IJCS | Volume 32, Nº5, September/October 2019

461 Linhares et al. Transesophageal echocardiography and CHA2DS2-Vasc score Int J Cardiovasc Sci. 2019;32(5):460-470 Original Article systemic thromboembolism occurs in patients whose score is ≥ 2 for males or ≥ 3 for females. 2 Left atrial appendage (LAA) represents one of the main sources of cardiac thrombi responsible for stroke in patients with AF 3-6 and this is probably due to the anatomical characteristics of this structure, which facilitate slower blood flow inside it. An interesting finding is that LAA thrombosis can occur even in patients with a lowered CHA 2 DS 2 VASc score (< 2) and this may be related to its morphology. Di Biase et al., 3 through magnetic resonance imaging and computed tomography, classified LAA into 4 different morphologies: chicken wing, windsock, cactus and cauliflower, according to their appearance in the imaging scans. With this characterization, these authors were able to correlate each morphological type with the risk of stroke, being greater in those whose morphology was different from the chicken wing morphology. 3 The more complex morphology of these non-chicken wing structures may explain the increased risk of atrial thrombosis. Two-dimensional transesophageal echocardiography (2D TEE) has been widely used to assess the left atrium and to characterize the structure and function of LAA. It is the method of choice to evaluate the presence of thrombus, as it provides a better imaging of these structures compared to transthoracic echocardiography, has greater accessibility and lower cost than magnetic resonance imaging and computed tomography, and can thus help identify risk factors for stroke. Agreement between the TEE findings and the CHA 2 DS 2 -VASc score is still poorly explored in our country. The objectives of this study were to evaluate, using TEE, the echocardiographic data with the CHA 2 DS 2 VASc score. Besides, to establish its usefulness in the characterization of the LAA morphology for thromboembolic risk stratification, based on the studies with magnetic resonance imaging. Material and methods A randomized retrospective study that evaluated data from patients undergoing outpatient transesophageal echocardiography between August 2012 and June 2015 in a specialized laboratory. All patients signed an informed consent form. The echocardiographic studywas performed by 2 experienced and qualified echocardiographers in state-of-the-art equipment with second harmonics (GE- Vingmed Ultrasound, Vivid 7 and S6, Horten, Norway and ESAOTE My Lab 70, Florence, Italy). The clinical indications of patients referred for the scan were: of the 207 patients with no history of atrial fibrillation, 50.2% interatrial septum evaluation, 25.2% thrombus investigation, 8.7% non-specific cardiac evaluation, 4, 4% thoracic aorta evaluation, 3.8% suspected endocarditis, 3.8% interventricular septum evaluation, 2% valvular heart disease evaluation, 1% cardiac tumor investigation, 0.5% syncope, 0.5% myocardiopathy. In patients with a history of atrial fibrillation, the indications were: 82.5% thrombus investigation, 10% interatrial septumevaluation and 7.5% aortic evaluation. Of all patients, 18.6% had a history of stroke/TIA and 5.7% had coronary artery disease. Using a questionnaire, considering the clinical data, the CHA 2 DS 2 -VASc score of each patient was quantified according to the European guidelines. 2 Two points were assigned to the risks considered larger represented by previous history of stroke, TIA or systemic embolism and age greater than 75 years, and one point for other risk factors. Patients with a history of stroke or transient ischemic attack, and individuals with sinus rhythm and history of atrial fibrillation were also identified. Two-dimensional transthoracic echocardiography with previous color Doppler was performed to determine ejection fraction and the degree of left atrial dilatation. Then, TEE was employed to look for thrombus and spontaneous contrast, and to determine blood flow velocity in LAA and its morphology. At the TEE scan, local anesthetic was initially administered via the oropharyngeal aspiration route with lidocaine spray and then light sedation with midazolam, dolanthine and propofol as required. Patientsweremonitored throughout the procedure with evaluation of vital signs, heart rate and pulse oximetry saturation. The room was equipped with oxygen therapy support, cardio-defibrillator, and equipment andmedications needed for any emergencies. After transesophageal catheterization, echocardiographic images were taken at the mid-esophagus level of the whole heart, mainly at the left atrial appendage and at the proximal sections at 0°, 60°, 90° and 120° and later at the transgastric section at 0° and 90°. We excluded patients with contraindication for outpatient ECOTE, patients undergoing left atrial appendage occlusion, patients with primary valvopathy with hemodynamic repercussion or undergoing valve surgery. The patients were separated into 3 different groups according to the CHA 2 DS 2 -VASc score: group 1: 0 and 1 point, group 2: 2 and 3 points and group 3: ≥ 4 points. The variables evaluatedwere: ejection fraction (Teichholz