approach was used, an 18-G needle was advanced below the sternum into the anterior mediastinal space. The inner micropuncture needle was advanced under fluoroscopic guidance into the anterior mediastinal space with confirmation of pericardial access with contrast injection. Transseptal catheterization and LAA angiography were performed. The endocardial magnet wire and EndoCATH balloon catheter were positioned in the anterior superior LAA. The epicardial magnet was inserted through the epicardial sheath and connected to the endocardial magnet. The LARIAT was advanced through the epicardial sheath over the epicardial magnet wire. The ostium of the LAA was verified with the balloon positioned at the ostium of the LAA, and the snare was closed. The suture was released from the snare and tightened with the TenSure device. The suture was cut with a suture-cutting device. Left atrial appendage exclusion success was defined as the successful closure of the LAA with the absence of a contrast leak on left atriogram and ≤5 mm jet as visualized by colour Doppler on transoesophageal echocardiography (TOE).¹¹

Clinical follow-up

Transoesophageal echocardiography to confirm LAA closure and detect intracardiac thrombus formation was performed at 30–90 days and 6–12 months post-LAA ligation. Incomplete LAA closure was defined as colour Doppler flow of >5 mm as based on the Munich LAA consensus document.¹¹ Clinical follow-up occurred at 1 month, 3 months, and 1-year post-LAA ligation.

Post-procedural medication

In all patients, colchicine 0.5 mg bid for at least 3 weeks was recommended. Additionally, non-steroidal anti-inflammatory drugs (NSAID) were prescribed for at least 2 weeks. Patients with an absolute contraindication to OAC therapy remained off OAC therapy. Patients with a CHA₂DS₂-VASc score of 2 or higher who could tolerate OAC therapy and were scheduled to undergo PVI were recommended to continue OAC therapy until the first TOE control, then based on results of TOE investigation.

Classification of adverse events

Adverse events were classified according to the Munich consensus document on LAA closure. Pericardial effusion related to epicardial access was defined as a major complication in case of aspiration of ≥500 mL bloody fluid, requiring transfusion or surgical intervention. Pericarditis was defined as severe if prolonged (0.4 weeks) anti-inflammatory therapy was required, in case of recurrent effusions, or surgical therapy was needed. Bleeding events were categorized according to the Munich consensus document and the Bleeding Academic Research Consortium (BARC) document.¹²

Statistical analysis

Continuous data were described as means and standard deviations. Statistics were computed with SPSS Statistics (Version 23, IBM, Armonk, NY, USA).

Results

Patient population

The referred population consisted of 186 patients (Figure 1). Thirty-eight patients (20.4%) were excluded due to LAA morphology exclusion criteria and seven patients were excluded at the time of the procedure due to a thrombus during TOE. The remaining 141 patients underwent attempted LARIAT procedure (Figure 1). Table 1 describes baseline characteristics of the 141 patients, in whom LAA ligation was performed.

Figure 1

Study overview. LA, left atrial; LAA, left atrial appendage; Pts, patients; TIA, transient ischemic attack; TOE, transoesophageal echocardiography.

Table 1

Patient characteristics

Screened patients	186
Screening failure
Total	45/186 (24.2)
LAA anatomy	38/186 (20.4)
Thrombus detected by TOE	7/186 (3.8)
N patients undergoing LAA ligation attempt	141
Age (years)	70.4 ± 12.2
N female patients	62/141 (44.0)
Type of AF
Paroxysmal	67/141 (47.5)
Persistent	30/141 (21.3)
Longstanding persistent	44/141 (31.2)
N patients with previous stroke	13/141 (9.2)
N patients with previous bleeding event	30/141 (21.3)
CHA2DS2-VASc score	3 [2; 4]
Reason for LAA ligation
Contraindication/intolerance to OAC	113/141 (80.1)
Prevention of LAA thrombus formation after LAA electrical isolation	23 (16.3)
Adjunctive to PVI	5 (3.5)

AF, atrial fibrillation; LAA, left atrial appendage; OAC, oral anticoagulation; PVI, pulmonary vein isolation; TOE, transoesophageal echocardiography.

