Concomitant Mechanical Aspiration and Appendage Closure for Recalcitrant Left Atrial Appendage Thrombi

Key Points

Question

How should persistent left atrial appendage (LAA) thrombi be managed when high-dose anticoagulation is ineffective or not advisable due to an increased risk of bleeding?

Findings

In this case series of 9 patients with persistent LAA thrombus and either conditions precluding anticoagulation intensification or thromboembolism with therapeutic anticoagulation, patients underwent temporary cerebral protection device placement when anatomically feasible, percutaneous mechanical vacuum thrombectomy using a 20F aspiration device with a 15-mm funneled ostium and mechanical LAA closure. In all patients, the procedure was successful without complications.

Meaning

These results suggest that temporary cerebral protection, percutaneous mechanical vacuum thrombectomy, and mechanical LAA closure is a promising treatment strategy for patients with persistent LAA thrombus despite optimal oral anticoagulation.

This case series examines a management strategy for resistant left atrial appendage (LAA) thrombus that uses percutaneous mechanical thrombus aspiration with cerebral protection, followed by LAA closure.

Abstract

Importance

It is well appreciated that a small yet high-risk subset of patients with atrial fibrillation (AF) develops persistent left atrial appendage (LAA) thrombus despite optimal oral anticoagulation (OAC). In patients with either a heightened risk of bleeding precluding enhanced doses of OAC to dissolve the thrombus, or thromboembolism despite optimal OAC, there are limited nonsurgical options.

Objectives

To introduce a novel management strategy for resistant LAA thrombus: percutaneous mechanical thrombus aspiration with cerebral protection, followed by LAA closure (LAAC).

Design, Setting, and Participants

This descriptive, prospective case series of 9 consecutive patients with persistent LAA thrombus was conducted between August 2023 and July 2024 at Mount Sinai Hospital. After placing a cerebral protection device (CPD) when anatomically feasible, balloon atrial septostomy was performed as needed to enhance transeptal access. A 20F mechanical aspiration device with a 15-mm funneled ostium was advanced to the LAA ostium, and manual vacuum aspiration of thrombus was performed. After ultrasonic confirmation of thrombectomy, an LAAC device was implanted.

Main Outcomes and Measures

Patient demographic, clinical, and intervention data were collected.

Results

Nine patients (mean [SD] age, 74.7 [13.6] years; median [IQR] CHA₂DS₂-VASc [congestive heart failure, hypertension, age 75 years or older, diabetes, stroke, vascular disease, age 65-74 years, and female sex] score, 5 [2-8] points) presented with LAA thrombus in the context of contraindications to OAC. CPD was placed in 8 patients; one had prohibitive aortic arch anatomy. Given the patient’s high thrombotic burden and limited alternatives, the decision was to proceed without CPD, based on a risk-benefit assessment. Balloon atrial septostomy was performed in 8 patients. In 8 patients, complete thrombus removal was achieved, followed by successful LAAC device placement. One patient had a small residual distal thrombus, presumably organized, which was strongly adherent to the LAA myocardium; this was compressed against the wall with the LAAC device. One patient had a small thrombus fragment recovered from the CPD postprocedure. During both the index procedure and during follow-up (range, 4-15 months), there were no complications.

Conclusions and Relevance

This cases series found that temporary cerebral protection, percutaneous mechanical vacuum thrombectomy, and mechanical LAA closure was feasible in patients with persistent LAA thrombus despite optimal OAC. Larger studies will be needed to fully assess the efficacy and safety of this approach.

Introduction

In patients with atrial fibrillation (AF) receiving oral anticoagulation (OAC), there is a high-risk subgroup that nonetheless develops left atrial appendage (LAA) thrombus. A meta-analysis of more than 14 000 patients undergoing transesophageal echocardiography (TEE) found a 2.7% incidence of LAA thrombus, with similar rates for vitamin K antagonists (VKA) (2.8%) or non-VKA OAC (NOAC) (3.1%).¹ Indeed, this resistance to OAC may contribute to the higher occurrence of subclinical cerebral emboli (leading to cognitive decline) at double the rate of manifest stroke in AF.²

The standard treatment for recalcitrant LAA thrombi is intensifying the antithrombotic regimen. Patients may switch to different NOACs or to VKA, and some additional antiplatelet agents. Intravenous anticoagulants may also accelerate thrombus resolution, especially when LAA closure (LAAC) is planned.

