Page 38 - Journal of Structural Heart Disease Volume 4, Issue 2

59
Meeting Abstracts
LATE PERFORATION OF THE LAA 4 MONTHS AFTER OCCLUDER IMPLANTATION. REASON FOR OR CAUSED BY A RESUSCITATION?
Karsten Schenke1, Stephan Geidel2, Joachim Schofer3, Christian Keller4, Groenefeld Gerian5
1 Ak Barmbek; Cardiology; Internal Medicine, 2 Ak St. Georg; Cardiac Surgery ; Department, 3 Medical Care Centre; Prof. Mathey, Prof. Schofer Gmbh; Hamburg University Cardiovascular Center, 4 Ak Altona; Internal Medicine; Cardiology, 5 Ak Barmbek; Internal Medicine; Cardiology
History and Physical: A 75-years old male with atrial brillation, a CHADS VASc Score of 4 and a HAS BLED Score of 4 was referred for an interventional closure of the left atrial appendage (LAA). The inter- vention was successful with an SJM® Amulet 18 mm device (panel 1). Nearly 4 months later the patient expressed acute dyspnoea and some minutes later had to be resuscitated. Pulseless electrical activity was documented by the rescue team. With ongoing thoracic compressions and application of catecholamines ROSC occurred after 10 minutes and the patient could be transported to a hospital. Ultrasound revealed a pericardial tamponade and a drainage tube was installed. Over the next hour two liters of blood were drained and conventional methods of stabilizing the coagulation did not stop the bleeding. In preparation of cardiac surgery a coronary angiogram was performed (panel 2). It revealed active leakage of contrast agent in the proximal circum ex artery and the patient was transferred to the cardiac surgery department. Intra-operatively a perforation of the tissue at the basis of the LAA close to the left main coronary artery was discovered. The occluder was excised through the opened left, and the LAA was closed by endocardial sutures. After that two sutures were used to seal another lesion at the LAA basis which stopped the bleeding.
Discussion: In review of the clinical information it remains unclear whether a primary perforation of the LAA led to the cardiac arrest. Since there were more than one laceration of the LAA it seems also possible that the thoracic compressions led to the perforation and consecutive pericardial tamponade. In the literature there are several reported cases of an early perforation of the pulmonary artery by an LAA occlusion device leading to cardiac tamponade. To our knowl- edge this is the rst reported case of a potential late lazeration of the circum ex artery by an LAA occluder.
Imaging: Panel 1: TOE of the implantation, Panel 2: Coronary angiogram.
Indication for Intervention: Pericardial tamponade due to a perfo- ration of the LAA by the occluder with suspected lazeration of the circum ex artery.
Intervention: Pericardiocentesis and introduction of a pig tail drain- age catheter. Surgical removal of the occluder and closure of the LAA basis by suture.
Learning Points of the Procedure: Check early for cardiac tamponade after LAA occlusion even when the procedure has been performed more than 3 months earlier, i.e. after an unexplained resuscitation.
3D MULTIMODAL IMAGE FUSION FRAMEWORK FOR LAA CLOSURE PROCEDURE
Giorgia Vivoli1, Emanuele Gasparotti1, Marco Rezzaghi2, Massimiliano Mariani2, Vincenzo Positano1, Luigi Landini3, Simona Celi1, Sergio Berti2
1 Biocardiolab; Bioengineering Unit; Fondazione Toscana "G. Monasterio", 2 Department of Interventional Cardiology; Fondazione Toscana "G. Monasterio"; Ospedale del Cuore, 3 Department of Information Engineering, University of Pisa; Bioengineering Unit; Fondazione Toscana "G. Monasterio"
Background: The left atrial appendage (LAA) is the main site respon- sible of thrombus formation in patients with non-valvular atrial brillation (NVAF). Oral anticoagulant therapy is the gold standard treatment to prevent cardioembolic events. In patients with NVAF and contraindications to the anticoagulation therapy, percutane- ous LAA closure is a strategy to reduce the cardioembolic risk. This procedure is particularly challenging because of LAA anatomical complexity.
Objective: The aim of this study was trifold: to develop a multimod- al-image-fusion technique in order to support the procedure’s plan- ning and execution; to assess the feasibility of LAA anatomical and functional characterization by a multimodal approach based on LAA 3D surface models and to develop an automatic tool able to quantify morphological parameters.
Methods: Three di erent imaging modalities were involved in this study: pre-operative ECG-gated cardiac CT, pre- and intra-operative 3D echocardiography (transesophageal echocardiography TEE or intracardiac echocardiography ICE) and intra-operative angiography (XA). LAA 3D models were generated using the volumetric CT and US acquisitions; for both CT and US datasets, multiple models were created, one for each of the ten phases in which the cardiac cycle was divided. A custom software was developed to register these models with US and angiographic images, through an algorithm based on the use of RAO/LAO and CRA/CAU C-arm rotation angles and on the identi cation of anatomical reperi. This algorithm was applied for all cardiac phases, allowing a dynamic spatial fusion. LAA anatomical and functional parameters was measured directly from 3D US and CT models: LAA volume (V-LAA) and ostium area (AO-LAA) were calcu- lated for each cardiac phase and LAA ejection fraction (LAA-EF) in a full cardiac cycle was measured for the LAA motility evaluation. The 3D models-based-method for LAA parameters extraction was vali- dated on 10 CT datasets from patients scheduled for percutaneous LAA closure procedure, availing the Simpson’s method at present the gold standard technique. The agreement between methods was assessed by paired t-test.
Results: This study has demonstrated the feasibility to fuse 3D LAA US and CT models with XA and 3D US images, without the use of dedicated commercial platform. The measurements extracted with our 3D-models-based-method has presented a high cor- relation with respect to the gold-standard technique (LA dia- stolic phase: V-LAA=10.61±5.17 cm3 vs 10.55±4.88 cm3, p=0.8319; AO-LAA=4.80±1.66 cm2 vs 4.77±1.53 cm2, p=0.5045; LA systolic phase: V-LAA= 7.51 ± 4.85 cm3 vs 7.46 ± 4.71 cm3 , p=0.7693; AO-LAA= 3.78 ± 1.72 cm2 vs 3.79 ± 1.69 cm2, p=0.8227; LAA-EF=33±16% vs 33±16%, p=0.9993).
Hijazi, Z
2017 LAA CSI Focus Abstracts