Procedural results and peri-procedural safety

Left atrial appendage closure was achieved in 138/141 patients (97.9%; Figure 1). Three patients (2.1%) did not undergo attempted deployment of the LARIAT⁺ device due to pericardial adhesion after previous epicardial ventricular tachycardia ablation (n = 1), a pericardial access related complication (n = 1) and due to multiple posterior LAA lobes (n = 1) and were excluded from the follow-up. Complete acute closure without intraprocedural leakage was achieved in 130 of the 138 (94.2%) patients undergoing LAA ligation.

Adverse events within 30 days after LAA ligation are listed in Table 2. Serious peri-procedural adverse events occurred in 2.8% (4/141 patients). One patient had a right ventricular (RV) perforation during pericardial access that led to tamponade. The tamponade was remedied by drainage of the pericardial space via a pigtail catheter passed through a steerable sheath. The patient was haemodynamically stable until apparent RV puncture with the steerable sheath occurred. The patient underwent surgical RV repair and surgical LAA closure, but did sustain a stroke during surgery. Another patient sustained a pneumothorax requiring a chest tube. One patient with low haemoglobin value before the procedure (7.9 g/dL) had a groin bleeding requiring transfusion of two PRBC. There was one patient with a late pericardial effusion of >500 cc resulting in pericardial drainage. There were two device-related complications (1.4%, 2/141 patients) in which the LAA was perforated with the endocardial magnet wire. In each case, the endocardial magnet wire was repositioned to the anterior aspect of the LAA with completion of the LARIAT⁺ procedure. Neither LAA perforation required a corrective intervention as the LAA was completely sealed with the LARIAT device. There were no acute or long-term adverse consequences of the LAA perforation.

Table 2

Adverse events

Serious adverse events	4 (2.8)
Cardiac perforation needing surgical repair	1 (0.7)
Groin bleeding requiring transfusion	1 (0.7)
Pericardial effusion needing puncture of >500 mL	1 (0.7)
Pneumothorax requiring chest tube	1 (0.7)
Device-related adverse events	2 (1.4%)
LAA perforation (no intervention required)	2 (1.4%)
Minor adverse events within 30 days of LAA ligation	19 (13.5%)
Superficial subxyphoid bleeding	1 (0.7)
Pericardial effusion (two required pericardiocentesis)	3 (2.1)
Pseudoaneurysm (conservative treatment)	1 (0.7)
Pleural effusion (no intervention required)	4 (2.8)
Pneumothorax (no intervention required)	1 (0.7)
Pericarditis greater than 2 days (conservative treatment)	8 (5.7)
Arterio-venous fistula (no intervention required)	1 (0.7)

LAA, left atrial appendage.

Table 3

Clinical and echocardiographic follow-up

N patients witd available clinical follow-up	111/138 (80.4)
Duration of follow-up (days)	180 ± 104
N patients with thrombo-embolic events	2/111 (1.8)
TIA	2/111 (1.8)
N patients with TOE	103/138 (74.6)
Mean timespan from LAA ligation to TOE (days)	176 ± 73
N patients with complete LAA closure	100/103 (97.1)
N patients with LA thrombus	2/103 (1.9)

LA, left atrium; LAA, left atrial appendage; TIA, transient ischemic attack; TOE, transoesophageal echocardiography.

Other minor adverse events, such as pneumothorax with no need for treatment, superficial subxyphoid bleeding, pleural effusions without need for intervention, pericardial effusions due to peri-procedural pericarditis in which two required pericardiocentesis, pericarditis not requiring intervention, pseudoaneurysm, and arterio-venous fistula not requiring a surgical intervention occurred in 13.5% (19/141) of patients (Table 2).