A challenging scenario arises when thrombus persists, systemic embolization occurs, or a heightened bleeding risk precludes OAC intensification. In these instances, LAAC alone may be an option—provided the procedure does not disrupt and embolize the thrombus. Indeed, for distally located thrombi, the LAAC device can be placed without advancing the delivery catheter or device deeper into the LAA.^3,4 However, this is not viable for large, proximally located thrombi. Surgical thrombectomy is another option, but is invasive and carries high risks, particularly in older patients with comorbidities. This case series presents a novel strategy: percutaneous vacuum aspiration thrombectomy with distal carotid artery protection, followed by mechanical LAAC (Figure 1).

Figure 1. Schematic Illustration of the Workflow for Mechanical Thrombectomy.

The diagram outlines the sequence of steps, including imaging to assess the size and location of the thrombus, followed by percutaneous mechanical thrombectomy with cerebral protection, and subsequent appendage closure. Arrows indicate the flow of the process from one step to the next. LAA indicates left atrial appendage.

Methods

The Icahn School of Medicine at Mount Sinai institutional review board approved this study. After written informed consent, under general anesthesia, the procedures were guided by fluoroscopy and intracardiac echocardiography (ICE; AcuNav [Biosense-Webster Inc] or ViewFlex Xtra [St Jude Medical Inc]), and also TEE in select patients. ICE imaging was from the right atrium (RA), pulmonary artery, or coronary sinus.

Thrombectomy Procedure

Right radial or ulnar artery access was obtained for deployment of a cerebral protection device (CPD; Sentinel [Boston Scientific Inc]). Aortography assessed aortic arch anatomy for CPD suitability. Coronary angiography was selectively performed. Activated clotting time of more than 300 seconds was maintained with heparin.

After ultrasound-guided femoral vein access, a 26F inflatable-seal sheath (Dryseal [Gore and Associates]) was advanced over a wire into the inferior vena cava. ICE-guided transeptal puncture was performed at the mid-anterior-inferior septum using an RF-activated wire. With the guidewire in a pulmonary vein, the thrombectomy device was advanced into the left atrium (LA), aided by septostomy predilatation using a peripheral angioplasty balloon (8- to 10-mm diameter) for facile advancement of the system.

The thrombectomy sheath and device were cautiously advanced to the LAA ostium without disrupting the thrombus. The funnel tip was positioned adjacent to the thrombus for mechanical suction, typically set to 30 mL per attempt. After thrombus evacuation, the sheath was withdrawn into the RA.

The thrombectomy sheath was removed, and a 12F LAAC delivery sheath was used to place the LAAC device (Watchman-FLX or Watchman-FLX Pro [Boston Scientific Inc]) in standard fashion. A figure-of-8 stitch was used for temporary hemostasis.

Procedure Conclusion

Pericardial effusion was excluded, and hemostasis was achieved using sutures, or closure devices for smaller sheaths. The CPD was removed and inspected for any captured material. Total blood loss was documented.

Mechanical Thrombectomy Device

The percutaneous vacuum aspiration device (AlphaVac [Angiodynamics]), developed for acute pulmonary embolectomy,⁵ enables targeted aspiration of thrombi without extracorporeal circulation, potentially offering a controlled method for thrombus removal, minimizing embolic risk. A single-hand operation handle induces a vacuum for manual aspiration through a torqueable funneled-tip 22F inner cannula (180° angle) inside a 26F outer cannula. The funnel tip entraps the clot before manual aspiration (10 or 30 mL per suction attempt) into a 250-mL waste bag, which can then be filtered and inspected for thrombi.

Data Analysis

Descriptive statistics were used to summarize the data. Categorical data are presented as counts. Continuous variables were expressed as mean for normally distributed data and as median for non-normally distributed data. Data were analyzed using Microsoft Excel 2016.

