All enrolled subjects underwent TTE and TEE after admission routinely. All parameters were determined by 2 experienced physicians. Any difference was resolved by a third independent observer. TTE was performed with a cardiovascular ultrasound system (IE 33 Elite, Royal Philips, Netherlands or Vivid E9, General Electric Company, US) to measure LA diameter (LAD), left ventricular ejection fraction (LVEF) and other parameters determined from the parasternal M-mode or 2D images. Left atrial volume index (LAVI) was defined as left atrial volume (LVA) standardized for body surface area (BSA). In detail, three LADs were measured from images optimized for the LA at end systole. D1 is the anterior-posterior (A-P) dimension measured perpendicular to the aortic root long axis in the parasternal long axis view, D2 is the superior inferior (S-I) dimension measured between the mitral annulus and the back wall of the left atrium, and D3 is the medial lateral (M-L) dimension orthogonal to D2 measured from the apical 4-chamber. LAV was calculated using the formula: (D1(A-P)×D2(S-I)×D3(M-L))×(0.523).16 (link) The BSA was calculated by the Mosteller formula.17 (link)
After informed consent, TEE was performed in all patients using an ultrasound system equipped with a TEE transducer (X7-2t, Philips Healthcare) under local anesthesia by lidocaine hydrochloride spray. Left atrial cavity and LAA were carefully evaluated in multiple planes. LAT was defined as a circumscribed intracavitary echo-dense mass showing acoustic characteristics distinct from the surrounding endocardium and pectinate muscles observed in multiple planes.18 (link) SEC was defined as dynamic “smoke-like” echoes with characteristic chaotic swirling during the cardiac cycle, with the severity of SEC 3+ or 4+ as reported previously.19 (link)