Clinical follow-up

Transoesophageal echocardiography was performed after LAA ligation in 103/138 patients (74.6%) after a mean of 181 ± 72 days after LAA ligation. Left atrial appendage closure without leakages >5 mm were documented in 100 of the 103 patients (97.1%). There were 81 cases without any leakage or leakages <2 mm according to TOE imaging (78.6%), 19 cases (18.4% of patients with TOE follow-up) with leakages ≥2 mm and <5 mm and 3 patients with a leakage ≥5 mm (Figure 2). Repeat TOE controls revealed stable leakage status in all three patients with leakages ≥5 mm. There was moderate leakage progression in 5/60 patients (8.3%) with no leakage/leakage <2 mm during the initial TOE control and leakage regression in 4/12 patients (28.6%) with a small leakage of ≥2 mm/<5 mm during the initial TOE control (Figure 2). There was no case of leakage ≥5 mm during repeat TOE controls if the initial TOE documented complete LAA exclusion or a leakage <5 mm (Figure 2).

Figure 2

Results of echocardiographic follow-up. Pts, patients; TOE, transoesophageal echocardiography.

Left atrial appendage thrombus on the ligated atrial side of the LAA stump was found in two patients at the 1–3 month TOE (1.9%). All thrombi resolved with OAC. Both patients had no leak and no thrombo-embolic event.

Clinical follow-up containing patients visits was available in 111/138 cases with a mean duration of 180 ± 104 days after LAA ligation. Two patients (1.8%) experienced a transient ischaemic attack (TIA) at 4 and 7 months follow-up, although there were no leakages observed at TOE investigations. There were two deaths during long-term follow-up, neither device nor procedure related (one septic shock and one malignant disease).

Discussion

The collective European experience with the LARIAT⁺ device demonstrates that LAA closure is feasible with acceptably low access complications and peri-procedural serious adverse events. The short-term efficacy of LAA closure with the LARIAT⁺ device is consistent with the initial observational studies of the LARIAT device⁵ and the recent US multi-centre registry.⁷

The safety profile in this series of patients is comparable with other LAA exclusion procedures^7,13 and results of a recent LARIAT US multi-centre registry.⁷ The strength of this series of patients is the inclusion of all consecutive patients within Europe undergoing LAA ligation with the LARIAT⁺ device and a complete listing of all adverse events. The previous registries and surveys on the LARIAT device did not include all centres or patients undergoing a LARIAT procedure during the designated enrolment period.^10,14–16 Therefore, the results presented in this report are more representative of an initial experience with the LARIAT procedure. The relatively low adverse event rate requiring a corrective intervention can be attributed to improved training based on lessons learned from previous LARIAT experience, use of the micropuncture needle for pericardial access and post-procedural use of colchicine and NSAIDs.⁷ Additionally, the LARIAT+ device has an improved handling and a larger snare, thus allowing for less variability in the procedure. Nevertheless, our results show that cardiac perforation can occur during epicardial LAA closure procedures, potentially resulting fatal patient outcome. Left atrial appendage ligation should therefore only be performed in experienced centres due to the complexity of the procedure. Additionally, two pericardial effusions were labelled as minor complications according to the Munich consensus statement, but would rather be categorized as major adverse events in most studies on LAA closure.

Although colchicine or NSAID was used post-LAA ligation, there was still a small percentage (8/141 patients, 5.7%) of patients that developed symptoms of pericarditis lasting longer than 2 days after the procedure. The post-LAA ligation inflammatory reaction is in response to LAA ischaemia.⁹ Further studies are required to determine whether other anti-inflammatory strategies are able to further reduce the reactive inflammatory response.