Results

Patient Characteristics

This report includes 9 consecutive patients with persistent LAA thrombus who underwent LAA thrombectomy and LAAC. The mean (SD) age was 74.7 (13.6) years (eTable 1 in Supplement 1) with either persistent (n = 7) or long-standing persistent (n = 2) AF, and the median (IQR) CHA₂DS₂-VASc score was 5 (2-8) points (including 2 patients with score of 2 but with hypertrophic cardiomyopathy and history of rheumatic heart disease). All patients had received an OAC, and 4 received warfarin immediately preprocedure.

The thrombectomy indication was substantial bleeding (n = 4; hematochezia or intracranial hematoma) or cerebral microhemorrhage (n = 2), preventing OAC escalation. One patient experienced a hemoglobin drop to 4.4 g/dL (to convert to g/L, multiply to 10), and another had persistent LAA thrombus despite 3 OAC trials, plus high fall risk. Three patients experienced systemic embolism despite OACs: (1) right coronary artery occlusion leading to a non-ST elevation myocardial infarction and acute heart failure, with intravenous anticoagulation achieving recanalization, confirming thromboembolism, (2) one patient had multiorgan embolization with stroke, kidney injury, and mesenteric ischemia requiring small and large intestine resection, and (3) thromboembolic vertebrobasilar artery stroke.

Outcomes

The procedural details and outcomes are in eTable 2 in Supplement 1. TEE or computed tomography revealed persistent LAA thrombus in all patients (Figures 1 and 2). Intraprocedural imaging in 1 patient revealed only dense “smoke” but without dissipation with catecholamines,⁶ prompting thrombectomy.

Figure 2. Left Atrial Appendage Thrombus: In Vivo Multimodality Imaging and Post-Evacuation.

Intracardiac echocardiography revealed LAA thrombi in different patients (arrowhead and asterisks, A-D). Computed tomography (CT) images of the LAA thrombus (E-G): the sagittal image showing a large thrombus prior to augmented anticoagulation (arrowhead, E), and transverse (arrowhead, F) and coronal (arrowhead, G) images of the LAA clot after intravenous anticoagulation. The aspirated thrombus fragments are shown next to measurement scales (H, I).

Coronary angiography (n = 2) revealed nonobstructive coronary artery disease. A CPD was successfully placed in 8 of 9 patients; complex aortic branch anatomy precluded placement in 1 patient (Figure 1). Dilatation of the atrial septostomy was required in 8 of 9 patients for facile passage of the thrombectomy cannula (Figure 3).

Figure 3. Ancillary Equipment Used During Mechanical Thrombectomy Cases.

Fluoroscopic image of balloon dilatation during atrial septostomy is shown (arrowhead, A). Intracardiac echocardiographic (ICE) image of the LAA reveals the thrombus (asterisk, B), the ICE and fluoroscopic image of mechanical aspiration funnel in the LAA (arrowhead, B and C), and a postaspiration image shows complete evacuation of the thrombus (D). Echocardiography and fluoroscopy reveal an LAAC device implanted after thrombectomy (yellow arrowheads E and F; and black arrowhead, G). Successful deployment of a septal occluder device is shown by fluoroscopy (blue arrowhead at septal occluder, G) and ICE (arrowhead at septal occluder, H).

Mechanical thrombectomy was successful in all 9 patients (Figures 1, 2, and 3), with full removal in 8 patients and leaving a small residual distal thrombus in 1 patient despite multiple attempts. Thrombus was identified in 6 of 9 waste bags, with the rest having potential disruption or dissolution of clot, or inadequate expression from the tubing apparatus. No thrombus was recovered from 7 of the 8 CPDs, but a small amount of pale material was recovered from patient 6, which was tissue-free upon pathological analysis.

Following thrombectomy, LAAC device implantation was successful in all patients, including the patient with residual thrombus, in whom the device compressed the thrombus against the wall (Figures 1 and 3). A septal occluder was placed in 1 patient due to a substantial atrial shunt (Figure 3).

