Computational Simulation to Plan for Percutaneous Left Atrial Appendage Closure

Atrial Fibrillation, Stroke Prevention

Atrial fibrillation, Left atrial appendage, Transcatheter closure, Stroke prevention, Computed tomography, Computational simulation, Preoperative planning

To ensure a successful percutaneous left atrial appendage (LAA) closure, it is important to select the correct size of the closure device used for each patient. To define and measure the size of the LAA, 2D transesophageal echocardiography (TEE) has been typically used. An increasing number of hospitals recently switched to measure the size of the LAA using a cardiac computed tomography (CT) scan. Although this CT scan helps to better define and measure the LAA, it is still difficult to determine the exact anticipated 'landing zone' or 'position' of the closure device. A novel strategy of preprocedural planning includes the use of preoperative computer simulations based on CT imaging (Feops HEARTguideTM), where the device is deployed with a computer simulation into the patient-specific LAA anatomy to provide the operator both optimal and suboptimal scenario's showing different sizes and positions of the closure device. The aim of this study is to assess whether use of FEops HEARTguide computer simulations based on cardiac CT-imaging can contribute to a better preprocedural planning and improved procedural outcomes of percutaneous LAA closure procedures with an Amplatzer Amulet device. The PREDICT-LAA trial will investigate the possible positive effect of a computer simulation using a CT scan of the LAA performed prior to the procedure. The hypothesis is that by using this new computer simulation, better planning of the intervention can be obtained.

To prevent stroke in patients with non-valvular atrial fibrillation (NVAF) and contraindication(s) to oral anticoagulant therapy, percutaneous left atrial appendage (LAA) closure is increasily being chosen as an alternative therapeutic option. To obtain a successful percutaneous LAA closure, an accurate preoperative planning is required to understand the LAA morphology and assess the correct size and optimal position of the device. The instructions for use (IFUs) of the medical device providers are still based on two-dimensional transesophageal echocardiography, whereas an increasing number of centers have shifted towards LAA sizing and planning based on cardiac computed tomography (CT) images. Other options are also available for preoperative planning, among others the use of 3D printed LAA models. However, this has a number of disadvantages, among others complex logistics and the possible inaccuracy in terms of mechanical response of the printing material used for the 3D printed model. Based on the FEops HEARTguideTM platform, a computational model for percutaneous LAA closure was developed and validated. The computer simulations are based on CT images, and can predict different possible LAA closure options for the patient, including optimal and suboptimal sizing and positioning of the LAA closure device. This can be achieved by modelling the mechanical interaction between the device and the 3D patient-specific LAA anatomy reconstructed from the CT images. The results are available as a web-based 3D-viewer and allow the physician to analyse the different device size and implant position options prior to the procedure. The PREDICT-LAA trial aims to study the possible added value of FEops HEARTguideTM patient-specific computational simulation in the planning of percutaneous LAA closure with the AmplatzerTM AmuletTM device, with special focus on procedural safety and efficiency as well as on clinical outcomes. The PREDICT-LAA study is a prospective, multicenter, randomized controlled trial. In total, 200 patients eligible for percutaneous LAA closure with an AmplatzerTM AmuletTM device (Abbott, USA) will be enrolled - 100 patients will be allocated to the computational simulation treatment arm and 100 patients to the standard treatment arm. All participants will be treated according to the standard of care of the center - additionally, the preoperative planning for patients allocated to the computational simulation treatment arm will include a careful analysis of the computational simulation results provided by the FEops HEARTguideTM platform. Patients with non-valvular atrial fibrillation who are referred to and deemed eligible for percutaneous LAA closure with an AmplatzerTM AmuletTM closure device can be included. Patients with reduced renal function, known contrast agent allergy, and/or suboptimal cardiac CT-image quality are excluded from this trial. The participant shall be informed about the aim and procedure of the PREDICT-LAA study and written informed consent shall be obtained in order to include the participant. All patients enrolled in the PREDICT-LAA study should have a post-procedural cardiac CT scan at three months after the LAA closure procedure to check for complete LAA closure, device-related thrombosis (DRT) and device position. The CT CoreLab for assessment of this post-procedural CT-scans will be performed at Rigshospitalet, Copenhagen, Denmark. The readers of the CT-scan will be blinded from the baseline data, randomisation arm and procedural data. The risk related to this study and, in particular, the CT-scan is low - especially as patients with renal insufficiency, iodine contrast allergy and/or contraindications for CT-scan are excluded. Concerning the radiation dose of maximally two CT-scans, we can report that patients will receive a radiation dose of 6 to 15 mSievert, depending on weight and heart frequency/rhythym. It can be calculated that lifetime risk to die from cancer, hereby, theoretically increases with maximally 0,06%. Thus, the patients' lifetime risk to die from cancer increases from 25.0% to 25.06%. The patients are informed about the aim and procedure of the study and are only included in case of written consent. Unknown side effects or risks associated with the study cannot be ruled-out. The protocol of the trial has been approved by local regional Ethics Committees in each participating country, the collection of data complies with the regulatory rules of the Danish Data Protection Agency, and the study is being conducted in compliance with good clinical practice and with the Helsinki II Declaration as adopted by the 18th World Medical Assembly in Helsinki, Finland, in 1964 and subsequent versions.

Belgium, Canada, Denmark, France, Italy, Spain, Sweden