Comparison to thoracoscopic and surgical left atrial appendage exclusion

The LARIAT procedure offers potential benefits like total LAA exclusion without leaving any foreign body, which is more comparable to surgical procedures than to interventional LAA closure. Accordingly, comparison of the LARIAT procedure not to interventional LAA closure, but rather to thoracoscopic or surgical LAA exclusion would be rational. Complication rates of thoracoscopic AF ablation and LAA closure are reported to be up to 27% in a study on patients undergoing thoracoscopic AF ablation and LAA closure with a surgically deployed clip device.¹⁷ Reports on patients undergoing sole thoracoscopic or surgical LAA exclusion are limited to case series making a comparison with the LARIAT procedure difficult at the current date. Additionally, it has to be noted that termination of OAC after surgical LAA exclusion is not recommended by European AF guidelines due to limited available systematic clinical data.¹⁸

Left atrial appendage as a trigger and substrate for atrial fibrillation

In patients with persistent and longstanding persistent AF, PVI as the sole ablative strategy may not be sufficient to maintain sinus rhythm.

Several studies have demonstrated that the LAA is an important trigger and/or substrate for AF in a subset of patients. Di Biase et al.¹⁹ demonstrated, that focal discharge originating in the LAA may induce AF and that electrical isolation of the LAA improves outcome. This is in line with the observation, that epicardial LAA exclusion results in a lower AF burden.^6,8

However, in a recent study by Rillig et al.,²⁰ electrical LAA isolation resulted in a high incidence of stroke and LAA thrombus formation. In this study, 50 patients underwent LAA isolation in addition to PVI for treatment of AF or atrial tachycardia. During a mean follow-up of 6.5 months, stroke or TIA occurred in 3 patients (6%) and LAA thrombus was identified on TOE in another 10 patients (20%). Importantly, all but two patients with thrombus or thrombo-embolic events were on OAC therapy. In a matched control group undergoing PVI alone without LAA isolation, no LAA thrombus was detected and no stroke occurred. This finding was recently confirmed by Di Biase et al.,²¹ who found LAA thrombus formation in patients with reduced LAA flow velocity after LAA isolation. These studies underscore the possible iatrogenic potential of LAA isolation to induce mechanical standstill leading to LAA thrombus formation and subsequent thromboembolism; thus pointing out the need to exclude the LAA. A recent study by Zender et al.²² demonstrated the feasibility of endocardial LAA closure in prevention of LAA thrombus formation and cardiogenic thromboembolism. The LARIAT device potentially leads to electrical LAA isolation with a reduced thrombo-embolic risk.^6,8 In a recent study, a combined approach of LAA isolation and subsequent LAA exclusion with the LARIAT device led to long-term maintenance of sinus rhythm in about 70% of a patient cohort comprised of PVI non-responders.⁸ Additionally, a high incidence of LAA thrombus formation after LAA isolation was documented in these patients but no LAA thrombus formation was observed after successful LAA exclusion with the LARIAT device.⁸

Clinical implications

Epicardial exclusion of the LAA has the potential benefits of both stroke prevention as well as potentially playing a role in the restoration and maintenance of sinus rhythm in patients with non-valvular AF.⁶ In high-risk AF patients for thrombo-embolic events with contraindication to OAC therapy, LAA ligation with the LARIAT device demonstrated a stroke and systemic embolism rate of 1.0% per year in this population that represented an 84% reduction in the expected stroke rate.⁵ Following LAA closure with the LARIAT device in our patient population, there were only two patients that had a TIA. The majority of patients in this study were AF patients at high risk for thrombo-embolic events, especially since a subset of these patients had LAA electrical isolation during a previous AF ablation procedure. Patients who have undergone LAA electrical isolation as part of an ablation strategy for the treatment of persistent AF have an incidence of thrombus formation in the LAA of more than 20% irrespective of the CHA₂DS₂-VASc score.^8,20,23 A previous study found that LAA ligation is an effective therapy to prevented thrombus formation following electrical LAA isolation.⁸ Nevertheless, the small number of patients included into this study and the retrospective study design do not allow to draw final conclusions about the effectiveness of LAA ligation for stroke prevention. Further prospective studies are needed to confirm these findings.