There were no complications during the procedures or on follow-up (4 to 15 months). Follow-up imaging in 8 of 9 patients revealed well-healed devices, with further imaging pending for 1 patient. A peri-device leak was observed in 1 patient, and a residual small ASD was found in 4 patients.

Discussion

In this case series of patients with persistent LAA thrombi: (1) mechanical thrombectomy successfully eliminated most thrombi with immediate LAA closure feasible, (2) the procedure was safe, and (3) it was feasible without the need for an extracorporeal circuit.

In a retrospective study of 7759 patients with symptomatic AF, TEE identified LAA thrombi in 5.8%,⁷ with TEE every 3 to 4 months (approximately 4 times) revealing thrombus resolution in only 56.3%. Interestingly, 10-year survival was higher in both (1) patients without thrombi (69%) vs with thrombi (31%) and (2) in patients with resolved thrombi (41%) vs persistent thrombi (17%). Albeit observational in nature, these findings suggest that managing persistent LAA thrombi acutely and definitively may potentially improve outcomes.

Patients with OAC-resistant LAA thrombi face limited therapeutic options, contributing to patient anxiety and requiring frequent monitoring. While LAAC and surgical thrombectomy are options in select patients, percutaneous mechanical thrombus aspiration emerges as a potentially optimal solution. Percutaneous aspiration of right-sided heart masses has been reported,⁵ but substantial technical challenges were anticipated for left-sided heart aspiration.⁸ A different vacuum aspiration system requiring extracorporeal bypass was used in case reports.^9,10,11,12 Our series demonstrated successful thrombus removal using a mechanical aspiration device, even with fibrous, organized thrombi resistant to OAC.

Cerebral protection was successfully implemented in 8 of 9 patients. In 7 patients, no debris was recovered, but a small amount of filamentous material was recovered in one CPD. This suggests that using the funneled thrombectomy device is potentially effective in preventing thromboembolism. Given the small LAA volume (approximately 8 mL),¹³ the large volume of blood aspirated (30 mL) with each suction may explain the effectiveness of the thrombectomy device (Video). In our experience with other procedures (eg, removal of LAAC device-related thrombi) using this aspiration device, we have extracted debris from the CPD and the left radial artery. Indeed, the CPD device does not provide complete embolic protection against the posterior brain circulation or the systemic circulation. Thus, it is crucial to document pulses in all extremities pre- and postthrombectomy. Future larger studies may consider both pre- and postprocedure brain MRI, and transient retrograde placement of a Watchman device in the ascending aorta for systemic protection.¹⁴

Video. Single-Handed Manual Aspiration of Thrombi With a Torqueable Funnel-Tip Cannula.

Download video file ^{(33.2MB, mp4)}

This video demonstrates the technique of single-handed manual aspiration of thrombi using a torqueable funnel-tip cannula. The video highlights how the suction cannula effectively aspirates both large chunks and smaller fragments of thrombi, showcasing the efficiency and control afforded by the design of the thrombectomy device.

While this technique shows promise, careful attention must be paid to catheter manipulation to avoid thrombus dislodgement, especially in proximal LAA locations.^3,4 Despite these complexities, our initial experience offers a potential solution to a challenging clinical scenario.

Limitations

This study has limitations. This case series was small, but it serves as a proof of concept for the feasibility of this strategy. Certainly, larger studies including multiple centers and operators, with control groups, are needed to confirm its safety and effectiveness. The follow-up period was limited, but late complications are unexpected. Additionally, the inherent risk of device-related thrombus and peridevice leak with LAAC mandates at least short-term antithrombotic therapy and potential reintervention.

Conclusion

This study found that temporary cerebral protection, percutaneous mechanical vacuum thrombectomy, and mechanical LAA closure was feasible in patients with persistent LAA thrombus despite optimal OAC. Larger studies are required to further assess the procedure’s safety and efficacy.

Supplement 1.

eTable 1. Patient Details

eTable 2. Procedure Details and Outcomes

Supplement 2.

Data Sharing Statement