A unique feature of the LARIAT⁺ device is absence of a foreign body which remains intracardially, which is in contrast to current endocardial LAA closure systems. This characteristic may potentially result in a lower frequency of device-related thrombus formation after LAA closure. We observed two cases of LA thrombus formation documented during echocardiographic follow-up investigations. The observed frequency is therefore comparable to the frequency observed in the Amulet observational registry.²⁴ Real-world data on patients undergoing LAA closure with the Watchman device from the European EWOLUTION registry found LA thrombi in 4.1% of patients with available imaging after LAA closure.²⁵ A multi-centre study from the US which compared epicardial LAA ligation with LAA closure with the Watchman device found LA thrombi in 1.5% and 2.7% of patients, respectively.²⁶ A non-standardized post-procedural OAC regime might have contributed to the occurrence of LA thrombosis after LAA ligation. Further studies are needed to evaluate the optimal OAC regime post-LAA ligation, the optimal timepoint of echocardiographic imaging and the status of LAA ligation in comparison to endocardial LAA closure with regards to LA thrombus formation after LAA closure.

Besides prevention of thromboembolism LAA ligation with the LARIAT⁺ device is used as an adjunctive therapy to PVI.^8,9 Epicardial exclusion of the LAA has been associated with LAA electrical isolation and termination of atrial tachycardias and AF.^8,20,23 Recent studies demonstrated, that LAA ligation combined with PVI results in favourable electrical and structural remodelling of the LA⁸ and that it has the potential to decrease arrhythmia recurrence. The aMAZE trial is a multi-centre, randomized trial testing the hypothesis that the LAA is a critical structure in the maintenance of persistent and longstanding persistent AF and that LAA ligation combined with PVI will decrease the recurrence of AF in patients with persistent and longstanding persistent AF.⁹

A further indication for LAA ligation using the LARIAT device is stroke prevention in patients at high thrombo-embolic risk and high bleeding risk. Although there are no randomized data evaluating this patients population for both endocardial and epicardial LAA closure device, a recent survey of the European Heart Rhythm Association revealed that the most common indication for LAA occlusion is stroke prevention in patients with an absolute contraindication to OAC therapy or a history of bleeding who are at high thrombo-embolic risk.²⁷ For this indication, the epicardial device has the benefit that no foreign body remains inside the heart.

Limitations

This observational study has limitations inherent to the study design. Patient inclusion criteria to perform CT screening for evaluation of suitability for LAA ligation depended on local inclusion criteria. The study offers therefore no information on the proportion of all patients with an indication to undergo LAA closure which is suitable for LAA ligation. The events were self-reported by the principal investigator of each institution. Additionally, clinical follow-up commonly included contact with the initial referring physicians, which could have affected the accuracy of long-term outcomes. Two pericardial effusions requiring pericardiocentesis after the procedure were noted as minor complications, which is in line with the Munich consensus document,¹¹ but would rather be major complications in most studies on interventional cardiac procedures. Nevertheless, the nomenclature proposed in the Munich consensus was applied in the study to enable improved comparability of future studies on LAA closure. Transoesophageal echocardiography follow-up data at 1–3 months were available in only 74.6% of the population, which is a major limitation of the study. The small patient number with a limited follow-up does not allow final conclusions about the effectiveness of LAA ligation for stroke prevention.

Conclusion

Initial experience with the LARIAT⁺ device shows that LAA ligation is feasible for LAA exclusion. Prospective studies with larger patient cohorts and longer follow-up periods are warranted to evaluate the safety and effectiveness of prevention of thromboembolism.

Funding

Part of analysed data delivered from Prof. Bartus was supported by the National Science Center (2014/13/D/NZ5/01351 and 2015/17/B/NZ5/00125).

Conflict of interest: T.F. received travel grants from SentreHeart. K.B. is a consultant for SentreHeart. R.J.L. is a consultant to and has equity in SentreHeart. R.T. received travel grants from SentreHeart. The other authors report no relevant conflicts of interest.

Author notes

Roland Richard Tilz and Thomas Fink contributed equally to the study.